[Bacula-users] Re-initialized database still shows Terminated Jobs

[Ralf Gross]

Hi,

after some time of testing I want now to go live with our new bacula
server. Before I do taht I want to clean to database.


I followed this chapter:
Re-initializing the Catalog Database
http://www.bacula.org/dev-manual/Installi_Configur_PostgreS.html#SECTION00523

cd 
./drop_bacula_tables
./make_bacula_tables
./grant_bacula_privileges

The information about volumes and jobs was cleaned, but I still see information
about old jobs when doing a 'status dir'.

*status dir
VU0EM005 Version: 2.0.2 (28 January 2007) x86_64-unknown-linux-gnu debian 4.0
Daemon started 16-Feb-07 17:31, 0 Jobs run since started.
 Heap: bytes=53,896 max_bytes=55,063 bufs=288 max_bufs=302
No Scheduled Jobs.


Running Jobs:
Console connected at 16-Feb-07 17:52
No Jobs running.


Terminated Jobs:
 JobId  LevelFiles  Bytes   Status   FinishedName

 2  Full 17,61887.98 M  Cancel   02-Feb-07 12:25 VU0EM004
 3  Full 58,077588.9 M  Cancel   02-Feb-07 12:36 VU0EM004
 4  Full  0 0   Cancel   02-Feb-07 12:40 VU0EM004
 7  Full 92,058733.3 M  Cancel   02-Feb-07 12:47 VU0EM004
 8  Full  0 0   Cancel   02-Feb-07 12:51 Bang
 9  Full 94,460854.0 M  Cancel   02-Feb-07 12:56 VU0EM004
10  Full 90,107704.3 M  Cancel   02-Feb-07 12:57 VU0EM004
12  Full103,727861.9 M  Cancel   02-Feb-07 13:43 VU0EM004
13  Full  1,670161.7 M  Cancel   02-Feb-07 13:44 Bang
14  Full   1,618,388231.7 G  OK   02-Feb-07 17:57 VU0EM004


Since I also see information about jobs done by a second bacula install with
'status client', I guess this information is stored somewhere else. How can I
remove these messages?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Re-initialized database still shows Terminated Jobs

[Ralf Gross]

Martin Simmons schrieb:
> > Since I also see information about jobs done by a second bacula install with
> > 'status client', I guess this information is stored somewhere else. How can 
> > I
> > remove these messages?
> 
> Yes, if you delete the "state" file named something like bacula-dir.9101.state
> in the Director's working dir.  It is probably best to stop bacula while doing
> this.  You might want to delete the state files from the other daemons as
> well.

Thanks, that did the trick. In fact, the state files were the only
files I didn't remove inside the working directory.


Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Retention Periods for long-term archival

[Ralf Gross]

Hi,

I had a very simple setup for the last 12 months.

* backup everything to tape and keep the last 2 weeks before
  recycling old volumes
* keep the monthly tape(s) for 3 years

I didn't care much about the retention periods of the monthly tapes. I
just set the Volume Retention to 3 years for the 3-year-pool and 16 days for
the regular backup.

In the Job resource I chose a file retention period of 30 days and a
job retention period of 6 months.

File Retention = 30 days
Job Retention = 6 month


With our new LTO3 40x lib we want to keep more backups and put some
tapes with selected data/files in the safe forever (at least until our IT
department can offer us more than just a 60 days period...). 

New setup:

* daily inc. backup (keep 14 days)
* weekly diff. backup (keep 4 weeks)
* monthly fill backup (keep 6 months)
* full backup of certain directories 3x year (keep forever)

Schedule {
  Name = "Regular Backup"
  Run = Level=Full Pool=Full 1st sun at 0:05
  Run = Level=Differential Pool=Differential FullPool=Full 2nd-5th sun at 0:05
  Run = Level=Incremental Pool=Incremental DifferentialPool=Differential 
FullPool=Full mon-fri at 0:05
}

# archive, only 3x year
Schedule {
  Name = "Archiv"
  Run = Level=Full Pool=Archiv 1st sat jan,may,sep at 6:00
}

I created differnt pools for regular backups and long term archival. I think
the retention times are reasonable. We are aware that we can't handle a _real_
long time archival (copy to new media, verify...), but we want to keep the
volumes for 30 years.

Pool {
  Name = Archiv
  Pool Type = Backup
  Recycle = no
  AutoPrune = no
  Volume Retention = 30 years
}

Pool {
  Name = Full
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Recycle Oldest Volume = yes
  Volume Use Duration = 96 hours
  Volume Retention = 180 days
}

Pool {
  Name = Differential
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Recycle Oldest Volume = yes
  Volume Use Duration = 96 hours
  Volume Retention = 31 days
}

Pool {
  Name = Incremental
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Recycle Oldest Volume = yes
  Volume Use Duration = 6 days
  Volume Retention = 14 days
}



So far so good. But I'm bit unsure about the "right" File and Job Retention
periods. I can set these times only in the client resource, but I would like to
differ the period if the job is a regular backup or an long term archival job.
For the regular backup I would choose 6 months for File/Job Retention, because
that's the longest period until a tape will be recycled again (full monthly
tape). For the archival job I would like to keep the file/job information
longer, not 30 years, but maybe 5 years for the file records.

How can I accomplish this? Any thoughts about this setup or configuration?

Thanks, Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Retention Periods for long-term archival

[Ralf Gross]

Kern Sibbald schrieb:
> > I created differnt pools for regular backups and long term archival. I
> > think the retention times are reasonable. We are aware that we can't handle
> > a _real_ long time archival (copy to new media, verify...), but we want to
> > keep the volumes for 30 years.
> 
> Be sure to archive the conf files as well as the full source for Bacula with 
> each major archive that you do.

Yes and keeping the right hardware in case of a restore might be a problem in
the future too.
 
> > So far so good. But I'm bit unsure about the "right" File and Job Retention
> > periods. I can set these times only in the client resource, but I would
> > like to differ the period if the job is a regular backup or an long term
> > archival job. For the regular backup I would choose 6 months for File/Job
> > Retention, because that's the longest period until a tape will be recycled
> > again (full monthly tape). For the archival job I would like to keep the
> > file/job information longer, not 30 years, but maybe 5 years for the file
> > records.
> >
> > How can I accomplish this? Any thoughts about this setup or configuration?
> 
> 1. You might want to use two different databases, one for regular backups and 
> one for archives.  This will clearly separate your archive data from your 
> regular data.

I changed the bacula db init scripts to create a new database bacula_archive
and can connect with a password to it with psql. 

pg_hba.conf

# TYPE  DATABASEUSERCIDR-ADDRESS  METHOD
local   bacula  baculamd5
local   bacula_archive bacula md5


# psql -U bacula bacula_archive
Passwort f?r Benutzer bacula: 
Dies ist psql 8.1.7, das interaktive PostgreSQL-Terminal.
[...]

But now bacula-dir does not start with the second catalog: 

20-Feb 22:07 bacula-dir:  Fatal error: Database not created or server not 
running.
20-Feb 22:07 bacula-dir:  Fatal error: Could not open Catalog "ArchiveCatalog",
database "bacula_archive".
20-Feb 22:07 bacula-dir:  Fatal error: Database not created or server not 
running.
20-Feb 22:07 bacula-dir ERROR TERMINATION
Please correct configuration file: /opt/bacula/etc/bacula-dir.conf

This is from the postgres log:

2007-02-20 22:07:28 CET FEHLER:  keine Berechtigung f?r Relation version
2007-02-20 22:07:28 CET LOG:  unerwartetes EOF auf Client-Verbindung


# Generic catalog service
Catalog {
  Name = MyCatalog
  dbname = bacula; user = bacula; password = ""
}

Catalog {
  Name = ArchiveCatalog
  dbname = bacula_archive
  user = bacula
  password = ""
}

I don't know what is wrong with the additional catalog. 
 
> 2. You might want to define two separate jobs/clients for each job to be 
> archived.  One uses the normal backup database and retention periods. The 
> other uses the archive database and has appropriate archival retention 
> periods -- why not 30 years in the beginning if you don't do archives very 
> often; you can always reduce it later if your archive database grows too 
> large.

If I can solve the above problem - and two bacula db's are are good to handle -
this will be a good solution.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Retention Periods for long-term archival

[Ralf Gross]

Ralf Gross schrieb:
> But now bacula-dir does not start with the second catalog: 
> 
> 20-Feb 22:07 bacula-dir:  Fatal error: Database not created or server not 
> running.
> 20-Feb 22:07 bacula-dir:  Fatal error: Could not open Catalog 
> "ArchiveCatalog",
> database "bacula_archive".
> 20-Feb 22:07 bacula-dir:  Fatal error: Database not created or server not 
> running.
> 20-Feb 22:07 bacula-dir ERROR TERMINATION
> Please correct configuration file: /opt/bacula/etc/bacula-dir.conf
> 
> This is from the postgres log:
> 
> 2007-02-20 22:07:28 CET FEHLER:  keine Berechtigung f?r Relation version
> 2007-02-20 22:07:28 CET LOG:  unerwartetes EOF auf Client-Verbindung

Ok, it's just too late to do these things. I could connect with psql
but had no rights on the tables. Most likely I ran the old privileges
script and not the modified new one...

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Retention Periods for long-term archival

[Ralf Gross]

Kern Sibbald schrieb:
> 1. You might want to use two different databases, one for regular backups and 
> one for archives.  This will clearly separate your archive data from your 
> regular data.

Basically this working now. But I'm a bit confused about the handling
of pools and the assigned volumes with a second catalog.

These are the pools/volumes I created before I added a second catalog.

*list pools
++--+-+-+--+-+
| poolid | name | numvols | maxvols | pooltype | labelformat |
++--+-+-+--+-+
|  2 | CleaningPool |   0 |   0 | Backup   | *   |
|  3 | Archiv   |   0 |   0 | Backup   | *   |
|  4 | Full |   0 |   0 | Backup   | *   |
|  5 | Differential |   0 |   0 | Backup   | *   |
|  6 | Incremental  |   0 |   0 | Backup   | *   |
|  7 | Scratch  |   0 |   0 | Backup   | *   |
++--+-+-+--+-+

Pool: Archiv
+-++---+-+
| mediaid | volumename | volstatus | enabled |
+-++---+-+ ...snip...
|  17 | 06D143L3   | Append|   1 |
+-++---+-+


Now with the second catalog I see the same pools if I switch to the new catalog
ArchiveCatalog in bconsole with 'use'.

*use
The defined Catalog resources are:
 1: MyCatalog
 2: ArchiveCatalog
Select Catalog resource (1-2): 2
Using Catalog "ArchiveCatalog"
Using Catalog name=ArchiveCatalog DB=bacula_archive

*list pools
++--+-+-+--+-+
| poolid | name | numvols | maxvols | pooltype | labelformat |
++--+-+-+--+-+
|  1 | Default  |   0 |   0 | Backup   | *   |
|  2 | CleaningPool |   0 |   0 | Backup   | *   |
|  3 | Archiv   |   0 |   0 | Backup   | *   |
|  4 | Full |   0 |   0 | Backup   | *   |
|  5 | Differential |   0 |   0 | Backup   | *   |
|  6 | Incremental  |   0 |   0 | Backup   | *   |
|  7 | Scratch  |   0 |   0 | Backup   | *   |
++--+-+-+--+-+

That's fine, but the pools are all empty.

*list volumes
Pool: Default
No results to list.
Pool: CleaningPool
No results to list.
Pool: Full
No results to list.
Pool: Differential
No results to list.
Pool: Incremental
No results to list.
Pool: Scratch
No results to list.
Pool: Archiv
No results to list.

Therefor the archive job was complaining that there is no volume in the Archiv 
pool.

21-Feb 09:55 VU0EM005: Job VU0EM003_Archiv.2007-02-21_09.53.40 waiting. Cannot
find any appendable volumes.

Please use the "label"  command to create a new Volume for:
Storage:  "LTO3" (/dev/nst0)
Media type:   LTO3
Pool: Archiv


To solve this, I deleted the Archiv pool in MyCatalog, relabled the tape and
added it to the Archiv pool of the ArchiveCatalog. In addition I deleted the
Full, Default, Differential, Incremental pools in ArchiveCatalog.

It is a bit confusing that the pools exist in both catalogs, but not the
volumes. What about the scratch pool, is there a way to share the volumes
between the catalogs/pools?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Retention Periods for long-term archival

[Ralf Gross]

Kern Sibbald schrieb:
> It is quite possible that Bacula will re-create all the pools the next time 
> the Director is restarted.

Indeed. And the pool id's now are different too. Is there any way that I can
prevent bacula to re-create the deleted pools? I'm sure at some point someone
(maybe me) will add a volume to the MyCatalog Archiv pool instead of the
ArchiveCatalog's pool.

*use
The defined Catalog resources are:
 1: MyCatalog
 2: ArchiveCatalog
Select Catalog resource (1-2): 1
Using Catalog "MyCatalog"
Using Catalog name=MyCatalog DB=bacula
*list pool
++--+-+-+--+-+
| poolid | name | numvols | maxvols | pooltype | labelformat |
++--+-+-+--+-+
|  1 | Default  |   0 |   0 | Backup   | *   |
|  2 | CleaningPool |   0 |   0 | Backup   | *   |
|  4 | Full |   8 |   0 | Backup   | *   |
|  5 | Differential |   4 |   0 | Backup   | *   |
|  6 | Incremental  |   4 |   0 | Backup   | *   |
|  7 | Scratch  |   7 |   0 | Backup   | *   |
|  8 | Archiv   |   0 |   0 | Backup   | *   |
++--+-+-+--+-+
*use
The defined Catalog resources are:
 1: MyCatalog
 2: ArchiveCatalog
Select Catalog resource (1-2): 2
Using Catalog "ArchiveCatalog"
Using Catalog name=ArchiveCatalog DB=bacula_archive
*list pool
++--+-+-+--+-+
| poolid | name | numvols | maxvols | pooltype | labelformat |
++--+-+-+--+-+
|  2 | CleaningPool |   0 |   0 | Backup   | *   |
|  3 | Archiv   |   1 |   0 | Backup   | *   |
|  7 | Scratch  |   0 |   0 | Backup   | *   |
|  8 | Default  |   0 |   0 | Backup   | *   |
|  9 | Full |   0 |   0 | Backup   | *   |
| 10 | Differential |   0 |   0 | Backup   | *   |
| 11 | Incremental  |   0 |   0 | Backup   | *   |
++--+-+-+--+-+


> > It is a bit confusing that the pools exist in both catalogs, but not the
> > volumes. What about the scratch pool, is there a way to share the volumes
> > between the catalogs/pools?
> 
> No.  Maybe it is something to think about for the future though.
> 
> By the way, handling of two catalogs is, for the most part, significantly 
> improved in version 2.0.x compared to 1.38.11.

I'm using 2.0.2.
 
> Pool definitions are currently "global".  At some point in the future, they 
> will need to be attached to a particular catalog ...

That would be great, at the moment it's a bit tricky to use 2 catalogs. 

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] [bacula 2.0.2] ACL's missing after restore

[Ralf Gross]

Hi,

I just tested a restore of files with ACL's. The restored file is missing the
ACL's. The file was restored to the same ACL capable filesystem.

org. file:

# getfacl acltest

# file: acltest
# owner: root
# group: root
user::rw-
user:rg:rwx
group::---
mask::rwx
other::---

restored file:

# getfacl acltest

# file: acltest
# owner: root
# group: root
user::rw-
group::rwx
other::---

This is the File Set with 'aclsupport = yes':

FileSet {
  Name = "VU0EM003 Full"
  Include {
Options {
  signature = MD5
  aclsupport = yes
}
Options {
   RegexDir = ".*000_KEIN_BACKUP.*"
   exclude = yes
}
File = /
File = /server
File = /public
File = /home
  }
  Exclude {
File = /cdrom
File = /media
File = /lost+found
File = /mnt
File = /dev
File = /sys
File = /proc
File = /tmp
File = /.journal
File = /.fsck
  }
}


The client's fd was compiled with acl support:

config.out:

ACL support:yes


There is only one unclear message about acl support in config.log:

configure:32849: checking for acl_get_file
configure:32905: gcc -o conftest -g -O2 -Wall   conftest.c -ldl  >&5
/tmp/ccq41UzW.o: In function `main':
/home/rg/bacula-2.0.2/conftest.c:164: undefined reference to `acl_get_file'
collect2: ld returned 1 exit status
configure:32911: $? = 1
configure: failed program was:
| /* confdefs.h.  */
| #define PACKAGE_NAME ""
[...]

client's bacula-fd:

# ldd /opt/bacula/sbin/bacula-fd
libacl.so.1 => /lib/libacl.so.1 (0x2ad67d53b000)
libz.so.1 => /usr/lib/libz.so.1 (0x2ad67d641000)
libpython2.4.so.1.0 => /usr/lib/libpython2.4.so.1.0 (0x2ad67d758000)
libutil.so.1 => /lib/libutil.so.1 (0x2ad67d996000)
librt.so.1 => /lib/librt.so.1 (0x2ad67da99000)
libpthread.so.0 => /lib/libpthread.so.0 (0x2ad67dba3000)
libdl.so.2 => /lib/libdl.so.2 (0x2ad67dcb8000)
libwrap.so.0 => /lib/libwrap.so.0 (0x2ad67ddbb000)
libssl.so.0.9.8 => /usr/lib/libssl.so.0.9.8 (0x2ad67dec5000)
libcrypto.so.0.9.8 => /usr/lib/libcrypto.so.0.9.8 (0x2ad67e00c000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x2ad67e28)
libm.so.6 => /lib/libm.so.6 (0x2ad67e47f000)
libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x2ad67e601000)
libc.so.6 => /lib/libc.so.6 (0x2ad67e70e000)
libattr.so.1 => /lib/libattr.so.1 (0x2ad67e94b000)
/lib64/ld-linux-x86-64.so.2 (0x2ad67d423000)
libnsl.so.1 => /lib/libnsl.so.1 (0x2ad67ea4f000)

# ls -l /lib/libacl.so*
lrwxrwxrwx 1 root root11 2007-02-09 16:54 /lib/libacl.so -> libacl.so.1
lrwxrwxrwx 1 root root15 2007-01-11 14:05 /lib/libacl.so.1 -> 
libacl.so.1.1.0
-rw-r--r-- 1 root root 26088 2006-07-25 01:38 /lib/libacl.so.1.1.0


I'm not sure when I added 'aclsupport = yes', but I restarted the director and
the client's fd afterwards. The inc. backup from where I restore the file from
was done after that change.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Restarting bacula-dir while a job is in progress?

[Ralf Gross]

Mike Hanby schrieb:
> I'm testing out bacula (a newbie) in an environment with Linux and Windows
> XP 64bit systems.
> 
> I'm running bacula 2.0.2 with -dir, -fd and -sd on a Linux server, and -fd
> on each of the Windows and other Linux systems. The backups are being stored
> onto a 500 GB hard drive.
> 
> I have a full backup running from one of the Windows machines, it's about 15
> GB's into the backup. I noticed that there are several other jobs in the
> queue "waiting execution".
> 
> I changed Maximum Current Jobs from 1 to 2 in bacula-dir.conf.
> 
> In order for this change to take affect, I need to restart bacula-dir.

Try issuing the reload command in bconsole. This way you don't have to
restart the dir.
 
> Do I need to pause the existing job before I restart bacula-dir, I don't see
> a command that will do this.
> 
> If I just restart bacula-dir, will the job resume normal operation once it
> restarts?

No, I'm pretty sure the job will end with an error message.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] [bacula 2.0.2] ACL's missing after restore

[Ralf Gross]

Ralf Gross schrieb:
> 
> I just tested a restore of files with ACL's. The restored file is missing the
> ACL's. The file was restored to the same ACL capable filesystem.
> 
> org. file:
> 
> # getfacl acltest
> 
> # file: acltest
> # owner: root
> # group: root
> user::rw-
> user:rg:rwx
> group::---
> mask::rwx
> other::---
> 
> restored file:
> 
> # getfacl acltest
> 
> # file: acltest
> # owner: root
> # group: root
> user::rw-
> group::rwx
> other::---

Some additional information. The above file is located on a xfs fs.
Today I tried to backup/restore a file from/to an ACL aware ext3 fs,
same result, no ACL's were restored. 

The server is running up to date debian etch with kernel
2.6.18-4-amd64. bacula 2.0.2 was compiled from source with libacl and
aclsupport for the file set is enabled.

Is there a way to check if bacula did back up the ACL's but can't
restore them, or if they weren't backed up at all?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] [bacula 2.0.2] ACL's missing after restore

[Ralf Gross]

Ralf Gross schrieb:
> Ralf Gross schrieb:
> > 
> > I just tested a restore of files with ACL's. The restored file is missing 
> > the
> > ACL's. The file was restored to the same ACL capable filesystem.
> > 
> > org. file:
> > 
> > # getfacl acltest
> > 
> > # file: acltest
> > # owner: root
> > # group: root
> > user::rw-
> > user:rg:rwx
> > group::---
> > mask::rwx
> > other::---
> > 
> > restored file:
> > 
> > # getfacl acltest
> > 
> > # file: acltest
> > # owner: root
> > # group: root
> > user::rw-
> > group::rwx
> > other::---
> 
> Some additional information. The above file is located on a xfs fs.
> Today I tried to backup/restore a file from/to an ACL aware ext3 fs,
> same result, no ACL's were restored. 
> 
> The server is running up to date debian etch with kernel
> 2.6.18-4-amd64. bacula 2.0.2 was compiled from source with libacl and
> aclsupport for the file set is enabled.
> 
> Is there a way to check if bacula did back up the ACL's but can't
> restore them, or if they weren't backed up at all?

Ok, I started the client's fd and all bacula daemons on the backup server with
-d400 and started one more backup.

client's fd:

VU0EM003: find_one.c:210 File : /home/ralfgro/acltest3
VU0EM003: find_one.c:261 Non-directory incremental: /home/ralfgro/acltest3
VU0EM003: backup.c:243 FT_REG saving: /home/ralfgro/acltest3
VU0EM003: backup.c:328 bfiled: sending /home/ralfgro/acltest3 to stored
VU0EM003: backup.c:1004 File /home/ralfgro/acltest3
attribs=P0A K+Gr IG4 B A A A A BAA I BF4X4B BF4X4B BF4YD0 A A C
attribsEx=
VU0EM003: backup.c:1021 >stored: attrhdr 76 1 0
VU0EM003: backup.c:1049 >stored: attr len=86: 76 3 /home/ralfgro/acltest3

on the director:

VU0EM005: catreq.c:417 dirdFrom src/filed/backup.c

 /* Send the buffer to the storage deamon */
   Dmsg2(400, "Backing up ACL type 0x%2x <%s>\n", acltype, jcr->acl_text);

If ACL's were backed up, shouldn't I see this messages somewhere in the debug
output?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] [bacula 2.0.2] ACL's missing after restore

[Ralf Gross]

Ralf Gross schrieb:
> Ok, I started the client's fd and all bacula daemons on the backup server with
> -d400 and started one more backup.
> 
> client's fd:
> 
> VU0EM003: find_one.c:210 File : /home/ralfgro/acltest3
> VU0EM003: find_one.c:261 Non-directory incremental: /home/ralfgro/acltest3
> VU0EM003: backup.c:243 FT_REG saving: /home/ralfgro/acltest3
> VU0EM003: backup.c:328 bfiled: sending /home/ralfgro/acltest3 to stored
> VU0EM003: backup.c:1004 File /home/ralfgro/acltest3
> attribs=P0A K+Gr IG4 B A A A A BAA I BF4X4B BF4X4B BF4YD0 A A C
> attribsEx=
> VU0EM003: backup.c:1021 >stored: attrhdr 76 1 0
> VU0EM003: backup.c:1049 >stored: attr len=86: 76 3 /home/ralfgro/acltest3
> 
> on the director:
> 
> VU0EM005: catreq.c:417 dird VU0EM005: catreq.c:418 dird BF4X4B BF4YD0 A A C
> VU0EM005: catreq.c:433 dird VU0EM005: catreq.c:434 dird BF4X4B BF4YD0 A A C
> VU0EM005: catreq.c:394 UpdCat msg=UpdCat Job=VU0EM003.2007-02-25_13.40.10 
> FileAttributes 
> VU0EM005: catreq.c:396 UpdCat VolSessId=2 VolSessT=1172406686 FI=77 Strm=1 
> data_len=81
> VU0EM005: catreq.c:400 Cached attr. Stream=1 fname=/home/ralfgro/acltest3
> 
> 
> >From src/filed/backup.c
> 
>  /* Send the buffer to the storage deamon */
>Dmsg2(400, "Backing up ACL type 0x%2x <%s>\n", acltype, jcr->acl_text);
> 
> If ACL's were backed up, shouldn't I see this messages somewhere in the debug
> output?

Well, it seems that I'm doing something wrong in the way I define the
FileSet. This is working and backing up all ACL's.

FileSet {
  Name = "VU0EM003 ACL Test"
Include {
  File = /home/ralfgro
Options {
 aclsupport = yes
 signature = MD5
 }
}
}


VU0EM003: backup.c:977 ACL of file: /home/ralfgro/acltest3
successfully backed up!
VU0EM003: backup.c:586 bfiled>stored:header 7217 3 0
VU0EM003: find_one.c:210 File : /home/ralfgro/acltest4
VU0EM003: backup.c:243 FT_REG saving: /home/ralfgro/acltest4
VU0EM003: backup.c:328 bfiled: sending /home/ralfgro/acltest4 to
stored
VU0EM003: backup.c:1004 File /home/ralfgro/acltest4
attribs=P0A K+Gs IG4 B A A A A BAA I BF4X4B BF4X4B BF4Yrq A A C
attribsEx=
VU0EM003: backup.c:1021 >stored: attrhdr 7218 1 0
VU0EM003: backup.c:1049 >stored: attr len=88: 7218 3
/home/ralfgro/acltest4
VU0EM003: backup.c:957 Backing up ACL type 0x 1 http://www.bacula.org/rel-manual/FileSet_Resource.html I couldn't
find the reason wyh the above FileSet should not backup ACL's.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] [bacula 2.0.2] ACL's missing after restore

[Ralf Gross]

Ralf Gross schrieb:
> This is my default FileSet for this client and does not backup ACL's, in
> /server or /home.
> 
> FileSet {
>   Name = "VU0EM003 Full"
>   Include {
> Options {
>   aclsupport = yes
>   signature = MD5
> }
> File = /
> File = /server
> File = /public
> File = /home
> 
> Options {
>RegexDir = ".*000_KEIN_BACKUP.*"
>exclude = yes
> }
>}
[snip]

The two Options{} declarations wasn't working as I expected. This FS seems to
work.

FileSet {
  Name = "VU0EM003 Full"
  Include {
Options {
  aclsupport = yes
  signature = MD5
}
File = /
File = /public
File = /home
  }
  Include {
Options {
  aclsupport = yes
  signature = MD5
  RegexDir = ".*000_KEIN_BACKUP.*"
  exclude = yes
}
File = /server
   }
[snip]

Thanks to Florian Heigl to pointing me in the right direction.

Ralf


-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] [bacula 2.0.2] ACL's missing after restore

[Ralf Gross]

Kern Sibbald schrieb:
> On Sunday 25 February 2007 16:49, Ralf Gross wrote:
> > Ralf Gross schrieb:
> > > This is my default FileSet for this client and does not backup ACL's, in
> > > /server or /home.
> > >
> > > FileSet {
> > >   Name = "VU0EM003 Full"
> > >   Include {
> > > Options {
> > >   aclsupport = yes
> > >   signature = MD5
> > > }
> > > File = /
> > > File = /server
> > > File = /public
> > > File = /home
> > >
> > > Options {
> > >RegexDir = ".*000_KEIN_BACKUP.*"
> > >exclude = yes
> > > }
> > >}
> >
> > [snip]
> >
> 
> The above does not work because the first Options does nothing.  The second 
> options does an exclude, and if nothing is excuded by the second Options, it 
> is included, *but* included with the Options used in the *last* Options to 
> appear within the Include { }.
> 
> The above *probably* would have written (I am unsure what you really wanted) 
> had you written it as:
> 
>  FileSet {
>Name = "VU0EM003 Full"
>Include {
>  File = /
>  File = /server
>  File = /public
>  File = /home
>  Options {
>aclsupport = yes
>signature = MD5
> RegexDir = ".*000_KEIN_BACKUP.*"
> exclude = yes
>   }
> }
> 
>  [snip]

After reading this...

http://www.bacula.org/rel-manual/FileSet_Resource.html#SECTION00159
However, one additional point is that in the case that no match was found,
Bacula will use the options found in the last Options resource. As a
consequence, if you want a particular set of "default" options, you should put
them in an Options resource after any other Options.

...I changed my config to that: 


FileSet {
  Name = "VU0EM003 Full"
  Include {
Options {
  RegexDir = ".*000_KEIN_BACKUP.*"
  exclude = yes
}
Options {
  aclsupport = yes
  signature = MD5
}
File = /
File = /public
File = /home
File = /server
  }
[snip]

This is working. I was a bit unsure if I could put everything in one Options{}
declaration, because the MD5/ACL options should be used for the files that are
not excluded (no suprise...). Therefore putting them to the exclude option
didn't look logical to me.

Background: we have differnt backup levels, the final level is no backup ;)
There is one 000_KEIN_BACKUP (english: no_backup) directory inside every samba
project share. The directory and it's content should be completely excluded
from backup.  People can decide to put there files they don't need to be backed
up or files that have already been a archived.


This is the archive FileSet. It should only backup the files in all
000_INS_ARCHIV (to_archive) directories. The 000_KEIN_BACKUP directories are
explicitly excluded. The archive directories will be backed up 2-3 times a year
and will be removed step by step from the file server.

The exclude/include is working, I didn't try the aclsupport option, but it
should work too.


FileSet 
  Name = "VU0EM003 Archiv"
  Include {
Options {
   RegexDir = ".*000_KEIN_BACKUP.*"
   exclude = yes
}
Options {
   signature = MD5
   aclsupport = yes
   RegexDir = ".*000_INS_ARCHIV.*"
}
File = /server/projekte
  }
[snip]


Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Info on setting up a second catalog?

[Ralf Gross]

Alan Davis schrieb:
> Questions for anyone who's currently using secondary catalogs:
>  
> I may be making this harder than it needs to be, but I need some
> guidance on just how the second catalog gets used.
>  
>  I've succeeded in creating the second db and I've added a second
> Catalog entry in the director config file. I can then execute the
> "use" command to switch catalogs :
>  
>  athos:sysdavis >bconsole
>  Connecting to Director athos.ruckus.com:9101
>  1000 OK: athos-dir Version: 1.39.22 (08 Septermber 2006)
>  Enter a period to cancel a command.
>  *use
>  The defined Catalog resources are:
>   1: MyCatalog
>   2: ArchiveCatalog
>  Select Catalog resource (1-2): 2
>  Using Catalog name=ArchiveCatalog DB=bacula_arc
>  
>  So far, so good. Now I'm stuck because I don't know what operations
>  can/will use the new catalog and what won't.
>  
>  Queries and list commands do use it, but trying to create a Pool in the
>  new catalog, for instance, doesn't work.

I had some problems with a second catalog too. 

http://thread.gmane.org/gmane.comp.sysutils.backup.bacula.general/32045/focus=32063

The pool's that I created before adding a second catalog do exist in
the second catalog too. Even if you delete a pool in your second catalog, it
will be there again after restart! The second catalog's pool is empty, because
the volumes inside the pools are not shared between the two catalogs. Thus
you'll have common pools (pool names), but you have to add volumes to each pool
separately.
  
>  Running a job prompts for the catalog to use, but doesn't use it -
>  all the  data goes into the default MyCatalog. I've tried
>  creating a second client entry that specifies the ArchiveCatalog
>  but that doesn't seem to make a difference either :

This is my config for the second catalog, I created an "alias" for my client.

# Generic catalog service
Catalog {
  Name = MyCatalog
  dbname = bacula; user = bacula; password = "xxx"
}

Catalog {
  Name = ArchiveCatalog
  dbname = bacula_archive
  user = bacula
  password = xxx
}

Client {
  Name = VU0EM003
  Address = 10.60.1.252
  FDPort = 9102
  Catalog = MyCatalog
  Password = ""
  File Retention = 30 days# 30 days
  Job Retention = 6 months# six months
  AutoPrune = yes # Prune expired Jobs/Files
}


Client {
  Name = VU0EM003-Archive
  Address = 10.60.1.252
  FDPort = 9102
  Catalog = ArchiveCatalog
  Password = "xxx"
  File Retention = 30 years   # 30 days
  Job Retention = 30 years# six months
  AutoPrune = no
}


I performed only one backup of the archive client so far. But this worked as
expected.

21-Feb 10:50 VU0EM005: Start Backup JobId 2, 
Job=VU0EM003_Archiv.2007-02-21_10.50.44
21-Feb 10:50 VU0EM005: Spooling data ...
21-Feb 11:56 VU0EM005: Job write elapsed time = 01:05:49, Transfer rate = 54.64 
M bytes/second
21-Feb 11:56 VU0EM005: Committing spooled data to Volume "06D143L3". Despooling 
216,015,495,393 bytes ...
21-Feb 12:47 VU0EM005: Despooling elapsed time = 00:51:08, Transfer rate = 
70.40 M bytes/second
21-Feb 12:53 VU0EM005: Sending spooled attrs to the Director. Despooling 
136,967,826 bytes ...
21-Feb 12:58 VU0EM005: Bacula 2.0.2 (28Jan07): 21-Feb-2007 12:58:17
  JobId:  2
  Job:VU0EM003_Archiv.2007-02-21_10.50.44
  Backup Level:   Full
  Client: "VU0EM003-Archive" 2.0.2 (28Jan07) 
x86_64-unknown-linux-gnu,debian,4.0
  FileSet:"VU0EM003 Full Projekte" 2007-02-21 10:50:44
  Pool:   "Archiv" (From Job resource)
  Storage:"NEC-T40A" (From Job resource)
  Scheduled time: 21-Feb-2007 10:50:08
  Start time: 21-Feb-2007 10:50:47
  End time:   21-Feb-2007 12:58:17
  Elapsed time:   2 hours 7 mins 30 secs
  Priority:   10
  FD Files Written:   463,754
  SD Files Written:   463,754
  FD Bytes Written:   215,709,046,419 (215.7 GB)
  SD Bytes Written:   215,806,131,401 (215.8 GB)
  Rate:   28197.3 KB/s
  Software Compression:   None
  VSS:no
  Encryption: no
  Volume name(s): 06D143L3
  Volume Session Id:  4
  Volume Session Time:1171980127
  Last Volume Bytes:  215,975,402,496 (215.9 GB)
  Non-fatal FD errors:0
  SD Errors:  0
  FD termination status:  OK
  SD termination status:  OK
  Termination:Backup OK





*use
The defined Catalog resources are:
 1: MyCatalog
 2: ArchiveCatalog
Select Catalog resource (1-2): 2
Using Catalog "ArchiveCatalog"
Using Catalog name=ArchiveCatalog DB=bacula_archive
*list volumes
Pool: Default
No results to list.
Pool: CleaningPool
No results to list.
Pool: Archiv
+-++---+-+-+--+
| mediaid | volumename | volstatus | enabled | volbytes| volfiles |
+-++---+-+

Re: [Bacula-users] How to evaluate number of JobQueue-Entries?

[Ralf Gross]

Jonas Wueste schrieb:
> The problem is that bconsole only can be run with root rights.
> I want to transmit the querys result via nrpe to my nagios-server and 
> the nrpe user on the local machine has no root privileges.
> So, is there a possibility  to get the value without root privileges?

Change the permissions, you don't have to run bconsole as root.

VU0EM005:/opt/bacula/sbin# su - bacula
Kein Verzeichnis, Anmeldung mit HOME=/
[EMAIL PROTECTED]:/$ bconsole
Connecting to Director VU0EM005:9101
1000 OK: VU0EM005 Version: 2.0.2 (28 January 2007)
Enter a period to cancel a command.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Cannot label tape from bconsole

[Ralf Gross]

Alfonso Amato - Sintecnos srl schrieb:
> I agree and your position is true.
> 
> Can you tell me what method you have in mind for changing permissions? A 
> shell line inside /etc/init.d/bacula-sd.conf (I use Fedora) with the 
> chmod previously said or a different command?

If your system supports udev, you can write your own rule. This is for
a debian etch system.

/etc/udev/rules.d/010-local.rules

SUBSYSTEM=="scsi_generic", SYSFS{model}=="T40A",
SYMLINK+="NEC-T40A", GROUP="tape", MODE="0660"
KERNEL=="st*",  GROUP="tape", MODE="0660"
KERNEL=="nst*", GROUP="tape", MODE="0660

VU0EM005:~# ls -l /dev/NEC-T40A 
lrwxrwxrwx 1 root root 4 2007-02-09 15:18 /dev/NEC-T40A -> sg11
VU0EM005:~# ls -l /dev/sg11 
crw-rw 1 root tape 21, 11 2007-02-09 15:18 /dev/sg11

VU0EM005:~# ls -l /dev/nst*
crw-rw 1 root tape 9, 128 2007-02-09 15:18 /dev/nst0
...

You can get the necessary information with udevinfo.

udevinfo -a -p $(udevinfo -q path -n /dev/$YOURDEVICE)

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] force bacula to do a specific type of backup

[Ralf Gross]

Oliver Lehmann schrieb:
> Arno Lehmann wrote:
> 
> > Apart from that, you should examine the job 385 in more detail, and 
> > compare with 534. I'm quite sure there must be a significant difference 
> > - either the job name or the client are not identical, I think.
> > 
> > I don't have a running Bacula ready at the moment, but I *think* you can 
> > use 'llist jobid=...' to get that information. Or query the catalog 
> > directly if you prefer that.
> > 
> > If you don't find the reason, please post the information here, and also 
> > add your job definitions, related to this broblem job / client.
> 
> What is important here: I changed *nothing* in bacula, I only
> changed *many* files on the client. That was all. And during the
> nightly scheduled incremental backup bacula started a full backup. A
> modification in the bacula config took not place. All I can assume
> is bacula calculated the amount of files / size from the affected
> files and came to the conclusion that the incremental backup to do
> would be nearly like a full backup, so it changed the job by its own
> to make a full backup to have the next incremental backups much more
> smaller. That sounds reasonable, but I don't want that kind of
> consideration ;)

I'm quite sure bacula will never decide to update a inc. backup to a
full one because of the amount of data that has changed.

The FileSet Id's of the two jobs are different. If the FileSet did
change since te last backup, the inc. will be promoted to a full one.

> *llist jobid=385
>JobId: 385
[...]
>   PriorJobId: 0
>FileSetId: 7
>  FileSet: avocado-files
> 
> *llist jobid=534 
>JobId: 534
[...]
>   PriorJobId: 0
>FileSetId: 8
>  FileSet: avocado-files

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] FileSet excludes with regexdir/wilddir (was [bacula 2.0.2] ACL's missing after restore)

[Ralf Gross]

Hi,

Kern Sibbald schrieb:
> The above *probably* would have written (I am unsure what you really wanted) 
> had you written it as:
> 
>  FileSet {
>Name = "VU0EM003 Full"
>Include {
>  File = /
>  File = /server
>  File = /public
>  File = /home
>  Options {
>aclsupport = yes
>signature = MD5
> RegexDir = ".*000_KEIN_BACKUP.*"
> exclude = yes
>   }
> }
>  [snip]

I'm facing a problem with an other FileSet.

The following should backup everything in all 000_INS_ARCHIV directories below
/server/projekte. There is at least one directory that matches the
regex/wilddir expression at the moment.

# ls -l /server/projekte/EAT/test
insgesamt 0
dr-xr-s--- 2 ralfgro ve 6 2007-02-27 18:18 000_ARCHIV.2007-02-27
drwxr-s--- 2 ralfgro ve 6 2007-02-27 18:18 000_INS_ARCHIV
drwxr-sr-x 2 rootve 6 2007-02-28 15:24 000_KEIN_BACKUP
drwxrws--- 2 ralfgro ve 6 2007-02-27 18:05 test


FileSet {
  Name = "VU0EM003-Archiv"
  Include {
Options {
   signature = MD5
   aclsupport = yes
   #RegexDir = ".*000_INS_ARCHIV.*"
   #wilddir = "/server/projekte/EAT/test/000_INS_ARCHIV*"
   wilddir = "*000_INS_ARCHIV*"
}
Options {
   RegexDir = ".*"
   exclude = yes
}
File = /server/projekte
  }
[snip]


# ./bregex -f /tmp/regex
Enter regex pattern: .*000_INS_ARCHIV.*
1: /server/projekte/EAT/test/000_INS_ARCHIV

# ./bwild -f /tmp/regex
Enter a wild-card: *000_INS_ARCHIV*
1: /server/projekte/EAT/test/000_INS_ARCHIV


*reload
*estimate job=VU0EM003-Archiv listing client=VU0EM003-Archiv  
fileset=VU0EM003-Archiv level=full
Using Catalog "ArchiveCatalog"
Connecting to Client VU0EM003-Archiv at 10.60.1.252:9102
drwxr-xr-x  11 root root   125 2007-02-18 10:57:58  /server/projekte
2000 OK estimate files=1 bytes=0

A real backup shows the same result.

I'm trying to follow this example:
http://www.bacula.org/rel-manual/FileSet_Resource.html#SECTION00153

FileSet {
  Name = "Full Set"
  Include {
 Options {
wilddir = "/home/a*"
wilddir = "/home/b*"
 }
 Options {
RegexDir = ".*"
exclude = yes
 }
 File = /home
  }
}



What am I missing in my VU0EM003-Archiv FileSet?

Thanks,
Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] FileSet excludes with regexdir/wilddir (was [bacula 2.0.2] ACL's missing after restore)

[Ralf Gross]

Martin Simmons schrieb:
> >>>>> On Wed, 28 Feb 2007 16:26:10 +0100, Ralf Gross said:
> > 
> > I'm facing a problem with an other FileSet.
> > 
> > The following should backup everything in all 000_INS_ARCHIV directories 
> > below
> > /server/projekte. There is at least one directory that matches the
> > regex/wilddir expression at the moment.
> > 
> > # ls -l /server/projekte/EAT/test
> > insgesamt 0
> > dr-xr-s--- 2 ralfgro ve 6 2007-02-27 18:18 000_ARCHIV.2007-02-27
> > drwxr-s--- 2 ralfgro ve 6 2007-02-27 18:18 000_INS_ARCHIV
> > drwxr-sr-x 2 rootve 6 2007-02-28 15:24 000_KEIN_BACKUP
> > drwxrws--- 2 ralfgro ve 6 2007-02-27 18:05 test
> > 
> > 
> > FileSet {
> >   Name = "VU0EM003-Archiv"
> >   Include {
> > Options {
> >signature = MD5
> >aclsupport = yes
> >#RegexDir = ".*000_INS_ARCHIV.*"
> >#wilddir = "/server/projekte/EAT/test/000_INS_ARCHIV*"
> >wilddir = "*000_INS_ARCHIV*"
> > }
> > Options {
> >RegexDir = ".*"
> >exclude = yes
> > }
> > File = /server/projekte
> >   }
> > [snip]
> The problem is that the Options clauses for directories control two things:
> whether to back up of that directory itself and whether to look inside for
> things to back up.  Your second options clause stops it looking at any
> directory in /server/projekte :-(

I see what you mean. Because /server/projekte/EAT matches the ".*"
exclude first, the include of the directories below do never match.

This can be confusing.

I'm still curious if there is a way to do this with Options{}.
 
> I think it is impossible to do what you want with the Options resources.  You
> have to use a File clause that finds the archive directories each time, e.g.
> 
> File = "\\|find /server/projekte -type d -name '*000_INS_ARCHIV*'"

This is working fine, I added  an additional  '-maxdepth 3' option to
the find command. This will be sufficient to find all directories we
need.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Different pools for incremental and full backups without prior full backup

[Ralf Gross]

Christian Nolte schrieb:
> [snip]
> Is it somehow possible to tell bacula to store the full backup which is
> run because no previous full backup has been made into the 'Full' pool
> and not the 'Incremental'?

Yes, you can.  

http://www.bacula.org/rel-manual/Configuring_Director.html

FullPool=Full
DifferentialPool=Differential
IncrementalPool=Incremental

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] How to schedule Full backup 1st sun of Jan, apr, jul, oct?

[Ralf Gross]

Mike Hanby schrieb:
> 
> I'm curious how I would set up a schedule to do:
> Full backups on the 1rst Sunday of Jan, Apr, Jul, Oct (i.e. quarterly)
> Differentials on all sundays in between
> incremental on mon-sat

I use this for my schedule.

Schedule {
  Name = "Archiv"
  Run = Level=Full Pool=Archiv 1st sat jan,may,sep at 6:00
}

Looking at the 'show jobs' output, this schedule seems to do what I want.

...
hour=6 
mday=0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
28 29 30 
month=0 4 8 
wday=6 
wom=0 
woy=0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 
mins=0
...

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] recommended autoloader

[Ralf Gross]

John Drescher schrieb:
> > I'd like to purchase an autoloader to run with bacula.  I need to do up
> > to 2tb a night
> 2TB in one night. You may have to get an LTO3 autoloader to do that. I
> hope your have a gigabit link to the data storage. I remember doing a
> 2TB backup in the past but I am sure it around 15 to 20 hours to
> complete.

How usefull is spooling in this case? The amount of data here is
growing too.  At the moment I'm spooling the data to disk before
writing it to tape, but I'm only backing up a few hundred Gb's.

spooling throughput:
Job write elapsed time ... Transfer rate = 51.15 M bytes/second

tape throughput :
Despooling elapsed time ... Transfer rate = 70.29 M bytes/second

overall throughput:
Rate: 27022.4 KB/s

For 2TB I would need ~18h with spooling and ~11h without. I'm not sure
how critical shoe shining is for LTO3 drives. But it would be
definitly nice if one could spool and despool data at the same time.
So that bacula starts despooling to tape right after filling the
buffer with a certain amount of data. Depending on the disks that are
use for spooling this would raise the overall performance, maybe not
to the max. despool rate, but over the current overall throughput.  I
think there was a discussion about this on the list last year, and
maybe a feature request too.


To get on topic again:

I can recommend the NEC-T40A LTO-3 changer. It starts with 24 slots
and can be expanded to 40 slots with a simple licence key (2 keys,
each for 8 slots). It's working fine with bacula.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Could not connect to Storage daemon - ERR=No error

[Ralf Gross]

Jeff Dickens schrieb:
[snip]
> *m
> 16-Mar 21:25 jeff-t40-fd: jeff-t40.2007-03-16_21.24.41 Warning: 
> ../../lib/bnet.c:864 Could not connect to Storage daemon on 
> packrat:9103. ERR=No error
> Retrying ...
> *
> 
> The server just sits there like this:
> 
> Running Jobs:
>  JobId Level   Name   Status
> ==
>   5831 Differe  jeff-t40.2007-03-16_21.24.41 is waiting for Client 
> jeff-t40-fd to connect to Storage jeff-t40-filestore
> 
> 
> and the client like this:
> 
> jeff-t40-fd Version: 2.0.3 (06 March 2007)  VSS Linux Cross-compile Win32
> Daemon started 16-Mar-07 20:32, 0 Jobs run since started.
>  Heap: bytes=19,076 max_bytes=20,872 bufs=83 max_bufs=94
>  Sizeof: boffset_t=8 size_t=4 debug=0 trace=1
> 
> Running Jobs:
> JobId 5831 Job jeff-t40.2007-03-16_21.24.41 is running.
> System or Console Job started: 16-Mar-07 21:24
> Files=0 Bytes=0 Bytes/sec=0
> Files Examined=0
> SDSocket closed.
> 
> 
> Any idea what might be going on ?

Can you reach the bacula-sd on port 9102 from your client?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] wishlist: ldap

[Ralf Gross]

HM schrieb:
> 
> If not already requested anywhere, LDAP support would be a sexy feature
> (configuration, job definitions, etc.).

http://www.bacula.org/?page=feature-request

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Strange "ERROR" on successful backup

[Ralf Gross]

[EMAIL PROTECTED] schrieb:
> I already read about that and the "Heartbeat Interval" has already been
> setup and changed to 90 seconds. I have Heartbeat setup on both the SD and
> the FD of this server. I actually had to setup the Heartbeat to handle a
> different problem with the backup. The Heartbeat solved that problem but
> then caused/showed this problem. I did attach the trace file. I'm not sure
> why you didn't see it.

I can see the attached bacula-sd.trace file, so at least it reached the
list.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Cleanning all

[Ralf Gross]

Jorge Gonz?lez schrieb:
> 
> Bacula it's been tested. I have a script to cleanning database but I 
> don't know what is wrong because when I run an 'status all' command in 
> console then past jobs are shown.
> 
> The script does:
> - Stops daemons
> - drops database
> - creates database & tables
> - deletes /var/bacula and all content in where backups are been saved

At this point you can remove the state files that are located in the
bacula working dir.

http://thread.gmane.org/gmane.comp.sysutils.backup.bacula.general/31916

> - starts daemons

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Hi,

bacula-dir 2.0.3, bacula-sd 2.0.3, bacula-fd 2.0.2, debian etch amd64,
postgres 8.1.7

Bacula is running fine here for a while. But this weekend bacula
chose to use fresh tape from the scratch pool for job #2 of a diff backup
instead of appending the data to the tape it was writing to a few
minutes before by the first of the two jobs.

Here are the two jobs:

Job {
  Name = "SMTCZB0003"
  Type = Backup
  Client = SMTCZB0003
  FileSet = "SMTCZB0003"
  Schedule = "Regular Backup"
  Storage = NEC-T40A
  Messages = "Regular Backup"
  Pool = Default
  SpoolData=yes
  Write Bootstrap = "/opt/bacula/var/bacula/working/SMTCZB0003.bsr"
  Priority = 5
  RunBeforeJob = "/usr/bin/ssh [snip]"
  RunAfterJob = "/usr/bin/ssh [snip]"
}

Job {
  Name = "VU0EM003"
  Type = Backup
  Client = VU0EM003
  FileSet = "VU0EM003 Full"
  Schedule = "Regular Backup"
  Storage = NEC-T40A
  Messages = "Regular Backup"
  Pool = Default
  SpoolData=yes
  Write Bootstrap = "/opt/bacula/var/bacula/working/VU0EM003.bsr"
  Priority = 10
}

..and the schedule:

Schedule {
  Name = "Regular Backup"
  Run = Level=Full Pool=Full 1st sun at 00:05
  Run = Level=Differential Pool=Differential FullPool=Full 2nd-5th sun at 00:05
  Run = Level=Incremental Pool=Incremental DifferentialPool=Differential 
FullPool=Full tue-sat at 00:05
}

Both jobs have the same start time, but different priorities.


llist volumes shows 

 volretention: 2,678,400 <--- (31 days)
   voluseduration: 345,600   <--- (4 days)

for all tapes in the diff pool



This is what happend this weekend:

Job #1:

08-Apr 00:05 VU0EM005: BeforeJob: run command "/usr/bin/ssh [snip]"
08-Apr 00:07 VU0EM005: Start Backup JobId 121, 
Job=SMTCZB0003.2007-04-08_00.05.00
08-Apr 00:08 VU0EM005: 3307 Issuing autochanger "unload slot 4, drive 0" 
command.
08-Apr 00:09 VU0EM005: 3304 Issuing autochanger "load slot 5, drive 0" command.
08-Apr 00:10 VU0EM005: 3305 Autochanger "load slot 5, drive 0", status is OK.
08-Apr 00:10 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
08-Apr 00:10 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 5.
08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3" (/dev/nst0), 
all previous data lost.

That's fine, I expected volume 06D124L3 to be overwritten at this time.

08-Apr 00:10 VU0EM005: Max configured use duration exceeded. Marking Volume 
"06D124L3" as Used.

But I don't know what happend at this point. Why does bacula mark volume
06D124L3 as used right after the job started. The configured duration time
is 4 days, I checked this.

08-Apr 00:10 VU0EM005: Spooling data ...
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
/var/lib/nfs/rpc_pipefs
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /dev
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /sys
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /boot
08-Apr 00:12 VU0EM005: Job write elapsed time = 00:02:23, Transfer rate = 6.766 
M bytes/second
08-Apr 00:12 VU0EM005: Committing spooled data to Volume "06D124L3". Despooling 
968,636,754 bytes ...
08-Apr 00:12 VU0EM005: Despooling elapsed time = 00:00:07, Transfer rate = 
138.3 M bytes/second
08-Apr 00:12 VU0EM005: Sending spooled attrs to the Director. Despooling 
1,543,784 bytes ...
08-Apr 00:12 VU0EM005: AfterJob: run command "/usr/bin/ssh [snip]"
08-Apr 00:12 VU0EM005: Bacula 2.0.3 (06Mar07): 08-Apr-2007 00:12:47
  JobId:  121
  Job:SMTCZB0003.2007-04-08_00.05.00
  Backup Level:   Differential, since=2007-04-01 00:07:10
  Client: "SMTCZB0003" i686-pc-linux-gnu,redhat,Enterprise 
release
  FileSet:"SMTCZB0003" 2007-02-20 22:25:00
  Pool:   "Differential" (From Run pool override)
  Storage:"NEC-T40A" (From Job resource)
  Scheduled time: 08-Apr-2007 00:05:00
  Start time: 08-Apr-2007 00:07:08
  End time:   08-Apr-2007 00:12:47
  Elapsed time:   5 mins 39 secs
  Priority:   5
  FD Files Written:   4,807
  SD Files Written:   4,807
  FD Bytes Written:   966,711,641 (966.7 MB)
  SD Bytes Written:   967,611,394 (967.6 MB)
  Rate:   2851.7 KB/s
  Software Compression:   None
  VSS:no
  Encryption: no
  Volume name(s): 06D124L3
  Volume Session Id:  40
  Volume Session Time:1174319761
  Last Volume Bytes:  968,583,168 (968.5 MB)
  Non-fatal FD errors:0
  SD Errors:  0
  FD termination status:  OK
  SD termination status:  OK
  Termination:Backup OK

08-Apr 00:12 VU0EM005: Begin pruning Jobs.
08-Apr 00:12 VU0EM005: No Jobs found to prune.
08-Apr 00:12 VU0EM005: Begin pruning Files.
08-Apr 00:12 VU0EM005: Pruned Files from 1 Jobs for client SMTCZB0003 from 
catalog.
08-Apr 00:12 VU0EM005: End auto prune.




Job #2:

08-Apr 00:12 VU0EM005: Start Backup JobId 122, J

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Me wrote:
>[snip]
> Job #1:
> 
> 08-Apr 00:05 VU0EM005: BeforeJob: run command "/usr/bin/ssh [snip]"
> 08-Apr 00:07 VU0EM005: Start Backup JobId 121, 
> Job=SMTCZB0003.2007-04-08_00.05.00
> 08-Apr 00:08 VU0EM005: 3307 Issuing autochanger "unload slot 4, drive 0" 
> command.
> 08-Apr 00:09 VU0EM005: 3304 Issuing autochanger "load slot 5, drive 0" 
> command.
> 08-Apr 00:10 VU0EM005: 3305 Autochanger "load slot 5, drive 0", status is OK.
> 08-Apr 00:10 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
> 08-Apr 00:10 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 5.
> 08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3" 
> (/dev/nst0), all previous data lost.
> 
> That's fine, I expected volume 06D124L3 to be overwritten at this time.
> 
> 08-Apr 00:10 VU0EM005: Max configured use duration exceeded. Marking Volume 
> "06D124L3" as Used.
> 
> But I don't know what happend at this point. Why does bacula mark volume
> 06D124L3 as used right after the job started. The configured duration time
> is 4 days, I checked this.
> 
> 08-Apr 00:10 VU0EM005: Spooling data ...
> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
> /var/lib/nfs/rpc_pipefs
> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /dev
> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /sys
> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /boot
> 08-Apr 00:12 VU0EM005: Job write elapsed time = 00:02:23, Transfer rate = 
> 6.766 M bytes/second
> 08-Apr 00:12 VU0EM005: Committing spooled data to Volume "06D124L3". 
> Despooling 968,636,754 bytes ...
> 08-Apr 00:12 VU0EM005: Despooling elapsed time = 00:00:07, Transfer rate = 
> 138.3 M bytes/second
> 08-Apr 00:12 VU0EM005: Sending spooled attrs to the Director. Despooling 
> 1,543,784 bytes ...
> 08-Apr 00:12 VU0EM005: AfterJob: run command "/usr/bin/ssh [snip]"
[snip]

Nobody? I'm a bit lost, my old bacula install worked as expected with
a similar config, so I don't think that I configured bacula-dir wrong.
This new install worked for the last 3 months, other volumes were
recycled and reused as expected.

What could lead to a volume that is recycled and then marked as used
just a couple of seconds afterwards? I double checked, the volume use
duration of this volume is set to 4 days.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Arno Lehmann schrieb:
> >> [snip]
> >> Job #1:
> >>
> >> 08-Apr 00:05 VU0EM005: BeforeJob: run command "/usr/bin/ssh [snip]"
> >> 08-Apr 00:07 VU0EM005: Start Backup JobId 121, 
> >> Job=SMTCZB0003.2007-04-08_00.05.00
> >> 08-Apr 00:08 VU0EM005: 3307 Issuing autochanger "unload slot 4, drive 0" 
> >> command.
> >> 08-Apr 00:09 VU0EM005: 3304 Issuing autochanger "load slot 5, drive 0" 
> >> command.
> >> 08-Apr 00:10 VU0EM005: 3305 Autochanger "load slot 5, drive 0", status is 
> >> OK.
> >> 08-Apr 00:10 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
> >> 08-Apr 00:10 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 
> >> 5.
> >> 08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3" 
> >> (/dev/nst0), all previous data lost.
> >>
> >> That's fine, I expected volume 06D124L3 to be overwritten at this time.
> >>
> >> 08-Apr 00:10 VU0EM005: Max configured use duration exceeded. Marking 
> >> Volume "06D124L3" as Used.
> >>
> >> But I don't know what happend at this point. Why does bacula mark volume
> >> 06D124L3 as used right after the job started. The configured duration time
> >> is 4 days, I checked this.
> >>
> >> 08-Apr 00:10 VU0EM005: Spooling data ...
> >> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
> >> /var/lib/nfs/rpc_pipefs
> >> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
> >> /dev
> >> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
> >> /sys
> >> SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
> >> /boot
> >> 08-Apr 00:12 VU0EM005: Job write elapsed time = 00:02:23, Transfer rate = 
> >> 6.766 M bytes/second
> >> 08-Apr 00:12 VU0EM005: Committing spooled data to Volume "06D124L3". 
> >> Despooling 968,636,754 bytes ...
> >> 08-Apr 00:12 VU0EM005: Despooling elapsed time = 00:00:07, Transfer rate = 
> >> 138.3 M bytes/second
> >> 08-Apr 00:12 VU0EM005: Sending spooled attrs to the Director. Despooling 
> >> 1,543,784 bytes ...
> >> 08-Apr 00:12 VU0EM005: AfterJob: run command "/usr/bin/ssh [snip]"
> > [snip]
> > 
> > Nobody?
> 
> Looks like that...

Hey, you're not nobody ;)
 
> > I'm a bit lost,
> 
> So am I. I checked your probelm description, and I don't see anything 
> obvious. I didn't expect anything simple, to be fair :-)

Thanks.
 
> The one thing I would do in your situation is - observe things. If this 
> repeats, I would try to get catalog data whenever a job starts, perhaps 
> even a dump of the Media table or something... just so you have some 
> data to look at. After the error has occured, it's likely Bacula 
> modified the catalog entries.

Well, I've a backup of the bacula database from last weeks full backup
(2007-04-01).

This shows the same value for voluseduration, 345600 (4 days).

COPY media (mediaid, volumename, slot, poolid, mediatype, mediatypeid,
labeltype, firstwritten, lastwritten, labeldate, voljobs, volfiles,
volblocks, volmounts, volbytes, volparts, volerrors, volwrites,
volcapacitybytes, volstatus, enabled, recycle, volretention,
voluseduration, maxvoljobs, maxvolfiles, maxvolbytes, inchanger,
storageid, deviceid, mediaaddressing, volreadtime, volwritetime,
endfile, endblock, locationid, recyclecount, initialwrite,
scratchpoolid, recyclepoolid, "comment") FROM stdin;

5   06D124L35   5   LTO30   0
2007-02-24 23:45:04 2007-02-25 11:04:06 2007-02-24 23:45:04
3   16  230427  1   14865371136 0   0   230428
0   Recycle 1   1   2678400 345600  0   0   0
1   2   0   0
0   0   15  20140   0   \N  0   0
\N


Because recycling and marking of the volume happend after the job
started (at the same minute 00:10), I've no idea if it would make
sense to dump the db in a run before skript before every backup.

 
> > my old bacula install worked as expected with
> > a similar config, so I don't think that I configured bacula-dir wrong.
> > This new install worked for the last 3 months, other volumes were
> > recycled and reused as expected.
> > 
> > What could lead to a volume that is recycled and then marked as used
> > just a couple of seconds afterwards? I double checked, the volume use
> > duration of this volume is set to 4 days.
> 
> Ok, double-checking is always valuable :-) I found the message "Max 
> configured..." only once in the source, so it's safe to assume the 
> message is not used out of context.

Ok.
 
> A look at the source dosn't reveal anything extraordinary, at east to 
> me. Unless you asume that the catalog data and the current system time 
> might be messed up... is your system time correct, does your catalog 
> know about the time zone, and whatever esle might be important?

I'm using an ntp server, so date and time is ok. The syslog doesn't
show anything unusual at the time it happend.

bacula=> select current_timestamp;
  now
  ---
 

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Arno Lehmann schrieb:
> > Well, I've a backup of the bacula database from last weeks full backup
> > (2007-04-01).
> > 
> > This shows the same value for voluseduration, 345600 (4 days).
> > 
> > COPY media (mediaid, volumename, slot, poolid, mediatype, mediatypeid,
> > labeltype, firstwritten, lastwritten, labeldate, voljobs, volfiles,
> > volblocks, volmounts, volbytes, volparts, volerrors, volwrites,
> > volcapacitybytes, volstatus, enabled, recycle, volretention,
> > voluseduration, maxvoljobs, maxvolfiles, maxvolbytes, inchanger,
> > storageid, deviceid, mediaaddressing, volreadtime, volwritetime,
> > endfile, endblock, locationid, recyclecount, initialwrite,
> > scratchpoolid, recyclepoolid, "comment") FROM stdin;
> > 
> > 5   06D124L35   5   LTO30   0
> > 2007-02-24 23:45:04 2007-02-25 11:04:06 2007-02-24 23:45:04
> > 3   16  230427  1   14865371136 0   0   230428
> > 0   Recycle 1   1   2678400 345600  0   0   0
> > 1   2   0   0
> > 0   0   15  20140   0   \N  0   0
> > \N
> > 
> > 
> > Because recycling and marking of the volume happend after the job
> > started (at the same minute 00:10), I've no idea if it would make
> > sense to dump the db in a run before skript before every backup.
> 
> I admit my suggestion is unlikely to reveal anything useful because the 
> most crucial information here are the retention times and voluseduration 
> etc., and these don't change with volume access. But you'll need some 
> hard data...
 
I just can't figure out what config option would cause a volume to be
recycled and set to used at the same minute if the volduration time is
set to 4 days (but this wouldn't be the first case of PEBKAC...).

> ... (time settings look correct)
> > At them moment I would consider this as a bug, but I would like to
> > here Kern's opinion on this before I open a bug report.
> 
> ... to confirm that assumption and give Kern something to work on.
> 
> By the way, I forgot to mention something more or less obvious: Run the 
> DIR with debug output enabled at a fairly high level. Perhaps that gives 
> us a clue why the volume is marked used too soon.

I'm a bit afraid of the amount of debug data that bacula will produce
until this problem occurs next time. 
 
> Finally, I don't recall if you run multple concurrent jobs to differend 
> drives. If that's the case, it might be that the volume in question was 
> marked used because another job accessed it. I understand that there are 
> cases where job timing is critical, and the current reservation scheme 
> isn't waterproof against two jobs accessing a single volume simultaneously.

No concurrent jobs, no differend drives. Just 2 jobs that have the
same start time, but different priorities. 

Anyway, thanks for your suggestions fo far!

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Kern Sibbald schrieb:
> > > At them moment I would consider this as a bug, but I would like to
> > > here Kern's opinion on this before I open a bug report.
> > 
> > ... to confirm that assumption and give Kern something to work on.
> 
> >From this email, I would probably close the bug report right away.  I don't 
> have any idea what system it is, what version of Bacula, nor do I see any 
> output generated by Bacula.  "llist volume=xxx" is best not a raw copy of 
> some stdin.  Also, there is no job output showing what happened with the 
> volume before, during and after its status was changed.

Sorry Kern, you seem to have missed my first mail in this thread.

(Message-ID: <[EMAIL PROTECTED]>, Date:
2007-04-10 07:29:05 GMT)

I'll cut & paste it again. If it's still missing important information, please
let me know.

## 

Hi,

bacula-dir 2.0.3, bacula-sd 2.0.3, bacula-fd 2.0.2, debian etch amd64,
postgres 8.1.7

Bacula is running fine here for a while. But this weekend bacula
chose to use fresh tape from the scratch pool for job #2 of a diff backup
instead of appending the data to the tape it was writing to a few
minutes before by the first of the two jobs.

Here are the two jobs:

Job {
  Name = "SMTCZB0003"
  Type = Backup
  Client = SMTCZB0003
  FileSet = "SMTCZB0003"
  Schedule = "Regular Backup"
  Storage = NEC-T40A
  Messages = "Regular Backup"
  Pool = Default
  SpoolData=yes
  Write Bootstrap = "/opt/bacula/var/bacula/working/SMTCZB0003.bsr"
  Priority = 5
  RunBeforeJob = "/usr/bin/ssh [snip]"
  RunAfterJob = "/usr/bin/ssh [snip]"
}

Job {
  Name = "VU0EM003"
  Type = Backup
  Client = VU0EM003
  FileSet = "VU0EM003 Full"
  Schedule = "Regular Backup"
  Storage = NEC-T40A
  Messages = "Regular Backup"
  Pool = Default
  SpoolData=yes
  Write Bootstrap = "/opt/bacula/var/bacula/working/VU0EM003.bsr"
  Priority = 10
}

..and the schedule:

Schedule {
  Name = "Regular Backup"
  Run = Level=Full Pool=Full 1st sun at 00:05
  Run = Level=Differential Pool=Differential FullPool=Full 2nd-5th sun at 00:05
  Run = Level=Incremental Pool=Incremental DifferentialPool=Differential 
FullPool=Full tue-sat at 00:05
}

Both jobs have the same start time, but different priorities.

llist volumes shows 

 volretention: 2,678,400 <--- (31 days)
   voluseduration: 345,600   <--- (4 days)

for all tapes in the diff pool

This is what happend this weekend:

Job #1:

08-Apr 00:05 VU0EM005: BeforeJob: run command "/usr/bin/ssh [snip]"
08-Apr 00:07 VU0EM005: Start Backup JobId 121, 
Job=SMTCZB0003.2007-04-08_00.05.00
08-Apr 00:08 VU0EM005: 3307 Issuing autochanger "unload slot 4, drive 0" 
command.
08-Apr 00:09 VU0EM005: 3304 Issuing autochanger "load slot 5, drive 0" command.
08-Apr 00:10 VU0EM005: 3305 Autochanger "load slot 5, drive 0", status is OK.
08-Apr 00:10 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
08-Apr 00:10 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 5.
08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3" (/dev/nst0), 
all previous data lost.

That's fine, I expected volume 06D124L3 to be overwritten at this time.

08-Apr 00:10 VU0EM005: Max configured use duration exceeded. Marking Volume 
"06D124L3" as Used.

But I don't know what happend at this point. Why does bacula mark volume
06D124L3 as used right after the job started. The configured duration time
is 4 days, I checked this.

08-Apr 00:10 VU0EM005: Spooling data ...
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into 
/var/lib/nfs/rpc_pipefs
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /dev
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /sys
SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /boot
08-Apr 00:12 VU0EM005: Job write elapsed time = 00:02:23, Transfer rate = 6.766 
M bytes/second
08-Apr 00:12 VU0EM005: Committing spooled data to Volume "06D124L3". Despooling 
968,636,754 bytes ...
08-Apr 00:12 VU0EM005: Despooling elapsed time = 00:00:07, Transfer rate = 
138.3 M bytes/second
08-Apr 00:12 VU0EM005: Sending spooled attrs to the Director. Despooling 
1,543,784 bytes ...
08-Apr 00:12 VU0EM005: AfterJob: run command "/usr/bin/ssh [snip]"
08-Apr 00:12 VU0EM005: Bacula 2.0.3 (06Mar07): 08-Apr-2007 00:12:47
  JobId:  121
  Job:SMTCZB0003.2007-04-08_00.05.00
  Backup Level:   Differential, since=2007-04-01 00:07:10
  Client: "SMTCZB0003" i686-pc-linux-gnu,redhat,Enterprise 
release
  FileSet:"SMTCZB0003" 2007-02-20 22:25:00
  Pool:   "Differential" (From Run pool override)
  Storage:"NEC-T40A" (From Job resource)
  Scheduled time: 08-Apr-2007 00:05:00
  Start time: 08-Apr-2007 00:07:08
  End time:   08-Apr-2007 00:12:47
  Elapsed time:   5 mins 39 secs
  Priority:   5
  FD Files Written:   4,807
  SD Files Written:   4,807
  FD Byt

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Kern Sibbald schrieb:
> > The main question is, why did bacula mark volumes 06D124L3 as used
> > right > after the first job started when the voluseduration is set
> > to 4 days?
> 
> I'm not sure that recycling a volume resets its use duration.  Perhaps the 
> use 
> duration expired at about the time or just after when the volume was 
> recycled.

The volume use duration is 4 days, 2007-02-25 was the first and last
time the volume was in use (weekly diff. backup) before the diff.
backup last sunday night.


The Volume Use Duration directive defines the time period that the
Volume can be written beginning from the time of first data write to
the Volume. If the time-period specified is zero (the default), the
Volume can be written indefinitely. Otherwise, the next time a job
runs that wants to access this Volume, and the time period from the
first write to the volume (the first Job written) exceeds the
time-period-specification, the Volume will be marked Used, which means
that no more Jobs can be appended to the Volume, but it may be
recycled if recycling is enabled. Using the command  status dir
applies algorithms similar to running jobs, so during such a command,
the Volume status may also be changed. Once the Volume is recycled, it
will be available for use again.


If I understand the last sentence correct, the volume use duration
should be reset after the volume was recycled. Otherwise the volume
could never be used again after the max volume use duration expired.

I'm also confused about this order:

1. run before job
2. start job 121
3. recycle volume 06D124L
4. mark volume 06D124L as used
5. ---> write job #1 to volume 06D124

08-Apr 00:05 VU0EM005: BeforeJob: run command "/usr/bin/ssh [snip]"
08-Apr 00:07 VU0EM005: Start Backup JobId 121, 
Job=SMTCZB0003.2007-04-08_00.05.00
08-Apr 00:08 VU0EM005: 3307 Issuing autochanger "unload slot 4, drive 0" 
command.
08-Apr 00:09 VU0EM005: 3304 Issuing autochanger "load slot 5, drive 0" command.
08-Apr 00:10 VU0EM005: 3305 Autochanger "load slot 5, drive 0", status is OK.
08-Apr 00:10 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
08-Apr 00:10 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 5.
08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3" (/dev/nst0), 
all previous data lost.
08-Apr 00:10 VU0EM005: Max configured use duration exceeded. Marking Volume 
"06D124L3" as Used.
08-Apr 00:10 VU0EM005: Spooling data ...
[snip]
08-Apr 00:12 VU0EM005: Job write elapsed time = 00:02:23, Transfer rate = 6.766 
M bytes/second
08-Apr 00:12 VU0EM005: Committing spooled data to Volume "06D124L3". Despooling 
968,636,754 bytes ...


I've used the 'Volume Use Duration' parameter in bacula 1.38.5 for
about one year. It always worked as (at least I) expected.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Arno Lehmann schrieb:
> > Both jobs have the same start time, but different priorities.
> 
> This should meant that effectively only that one job was active. It 
> looks something went wrong in the volume selection process, or - ah, 
> inspiration! - you did a 'status dir' while the tape was already 
> reserved for the job but not actually accessed AND the DIR needed to 
> trigger recycling etc., which in turn activated the pruning / volume use 
> time checks... just a wild idea. Or, to finish the thought, you have a 
> misconfiguration, wrong volume metadata, or a catalog problem, perhaps
> time-related. Which is what we discussed and everything looks good.

Well, I've 2 cron jobs, one of them does a 'status dir' every day. But
this happend hours before the backup took place. I do this exactly to
trigger pruning etc. 

5 18 * * * /bin/echo -e "use Catalog=MyCatalog\n status dir" | 
/opt/bacula/sbin/bconsole -c /opt/bacula/etc/bconsole.conf -s >/dev/null
0 7 * * 1 /opt/scripts/bacula/bacula_weekly_status.sh 2>&1 >/dev/null

But I can't see how this could have caused the problem.

Ralf

BTW: I use 2 catalogs, but this volume is of course not part of the
second catalog.

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Ralf Gross schrieb:
> > > Both jobs have the same start time, but different priorities.
> > 
> > This should meant that effectively only that one job was active. It 
> > looks something went wrong in the volume selection process, or - ah, 
> > inspiration! - you did a 'status dir' while the tape was already 
> > reserved for the job but not actually accessed AND the DIR needed to 
> > trigger recycling etc., which in turn activated the pruning / volume use 
> > time checks... just a wild idea. Or, to finish the thought, you have a 
> > misconfiguration, wrong volume metadata, or a catalog problem, perhaps
> > time-related. Which is what we discussed and everything looks good.
> 
> Well, I've 2 cron jobs, one of them does a 'status dir' every day. But
> this happend hours before the backup took place. I do this exactly to
> trigger pruning etc. 
> 
> 5 18 * * * /bin/echo -e "use Catalog=MyCatalog\n status dir" | 
> /opt/bacula/sbin/bconsole -c /opt/bacula/etc/bconsole.conf -s >/dev/null
> 0 7 * * 1 /opt/scripts/bacula/bacula_weekly_status.sh 2>&1 >/dev/null

I forgot something. I use the nagios plugin to check the state every
30 minutes. This plugin also does a 'status dir current'.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Kern Sibbald schrieb:
> On Wednesday 11 April 2007 20:28, Ralf Gross wrote:
> > I'm also confused about this order:
> > 
> > 1. run before job
> > 2. start job 121
> > 3. recycle volume 06D124L
> > 4. mark volume 06D124L as used
> > 5. ---> write job #1 to volume 06D124
> > 
> > 08-Apr 00:05 VU0EM005: BeforeJob: run command "/usr/bin/ssh [snip]"
> > 08-Apr 00:07 VU0EM005: Start Backup JobId 121, 
> Job=SMTCZB0003.2007-04-08_00.05.00
> > 08-Apr 00:08 VU0EM005: 3307 Issuing autochanger "unload slot 4, drive 0" 
> command.
> > 08-Apr 00:09 VU0EM005: 3304 Issuing autochanger "load slot 5, drive 0" 
> command.
> > 08-Apr 00:10 VU0EM005: 3305 Autochanger "load slot 5, drive 0", status is 
> OK.
> > 08-Apr 00:10 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
> > 08-Apr 00:10 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 5.
> > 08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3" 
> (/dev/nst0), all previous data lost.
> > 08-Apr 00:10 VU0EM005: Max configured use duration exceeded. Marking 
> Volume "06D124L3" as Used.
> > 08-Apr 00:10 VU0EM005: Spooling data ...
> > [snip]
> > 08-Apr 00:12 VU0EM005: Job write elapsed time = 00:02:23, Transfer rate = 
> 6.766 M bytes/second
> > 08-Apr 00:12 VU0EM005: Committing spooled data to Volume "06D124L3". 
> Despooling 968,636,754 bytes ...
> > 
> As I said (or rather implied) perhaps when the volume was recycled the use 
> duration expired "before" the First Write Time was properly updated, or maybe 
> it wasn't even updated.  
> 
> If you can specifically show that to be the case, then this would be a bug.

How would I do this, any hints? I'm running the director with -d 200
at the moment. Is this enough debug level? Or could I lower the debug
level? BTW, is there a way to add a timestamp to the debug output?

I 'hope' the same problem will appear again next weekend when the next
volume will be recycled (06D125L3).

Pool: Differential (trimmed output)
+---+--+-+--++---+---+
|mediaid|volumename|volstatus|volbytes  |volretention|recycle|lastwritten   
 |
+---+--+-+--++---+---+
| 5 |06D124L3  |Used |   968,583,168| 2,678,400  |  1|2007-04-08 
00:12:44|
| 6 |06D125L3  |Used |17,191,802,880| 2,678,400  |  1|2007-03-11 
00:33:24|
| 7 |06D126L3  |Used | 6,299,725,824| 2,678,400  |  1|2007-03-19 
21:07:48|
| 8 |06D127L3  |Used | 9,633,254,400| 2,678,400  |  1|2007-03-25 
00:22:52|
|19 |06D137L3  |Append   | 3,497,001,984| 2,678,400  |  1|2007-04-08 
00:18:03|
+---+--+-+--++---+---+


Anything I can check before the next diff. backup on sunday? Any
special debug commands? 


> > I've used the 'Volume Use Duration' parameter in bacula 1.38.5 for
> > about one year. It always worked as (at least I) expected.
> 
> Agreed, but three things:
> 
> - Code is in constant change, and sometimes in changing it new bugs are 
> introduced.
> 
> - Version 2.0.x checks more carefully and more often for expired  volume 
> parameters than previous versions.  
> 
> - Maybe you just fell into a new untested case during the recycle.  Code can 
> work "correctly" for years, then something different happens, and bang things 
> go wrong.


Of course! If it's a bug I would like to help fix it. If it's a
missconfiguration, I would like to find the error.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Martin Simmons schrieb:
> >>>>> On Wed, 11 Apr 2007 20:28:48 +0200, Ralf Gross said:
> > 
> > Kern Sibbald schrieb:
> > > > The main question is, why did bacula mark volumes 06D124L3 as used
> > > > right > after the first job started when the voluseduration is set
> > > > to 4 days?
> > > 
> > > I'm not sure that recycling a volume resets its use duration.  Perhaps 
> > > the use 
> > > duration expired at about the time or just after when the volume was 
> > > recycled.
> > 
> > The volume use duration is 4 days, 2007-02-25 was the first and last
> > time the volume was in use (weekly diff. backup) before the diff.
> > backup last sunday night.
> > 
> > <http://www.bacula.org/rel-manual/FileSet_Resource.html#PoolResource>
> > The Volume Use Duration directive defines the time period that the
> > Volume can be written beginning from the time of first data write to
> > the Volume. If the time-period specified is zero (the default), the
> > Volume can be written indefinitely. Otherwise, the next time a job
> > runs that wants to access this Volume, and the time period from the
> > first write to the volume (the first Job written) exceeds the
> > time-period-specification, the Volume will be marked Used, which means
> > that no more Jobs can be appended to the Volume, but it may be
> > recycled if recycling is enabled. Using the command  status dir
> > applies algorithms similar to running jobs, so during such a command,
> > the Volume status may also be changed. Once the Volume is recycled, it
> > will be available for use again.
> > 
> > 
> > If I understand the last sentence correct, the volume use duration
> > should be reset after the volume was recycled. Otherwise the volume
> > could never be used again after the max volume use duration expired.
> 
> That is almost correct.
> 
> The volume use duration itself is never reset, because it holds the difference
> between two times.  Instead, the volume has a "firstwritten" timestamp, which
> is updated when the first job is written.

Ok, then something seems to be wrong with the firstwritten time of this volume.
It's still the date/time when the first differential job was written to
the tape.

*llist volume pool=Differential
  mediaid: 5
   volumename: 06D124L3
 slot: 5
   poolid: 5
mediatype: LTO3
 firstwritten: 2007-02-24 23:45:04
  lastwritten: 2007-04-08 00:12:44
labeldate: 2007-04-08 00:07:08
[...]

This is an example of an old tape from my 1.38.5 server:

*llist volumes pool=Bang
  mediaid: 25
   volumename: Bang1
 slot: 1
   poolid: 1
mediatype: AIT-2
 firstwritten: 2007-02-18 02:59:10
  lastwritten: 2007-02-18 03:12:47
labeldate: 2007-02-18 02:59:10
[snip]

This tape was reused every 2 weeks.
 
> "Max configured use duration exceeded" means that the firstwritten + use
> duration time is in the past, so as Kern said, this could happen if Bacula
> fails to update the firstwritten because of a bug.
> 
> Do you know when 06D124L3 was marked as recycled?  Look for Director messages
> like
> 
> Pruned nnn Jobs on Volume "06D124L3" from catalog.
> There are no Jobs associated with Volume "06D124L3". Marking it purged.
> Recycled volume "06D124L3"

I can't find any message about this in the log file.  There are only messages
about pruned files. Nothing about pruned volumes...

11-Apr 00:09 VU0EM005: Pruned Files from 1 Jobs for client VU0EM003 from catalog

This is very differnt from my old 1.38.5 install, where I can find the expected
messages about pruned volumes.


pool config:

Pool {
  Name = Differential
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Recycle Oldest Volume = yes
  Volume Use Duration = 96 hours
  Volume Retention = 31 days
}

double check:

  poolid: 5
name: Differential
 numvols: 5
 maxvols: 0
 useonce: 0
  usecatalog: 1
 acceptanyvolume: 0
volretention: 2,678,400
  voluseduration: 345,600
  maxvoljobs: 0
 maxvolbytes: 0
   autoprune: 1
 recycle: 1
pooltype: Backup
 labelformat: *
 enabled: 1
   scratchpoolid: 0
   recyclepoolid: 0
   labeltype: 0


> Note that the "all previous data lost" message comes from the SD, not the
> Director.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Ralf Gross schrieb:
> > Do you know when 06D124L3 was marked as recycled?  Look for Director 
> > messages
> > like
> > 
> > Pruned nnn Jobs on Volume "06D124L3" from catalog.
> > There are no Jobs associated with Volume "06D124L3". Marking it purged.
> > Recycled volume "06D124L3"
> 
> I can't find any message about this in the log file.  There are only messages
> about pruned files. Nothing about pruned volumes...
> 
> 11-Apr 00:09 VU0EM005: Pruned Files from 1 Jobs for client VU0EM003 from 
> catalog
> 
> This is very differnt from my old 1.38.5 install, where I can find the 
> expected
> messages about pruned volumes.
> 
> 
> pool config:
> 
> Pool {
>   Name = Differential
>   Pool Type = Backup
>   Recycle = yes
>   AutoPrune = yes
>   Recycle Oldest Volume = yes
>   Volume Use Duration = 96 hours
>   Volume Retention = 31 days
> }
> 
> double check:
> 
>   poolid: 5
> name: Differential
>  numvols: 5
>  maxvols: 0
>  useonce: 0
>   usecatalog: 1
>  acceptanyvolume: 0
> volretention: 2,678,400
>   voluseduration: 345,600
>   maxvoljobs: 0
>  maxvolbytes: 0
>autoprune: 1
>  recycle: 1
> pooltype: Backup
>  labelformat: *
>  enabled: 1
>scratchpoolid: 0
>recyclepoolid: 0
>labeltype: 0


As I can't find anything about pruned/pured volumes in the log file, I'm now
looking for volumes that should have be purged already.  But I can't find one.
(although the state of volume 06D124L3 was Recycle on 2007-04-01!)

Both clients that write to these volumes have these File/Job retention times.

  File Retention = 30 days
  Job Retention = 6 month

This is to keep the job information for the volume retention time of the
full pool. 


Example 1: volume 06D125L3, differential pool, volretention 31 days

Shouldn't this volume have been purged already? 

Pool: Differential (trimmed output)
+---+--+-+--++---+---+
|mediaid|volumename|volstatus|volbytes  |volretention|recycle|lastwritten   
 |
+---+--+-+--++---+---+
| 5 |06D124L3  |Used |   968,583,168| 2,678,400  |  1|2007-04-08 
00:12:44|
| 6 |06D125L3  |Used |17,191,802,880| 2,678,400  |  1|2007-03-11 
00:33:24|
| 7 |06D126L3  |Used | 6,299,725,824| 2,678,400  |  1|2007-03-19 
21:07:48|
| 8 |06D127L3  |Used | 9,633,254,400| 2,678,400  |  1|2007-03-25 
00:22:52|
|19 |06D137L3  |Append   | 3,497,001,984| 2,678,400  |  1|2007-04-08 
00:18:03|
+---+--+-+--++---+---+



*llist volumes pool=Differential
  mediaid: 6
   volumename: 06D125L3
 slot: 6
   poolid: 5
mediatype: LTO3
 firstwritten: 2007-03-10 23:55:04
  lastwritten: 2007-03-11 00:33:24
labeldate: 2007-03-10 23:55:04
  voljobs: 3
 volfiles: 19
volblocks: 266,489
volmounts: 1
 volbytes: 17,191,802,880
volerrors: 0
volwrites: 266,490
 volcapacitybytes: 0
volstatus: Used
  enabled: 1
  recycle: 1
 volretention: 2,678,400
   voluseduration: 345,600
   maxvoljobs: 0
  maxvolfiles: 0
  maxvolbytes: 0
inchanger: 1
  endfile: 18
 endblock: 4,146
 volparts: 0
labeltype: 0
storageid: 2
 deviceid: 0
   locationid: 0
 recyclecount: 0
 initialwrite: 
scratchpoolid: 0
recyclepoolid: 0
  comment: 



Example 2: volume 06D121L3, incremental pool, volretention 26 days

Pool {
  Name = Incremental
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Recycle Oldest Volume = yes
  Volume Use Duration = 20 days
  Volume Retention = 26 days
}

Pool: Incremental (trimmed output)
+---+--+-+--++-+-+-+
|mediaid|volumename|volstatus| volbytes |volretention| recycle |mediatype| 
lastwritten |
+---+--+-+--++-+-+-+
| 1 |06D121L3  | Full|17,805,699,072|  2,246,400 |   1 | LTO3| 
2007-02-25 16:26:36 |
| 2 |06D122L3  | Used|70,216,538,112|  2,246,400 |   1 | LTO3| 
2007-03-17 00:09:27 |
| 3 |06D123L3  | Used|17,931,110,400|  2,246,400 |   1 | LTO3| 
2007-04-07 00:08:24 |
| 4 |06D120L3  | Append  | 2,433,263,616|  2,246,400 |   1 | LTO3| 
2007-04-12 00:09:35 |
+---+--+-+--++-+-+-+

*llist volumes pool=Incremental
Sel

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Kern Sibbald schrieb:
> > Of course! If it's a bug I would like to help fix it. If it's a
> > missconfiguration, I would like to find the error.
> 
> The way I would proceed if I had the time to work on this would be to setup 
> a "dummy" backup, either on a separate machine, or with a different Job name 
> on your production machine.  I would then write a volume, then recycle it 
> (purge, ...), and at each step do a "llist volume=xxx" and see what happens.  
> You could even set a pretty short volume use duration and see what happens 
> when 1. that use duration is normally exceeded, and 2. what happens just 
> after a recycle and the use duration is exceeded from the previous 
> firstwritten.
> 
> That would show the evolution of the volume, and might reproduce the "bug".

I did set up a test job with these settings:

Pool {
  Name = Test
  Pool Type = Backup
  Recycle = yes
  AutoPrune = yes
  Recycle Oldest Volume = yes
  Volume Use Duration = 5 minutes
  Volume Retention = 30 minutes
}

Client {
  Name = VU0EM005-Test
  Address = 10.60.1.250
  FDPort = 9102
  Catalog = MyCatalog
  Password = "snip"
  File Retention = 30 minutes
  Job Retention = 60 minutes
  AutoPrune = yes
}

Job {
  Name = "VU0EM005-Test-1"
  Type = Backup
  Client = VU0EM005-Test
  FileSet = "VU0EM005-Test"
 Schedule = "Test"
  Storage = NEC-T40A
  Messages = Standard
  Pool = Test
  SpoolData=yes
  Priority = 5
}

Job {
  Name = "VU0EM005-Test-2"
  Type = Backup
  Client = VU0EM005-Test
  FileSet = "VU0EM005-Test"
 Schedule = "Test"
  Storage = NEC-T40A
  Messages = Standard
  Pool = Test
  SpoolData=yes
  Priority = 10
}

Schedule {
  Name = "Test"
  Run = Level=Full hourly at 0:05
  Run = Level=Full hourly at 0:20
  Run = Level=Full hourly at 0:35
  Run = Level=Full hourly at 0:50
}


the backup starts:


12-Apr 13:05 VU0EM005: Start Backup JobId 132, 
Job=VU0EM005-Test-1.2007-04-12_13.05.00
12-Apr 13:05 VU0EM005: 3307 Issuing autochanger "unload slot 19, drive 0" 
command.
12-Apr 13:06 VU0EM005: 3304 Issuing autochanger "load slot 20, drive 0" command.
12-Apr 13:06 VU0EM005: 3305 Autochanger "load slot 20, drive 0", status is OK.
12-Apr 13:06 VU0EM005: 3301 Issuing autochanger "loaded? drive 0" command.
12-Apr 13:06 VU0EM005: 3302 Autochanger "loaded? drive 0", result is Slot 20.
12-Apr 13:06 VU0EM005: Wrote label to prelabeled Volume "06D139L3" on device 
"LTO3" (/dev/nst0)
12-Apr 13:06 VU0EM005: Spooling data ...
12-Apr 13:07 VU0EM005: Job write elapsed time = 00:00:40, Transfer rate = 38.75 
M bytes/second
12-Apr 13:07 VU0EM005: Committing spooled data to Volume "06D139L3". Despooling 
1,551,808,127 bytes ...

[snip]
12-Apr 13:10 VU0EM005: Max configured use duration exceeded. Marking Volume 
"06D139L3" as Used.

at 13:25

*llist volume=06D139L3
  mediaid: 21
   volumename: 06D139L3
 slot: 20
   poolid: 10
mediatype: LTO3
 firstwritten: 2007-04-12 13:05:03
  lastwritten: 2007-04-12 13:08:25
labeldate: 2007-04-12 13:05:03
  voljobs: 2
 volfiles: 4
volblocks: 48,102
volmounts: 1
 volbytes: 3,103,220,736
volerrors: 0
volwrites: 48,103
 volcapacitybytes: 0
volstatus: Used
  enabled: 1
  recycle: 1
 volretention: 1,800
   voluseduration: 300
   maxvoljobs: 0
  maxvolfiles: 0
  maxvolbytes: 0
inchanger: 1
  endfile: 3
 endblock: 8,550
 volparts: 0
labeltype: 0
storageid: 2
 deviceid: 0
   locationid: 0
 recyclecount: 0
 initialwrite: 
scratchpoolid: 0
recyclepoolid: 0
  comment: 

[snip]
12-Apr 13:41 VU0EM005: Pruned 2 Jobs on Volume "06D139L3" from catalog.
12-Apr 13:41 VU0EM005: Recycled volume "06D139L3"

at 13:42

*llist volume=06D139L3
  mediaid: 21
   volumename: 06D139L3
 slot: 20
   poolid: 10
mediatype: LTO3
 firstwritten: 1970-01-01 01:00:00
  lastwritten: 2007-04-12 13:08:25
labeldate: 2007-04-12 13:05:03
  voljobs: 0
 volfiles: 0
volblocks: 0
volmounts: 1
 volbytes: 1
volerrors: 0
volwrites: 48,103
 volcapacitybytes: 0
volstatus: Recycle
  enabled: 1
  recycle: 1
 volretention: 1,800
   voluseduration: 300
   maxvoljobs: 0
  maxvolfiles: 0
  maxvolbytes: 0
inchanger: 1
  endfile: 3
 endblock: 8,550
 volparts: 0
labeltype: 0
storageid: 2
 deviceid: 0
   locationid: 0
 recyclecount: 1
 initialwrite: 
scratchpoolid: 0
recyclepoolid: 0
  comment: 



This looks ok to me.



I've no idea why none of the volumes in the incr/diff pools were pured until
now, there is at least no message about it in the log. 

So I manually started a differential backup.


12-Apr 14:23 VU0EM005: Start Backup JobId 138, Job=VU0EM003.2007-04-12_14.23.42
12-Apr 14:23 VU0EM005: Pr

Re: [Bacula-users] Why did bacula mark a volume as used (voluseduration)?

[Ralf Gross]

Arno Lehmann schrieb:
> > I've no idea why none of the volumes in the incr/diff pools were pured until
> > now, there is at least no message about it in the log. 
> > 
> > So I manually started a differential backup.
> > 
> > 
> > 12-Apr 14:23 VU0EM005: Start Backup JobId 138, 
> > Job=VU0EM003.2007-04-12_14.23.42
> > 12-Apr 14:23 VU0EM005: Pruned 3 Jobs on Volume "06D125L3" from catalog.
> > 12-Apr 14:23 VU0EM005: ua_purge.c:611 All records pruned from Volume 
> > "06D125L3"; marking it "Purged"
> > 12-Apr 14:23 VU0EM005: Recycled volume "06D125L3"
> > 
> > Now the expected volume 06D125L3 was marked as purged...
> 
> That is normal as the recycling/pruning process is triggered only when 
> Bacula looks for a volume. Pruning does not take place after the 
> retention times are expired but only when needed.
> 
> (This is to make sure your data is available as long as possible.) This 
> means that, as long as unused volumes are available or can be 
> automatically created, no recycling will take place.

I've read about that, but why was there no message about the recycled
volume from last weekend?

Martin Simmons wrote that this message comes from the sd.

08-Apr 00:10 VU0EM005: Recycled volume "06D124L3" on device "LTO3"
(/dev/nst0), all previous data lost.

It's possible that _I_ purged this volume manually and therefor no
message about pruning/purging appears in the log file.

Because the volumes are just beginning to be reused, it might be
completely ok that no volume were purged before.

It's funny, today the first "Prune" messages appear in the log.

12-Apr 15:26 VU0EM005: Pruned 14 Jobs on Volume "06D121L3" from catalog.
12-Apr 15:26 VU0EM005: Pruned 30 Jobs on Volume "06D122L3" from catalog.
12-Apr 15:26 VU0EM005: Recycled volume "06D121L3"


But all of this does not explain, why recycling and marking the volume
as used happend at the same minute  :(

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] question about dbcheck (orphaned Path records)

[Ralf Gross]

Hi,

bacula-dir 2.0.3, bacula-sd 2.0.3, bacula-fd 2.0.2, debian etch amd64,
postgres 8.1.7

I'm running dbcheck at the moment. The bacula postgres db is 1 GB. dbcheck
needs very long time to check the orphaned Path entries. 

Checking for orphaned Path entries. This may take some time!
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
[snip, same output for about an hour]

After one hour I stopped dbcheck and enabled the verbose output. It
shows the same output for every orphaned path (I only check the first 10
paths...)

Found 5472 orphaned Path records.
Print them? (yes/no): yes 
[snip]
/home/rg/mysql.bacula/benchmark/
/home/rg/mysql.bacula/
/home/rg/mysql.bacula/bacula/
/home/rg/mysql.bacula/lost+found/
/home/rg/mysql.bacula/mysql/
Found 5472 orphaned Path records.
Print them? (yes/no): yes
[snip]
/home/rg/mysql.bacula/benchmark/
/home/rg/mysql.bacula/
/home/rg/mysql.bacula/bacula/
/home/rg/mysql.bacula/lost+found/
/home/rg/mysql.bacula/mysql/
Found 5472 orphaned Path records.
Print them? (yes/no): yes

...and so on.


Orphaned Path records. This condition happens any time a directory is
deleted from your system and all associated Job records have been
purged. During standard purging (or pruning) of Job records, Bacula
does not check for orphaned Path records. As a consequence, over a
period of time, old unused Path records will tend to accumulate and
use space in your database. This check will eliminate them. It is
recommended that you run this check at least once a year.


I know that dbcheck needs some time. But the db is not very large yet
and all index are there. dbcheck seems to loop somehow.

bacula=> select * from pg_indexes where tablename='file';
 schemaname | tablename |   indexname| tablespace | 
indexdef  
+---+++---
 public | file  | file_fp_idx|| CREATE INDEX 
file_fp_idx ON file USING btree (filenameid, pathid)
 public | file  | file_jobid_idx || CREATE INDEX 
file_jobid_idx ON file USING btree (jobid)
 public | file  | file_pkey  || CREATE UNIQUE INDEX 
file_pkey ON file USING btree (fileid)
(3 Zeilen)

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] question about dbcheck (orphaned Path records)

[Ralf Gross]

Ralf Gross schrieb:
> bacula-dir 2.0.3, bacula-sd 2.0.3, bacula-fd 2.0.2, debian etch amd64,
> postgres 8.1.7
> 
> I'm running dbcheck at the moment. The bacula postgres db is 1 GB. dbcheck
> needs very long time to check the orphaned Path entries. 
> 
> Checking for orphaned Path entries. This may take some time!
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> Found 5472 orphaned Path records.
> [snip, same output for about an hour]

After I toggled the modify database flag, the 5472 orphaned path
records were deleted in one step and the complete dbcheck finished in
less then one minute.

Is it possible that dbcheck loops for a much longer time without the
modify flag enabled?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] question about dbcheck (orphaned Path records)

[Ralf Gross]

Kern Sibbald schrieb:
> > Is it possible that dbcheck loops for a much longer time without the
> > modify flag enabled?
> 
> Yes, I think there is a bug here.  I would appreciate it if you would submit 
> a 
> bug report ...

I'll open a bug report in the next couple of days. Thanks.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] appropriate data spool size for tape

[Ralf Gross]

Dave schrieb:
> I'm getting my Quantum Dlt 4000 going with bacula. I've included my 
> definitions below and the job that i'm running. I'm not getting very good 
> performance with the spool, the drive starts and stops quite frequently, it 
> is on an Adaptec Ultrawide 2940 scsi controller. I was wondering the optimal 
> size of a tape data spool?
> Thanks.
> [snip] 
> Maximum Spool Size = 1024
> Maximum Job Spool Size = 1024


Maximum Spool Size = size  Where size is a the maximum spool size for
all jobs specified in bytes.
Maximum Job Spool Size = size  Where size is the maximum spool file
size for a single job specified in bytes.

1024 = 1KB, this is not much for a spool file.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] appropriate data spool size for tape

[Ralf Gross]

Dave schrieb:
[pleas always answer to the list]
>Thanks for your reply. My brain is fuzzy, do you use data spooling and 
> if so what size value do you have? Do you also use the same tape drive as 
> i? If so, does my hardware definition look correct?

I've a LTO-3 drive and have no special hardware config in my
bacula-sd.conf. For spooling I use an extra RAID 0 device with ~440 GB
free disk space. I don't run concurrent jobs, so I can use the whole
device for a job.

Maximum Spool Size = 420g
Maximum Job Spool Size = 420g
Spool Directory = /bacula-spool/

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bacula-users email list

[Ralf Gross]

Jerry Amundson schrieb:
> On 4/16/07, Alan Brown <[EMAIL PROTECTED]> wrote:
> > On Sat, 14 Apr 2007, Kern Sibbald wrote:
> >
> > > This decision is motivated by the fact that the number of emails has 
> > > grown to
> > > be quite large, and hence to read them all requires a good deal of time.
> >
> > I think there is a big need for a bacula-newbies list...
> >
> > The same questions do tend to be cropping up repeatedly.
> 
> That, I think, would be better served by a decently organized and
> maintained Wiki/FAQ.

A (unofficial?) Wiki does already exist.

http://paramount.ind.wpi.edu/wiki/doku.php

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Why does dbcheck run forever, even with indexes?

[Ralf Gross]

Arno Lehmann schrieb:
> > Select function number: 16
> > Checking for Filenames with a trailing slash
> > Found 0 bad Filename records.
> > Checking for Paths without a trailing slash
> > Found 0 bad Path records.
> > Checking for duplicate Filename entries.
> > Found 0 duplicate Filename records.
> > Checking for duplicate Path entries.
> > Found 0 duplicate Path records.
> > Checking for orphaned JobMedia entries.
> > Found 0 orphaned JobMedia records.
> > Checking for orphaned File entries. This may take some time!
> > Found 103804 orphaned File records.
> > Checking for orphaned Path entries. This may take some time!
> > snip
> 
> There was recently a report that dbcheck loops infinitely when checking. 
> To me, it sounded like the change to have checks done in batches, not 
> for the whole database at once, doesn't quite work... if you set dbcheck 
> to fix problems it encounters, does it finish? (Make a dump before...)

I had this problem and opened a bug report (822). But my dbcheck
output looked different.

Checking for orphaned Path entries. This may take some time!
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
Found 5472 orphaned Path records.
[snip, same output for about an hour]

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Job history after db refresh

[Ralf Gross]

Kyle Tucker schrieb:
> Today I implemented an all-new backup scheme on my
> 2.0.2 Bacula installation. I ran drop_mysql_tables
> and then make_mysql_tables successfully and my first
> job started at job number 1. But if I run the command
> 'status client=' from bconsole, it has all the old
> jobs I ran from the previous db installation. Where
> is this old data coming from? 

Look for files in the bacula working directory that look like this:

bacula-dir.9101.state

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Incrementals upgraded to fulls

[Ralf Gross]

Kyle Tucker schrieb:
>   On my newly refreshed 2.0.2 Bacula installation,
> I ran fulls on all my clients initially. Last night the
> daily incrementals ran and all clients were upgraded to
> full as Bacula determined their was no previous full in
> the catalog for those clients, despite them all using the 
> same FileSet as the fulls. I do use different pools for
> fulls than incrementals. 
> 
> Would that cause Bacula to think no full had been done? 
> If so, can I work around this?
> If not, would this occur only initially or based on some
> schedule or retention settings?

Did you change the FileSet?


Any change to the list of the included files will cause Bacula to
automatically create a new FileSet (defined by the name and an MD5
checksum of the Include/Exclude contents). Each time a new FileSet is
created, Bacula will ensure that the next backup is always a Full
save.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] listing a catalog

[Ralf Gross]

Jean-François Leroux schrieb:
> I've created several catalogs in my testing setup. I'd like to browse these,
> now but I can't find the command to browse a given catalog only. Isn't it
> possible?

I'm not sure what you mean with 'browse', but you can switch the
catalog in bconsole with the 'use' command.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Force upgraded Jobs to another Pool?

[Ralf Gross]

Stern, Falk schrieb:
> 
> Is it possible to force upgraded jobs (upgraded from differential to
> full) to a different pool?

Yes, with FullPool, DifferentialPool...
 
> Currently differential jobs go to disk and full backups should go to
> tape (defined by schedule)
> 
> My schedule currently looks like this:
> 
> Schedule {
> Name= "WeeklyCycle"
> Run = Level=Full Pool=FullBackup first Sunday at 
> 1:00
> Run = Level=Differential Pool=DiffBackup 
> Monday-Saturday at 1:00
> Run = Level=Differential Pool=DiffBackup Sunday 
> at 1:00
> 
> }

Schedule {
  Name = "Regular Backup"
  Run = Level=Full Pool=Full 1st sun at 00:05
  Run = Level=Differential Pool=Differential FullPool=Full 2nd-5th sun at 00:05
  Run = Level=Incremental Pool=Incremental DifferentialPool=Differential 
FullPool=Full tue-sat at 00:05
}

 
Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Purge, relabel - more voodoo needed

[Ralf Gross]

John Drescher schrieb:
> > It purges (deletes) all Files from a Job,
> > JobId, Client or Volume; or it purges (deletes)
> > all Jobs from a Client or Volume without regard
> > for retention periods. Normally you should use the
> > PRUNE command, which respects retention periods.
> > You have the following choices:
> >  1: files
> >  2: jobs
> >  3: volume
> > Choose item to purge (1-3): 3
> > Automatically selected Pool: Default
> > No results to list.
> > Enter MediaId or Volume name: WHAT THE F**K DO YOU WANT ME TO TYPE
> > sql_get.c:954 Media record for Volume "WHAT THE F**K DO YOU WANT ME TO
> > TYPE" not found.
> > *
> 
> Ok.It looks like you have deleted all media from the Default pool. I
> assume the backup complains that the tape you want to use is already
> bacula labeled or something like that so it will not use it.
> 
> To get around this you need to wipe the label off the tape externally
> using the mt command:
> mt -f yourtapedrive rewind
> mt -f yourtapedrive weof
> 
> This is usually:
> 
> mt -f /dev/nst0 rewind
> mt -f /dev/nst0 weof
> 
> Then you should be able to label the tape in bacula.

To check if the tape really has an old bacula lable, he could also use
btape with the readlable command.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bweb

[Ralf Gross]

Lonny Selinger schrieb:
> I'm sure this has been covered but I can't for the life of me find
> it! ... I had bweb installed before and everything was working
> great. Then I decided to start from scratch and got everything
> installed again ... but now when I go to the list of jobs or click
> the "Jobs" link ... it takes a long time for it to come back and it
> comes back with nothing but a blank set. I've verified the bconsole
> command because I knew this was a thorn for some people:
> 
> 'bconsole' => '/usr/sbin/bconsole -n -c /etc/bacula/bconsole.conf',
> 
> # /usr/sbin/bconsole -n -c /etc/bacula/bconsole.conf Connecting to
> Director phantom2:9101 1000 OK: btrtg-dir Version: 2.0.3 (06 March
> 2007) Enter a period to cancel a command.  * 
> 
> I'm sure it's a silly path or permissions thing but I thought I'd
> throw it out there in case anyone else has any ideas. 

What database are you using? I think I had a similar problem when I
started from scratch. I forgot to import bweb-postgresql.sql again.

Any errors in the apache error log?

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bweb

[Ralf Gross]

Lonny Selinger schrieb:
> - "Ralf Gross" <[EMAIL PROTECTED]> wrote:
> > Lonny Selinger schrieb:
> > > I'm sure this has been covered but I can't for the life of me find
> > > it! ... I had bweb installed before and everything was working
> > > great. Then I decided to start from scratch and got everything
> > > installed again ... but now when I go to the list of jobs or click
> > > the "Jobs" link ... it takes a long time for it to come back and it
> > > comes back with nothing but a blank set. I've verified the bconsole
> > > command because I knew this was a thorn for some people:
> > > 
> > > 'bconsole' => '/usr/sbin/bconsole -n -c /etc/bacula/bconsole.conf',
> > > 
> > > # /usr/sbin/bconsole -n -c /etc/bacula/bconsole.conf Connecting to
> > > Director phantom2:9101 1000 OK: btrtg-dir Version: 2.0.3 (06 March
> > > 2007) Enter a period to cancel a command.  * 
> > > 
> > > I'm sure it's a silly path or permissions thing but I thought I'd
> > > throw it out there in case anyone else has any ideas. 
> > 
> > What database are you using? I think I had a similar problem when I
> > started from scratch. I forgot to import bweb-postgresql.sql again.
> 
> I'm running MySQL

I think my problme with bweb-postgresql.sql was at an other point of
my install. But I had the problem you described after trying the svn
code, I think it was after the 2.0.3 release. I switched back to the
bweb code which comes with 2.0.3.
  
> > Any errors in the apache error log?
> 
> There isn't a thing :( I had it tailed while I was trying to connect
> and nothing showed up. I was looking when it files through the web
> interface and also when it was successful running it as root from
> command line but neither post an error.
> 
> Thanks for the suggestions! ... I'll keep digging.

Did you enable the debug option in the bweb config file? Can your
apache user (www-data?) access the bacula files and execute bconsole?

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] One tape, multiple volumes

[Ralf Gross]

Leonardo Batista schrieb:
> 
> i need to make backup every day with 1 week retention but using just 1 DLT
> tape. What the better way to do this with a different volume to each backup?
> 
> if a restore is necessary, how mount the tape with a prev. volume?

In bacula, if you are using tapes, one volume is equivalent to one
tape. Maybe I missing something, but if you really have only one tape,
it will get hard to do a reasonable backup.


Ralf


-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bacula 2.0 scheduling problem

[Ralf Gross]

Jamie ffolliott schrieb:
> I have a problem with bacula incorrectly running the schedule, at least as I
> see it, and end up with two jobs getting scheduled for the same time.  
> Using bacula 2.0.1.  I believe this has persisted as a problem since 1.36,
> and has to do with using the week # in a schedule, and the result I'm
> getting looks like a bug to me.
>  
> My schedule has Elm1 set to run:
> Run = Level=Full Pool=pool1 w19 mon at 1:05
>  
> Elm3 has schedules to run:
> Run = Level=Differential Pool=pool3 w18 mon at 1:05
> Run = Level=Incremental Pool=pool3 w18 tue-sun at 1:05
>  
> However what I'm seeing on Monday May 7th, of week 19, is this on the
> director status:
>  
>  760 Differe  Elm3.2007-05-07_01.05.01 is waiting for Client elm-fd to
> connect to Storage carbon
>  762 FullElm1.2007-05-07_01.05.03 is waiting on max Client jobs
>  
> Notice both are scheduled for the same day, which is impossible per the
> above schedule.  I should only see Elm1 scheduled, not Elm3.
>  
> Is this because the director's scheduling ran on Sunday May 6th, of week 18?
> So it incorrectly made a match on the Elm3 schedule for Mon of week 18?
> (that would be a bug)

What does 'show schedules' show for these two schedules?

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bacula 2.0 scheduling problem

[Ralf Gross]

Jamie ffolliott schrieb:
> > What does 'show schedules' show for these two schedules?
> 
> For the correct schedule for Elm1, I see this:
> 
>  --> Pool: name=pool1 PoolType=Backup
>   use_cat=1 use_once=0 cat_files=1
>   max_vols=0 auto_prune=1 VolRetention=27 days
>   VolUse=0 secs recycle=1 LabelFormat=Vol-
>   CleaningPrefix=*None* LabelType=0
>   RecyleOldest=1 PurgeOldest=0 MaxVolJobs=0 MaxVolFiles=0
>   MigTime=0 secs MigHiBytes=0 MigLoBytes=0
>   --> Run Level=Full
>   hour=1
>   mday=0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
> 25 26 27 28 29 30
>   month=0 1 2 3 4 5 6 7 8 9 10 11
>   wday=1
>   wom=0 1 2 3 4
>   woy=19
>   mins=5
> 
> For the incorrect "differential" schedule for Elm3, I see this:
> 
>  --> Pool: name=pool3 PoolType=Backup
>   use_cat=1 use_once=0 cat_files=1
>   max_vols=0 auto_prune=1 VolRetention=27 days
>   VolUse=0 secs recycle=1 LabelFormat=Vol-
>   CleaningPrefix=*None* LabelType=0
>   RecyleOldest=1 PurgeOldest=0 MaxVolJobs=0 MaxVolFiles=0
>   MigTime=0 secs MigHiBytes=0 MigLoBytes=0
>   --> Run Level=Differential
>   hour=1
>   mday=0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
> 25 26 27 28 29 30
>   month=0 1 2 3 4 5 6 7 8 9 10 11
>   wday=1
>   wom=0 1 2 3 4
>   woy=18
>   mins=5
> 
> I've got a preset schedule designed for a 2-week rotation of 3 USB drives (6
> weeks of backups), allowing one drive to be rotated off site for 2 weeks at
> a time.  Different clients do their full/diff backup on different days of
> the week so that we can ensure it gets done during certain hours.
> 
> So to achieve this, I had to use the Week of Year.  The full output of 'show
> schedules' is quite long.  If you want the full configuration file for the
> schedules, I'll paste that.  However the above should be a pretty good
> example of the problem.

I don't see anything obvious that would indicate a configuration
error, maybe you should file a bug report.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] problem starting bacula director at boot and sd permission problem.

[Ralf Gross]

Don MacArthur schrieb:
> 
> I've been running bacula for like this for about a year (recently
> upgraded from 1.38.7).  I've always started the services manually, but
> now I have other admins working with the system also, and want it to
> autostart at boot.  
> 
> Symptoms:
> When I try to start the director via "bacula-ctl-dir start" I get the
> following error -
> "Starting the Bacula Director daemon
> 07-May 16:18 bacula-dir: ERROR TERMINATION at parse_conf.c:791
> Config error: Cannot open config file "/etc/bacula/bacula-dir.conf":
> Permission denied"

As which user did you start bacula-dir? The director doesn't have to
run as root user, but then you have to set the permissions correctly.
 
> My most extreme effort to solve this was to grant bacula ownership and
> to set the the permissions to 777.  But, it still fails.  

Only on the config file or on the whole path to the file?
 
> More Symptoms:
> When I ran the director as root I started the SD and FD via bacula-
> ctl-... and the backups failed because of permissions to the tape
> drive.  

If you start the sd as user bacula, the user must be in the tape
group (debian), and you should check if the tape group has access to
the tape device (I think 660 on /dev/nst0, or so). I had to create an
udev rule to get this right.
 
> So, do I need to change the bacula account (a group membership?), mod
> the objects it wants, or am I going in the wrong direction completely?

It's the right direction. Check the ownership/permissions of the tape
device and as which user your daemons run. The fd must run as root if
it should be able to backup all data on a system. The other daemons
can run as user bacula.

http://www.bacula.org/dev-manual/Bacula_Security_Issues.html#SECTION00463

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Rif: Bacula Network Performance

[Ralf Gross]

Jordan Desroches schrieb:
> So it doesn't look like the MySQL process is tanking the system at all. I
> increased cache sizes in MySQL with no effect, along with buffer sizes in
> sysctl.conf .  I also tried using SQLite , and got marginally better
> throughput. Has anyone seen significantly >30 MB/s over a gig line on
> backup?

Spooling transfer rate:
Job write elapsed time = 01:17:50, Transfer rate = 49.64 M bytes/second

Despooling to tape:
Despooling elapsed time = 00:55:10, Transfer rate = 70.11 M bytes/second

Average:
Rate:   26546.5 KB/s

I guess my average transfer rate could be higher, if I would not use
spooling. I'm not running concurrent jobs.

Ralf
~


-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula documents

[Ralf Gross]

Kern Sibbald schrieb:
> 
> Currently all the Bacula documents (web site, technotes, manual, manual 
> translations, ...) are all in the same SVN repository directory.  The result 
> is that when it is downloaded, and built to contain the final documents 
> needed by most release systems, the tar.gz file is over 30 MB, which means I 
> am no longer able to load it on Source Forge -- their FTP site drops the line 
> roughly every 10-20 MB.
> 
> As a consequence, I am considering to split it into a few different 
> directories.  My current thinking is:
> 
>  docs-en   English manual and developers guide
>  docs-fr French
>  docs-de   German
>   ...
>  docs-web  Web site documents
>  docs-techlogs  technical logs
> 
> Comments?

Sounds reasonable to me.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Spooling directory management

[Ralf Gross]

Ivan Adzhubey schrieb:
> 
> I have Bacula configured to use spooling to disk. I can see a few rather old 
> (timestamped from 6 months to 2 years ago) spool files sitting in the 
> spooling directory. Can I assume these are leftovers from some crashed backup 
> jobs and thus can I safely remove them? They use quite some space. My 
> understanding is that after a full successful backup cycle, Bacula should 
> leave spooling directory empty, right? It first spools, then dumps everything 
> to tape, then deletes spool files, am I correct?

That's how it works here, it should be ok to remove these files.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Feature: Reduce Backup Size on RAID

[Ralf Gross]

Mark Best schrieb:
> 
> I'm not sure what term is used to describe the feature but basically the
> backup server only saves 1 copy of a file. Using MD5 it checks to see if
> the file is different if its not it considers it to be the same file and
> doesn't backup multiple copies of the file. 
> 
> For example most *nixes have the /bin/ls binary file. Assuming it's the
> same 'ls' that another system has the backup server only keeps 1 copy of
> the binary but 2 entries in the Database - one for each PC. 
> ...

I don't think this is possible with bacula. BackupPC with its pooling
feature could be an alternative (maybe you know BackupPC already).

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Problems using smartmontools as Tape Alert Command

[Ralf Gross]

Ryan Novosielski schrieb:
> 
> Could someone tell me where I'm going wrong here? Here is my line that
> I've added to the config files, stolen directly from those who know what
> they're doing:
> 
> Alert Command = "sh -c '/opt/csw/sbin/smartctl -H -l error %c'"

Try something like this:

Alert Command = "sh -c 'smartctl -H -d scsi -l error %c'"

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Problems using smartmontools as Tape Alert Command

[Ralf Gross]

Ryan Novosielski schrieb:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
> 
> Ralf Gross wrote:
> > Ryan Novosielski schrieb:
> >> Could someone tell me where I'm going wrong here? Here is my line that
> >> I've added to the config files, stolen directly from those who know what
> >> they're doing:
> >>
> >> Alert Command = "sh -c '/opt/csw/sbin/smartctl -H -l error %c'"
> > 
> > Try something like this:
> > 
> > Alert Command = "sh -c 'smartctl -H -d scsi -l error %c'"
> 
> I'll certainly give it a shot, but doesn't it look like the %c is never
> getting expanded to anything? Does anyone know where in the manual these
> % variables are discussed?

It's in 'CHAPTER 28. AUTOCHANGER SUPPORT'
...
%c = changer device name
...

Of course you can use the device path instead of %c.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Problems using smartmontools as Tape Alert Command

[Ralf Gross]

Ryan Novosielski schrieb:
>  Alert Command = "sh -c '/opt/csw/sbin/smartctl -H -l error %c'"
> >>> Try something like this:
> >>>
> >>> Alert Command = "sh -c 'smartctl -H -d scsi -l error %c'"
> >> I'll certainly give it a shot, but doesn't it look like the %c is never
> >> getting expanded to anything? Does anyone know where in the manual these
> >> % variables are discussed?
> > 
> > It's in 'CHAPTER 28. AUTOCHANGER SUPPORT'
> > ...
> > %c = changer device name
> > ...
> > 
> > Of course you can use the device path instead of %c.
> 
> I don't have an autochanger. Would that mean I'd have to type the path
> in manually, or use a different variable?

First I would try the command on the command line. I think you can use
%a for the archive device.

14.3. DEVICE RESOURCE
Alert Command
...
Note, it is not necessary to have an autochanger to use this command.
The example below uses the tapeinfo program that comes with the
mtx package, but it can be used on any tape drive. However, you will
need to specify a Changer Device directive in your Device resource
(see above) so that the generic SCSI device name can be edited into
the command (with the %c).


Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] change FileSet name, where's the new filesetid?

[Ralf Gross]

Hi,

just for confirmation: if I change the name of a FileSet (not the file
list), the next job will be a full backup? 

>From the docs:
Any change to the list of the included files will cause Bacula to automatically
create a new FileSet (defined by the name and an MD5 checksum of the
Include/Exclude contents). Each time a new FileSet is created, Bacula will
ensure that the next backup is always a Full save.

Where can I see the current FileSet ID? With 'llist jobs' I get the filesetids
of completed jobs, but where can I check the current filesetid?

I tried to get it from the db.

bacula=> select * from fileset;
 filesetid |fileset|  md5   | createtime
---+---++-
 1 | SMTCZB0003| K98Oq//ZH2plCHkth6UKkC | 2007-02-20 22:25:00
 2 | VU0EM003 Full | yB/6TU1SCB+HyTpYeFwiGD | 2007-02-20 22:25:01
 3 | Catalog   | z4/v/7/uQ+xwMkF9v//4VD | 2007-02-25 00:10:00
 5 | VU0EM003 Full | I3oXN/QmSUZEs2/H7i+C4C | 2007-02-25 16:22:49
 7 | Test-Set  | 4H+m384yU9+n7//wR6/KjA | 2007-03-01 08:43:47
(5 Zeilen)

I changed the 'VU0EM003 Full' FileSet to 'VU0EM003' and issued reload in
bconsole before. But I can't find the modified FileSet name or id in the db.
Show fileset returns the correct modified name.

* show fileset
...
FileSet: name=VU0EM003
...


Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] change FileSet name, where's the new filesetid?

[Ralf Gross]

[resent, because of sf problems with the mailing lists]

Hi,

just for confirmation: if I change the name of a FileSet (not the file
list), the next job will be a full backup? 

>From the docs:
Any change to the list of the included files will cause Bacula to automatically
create a new FileSet (defined by the name and an MD5 checksum of the
Include/Exclude contents). Each time a new FileSet is created, Bacula will
ensure that the next backup is always a Full save.

Where can I see the current FileSet ID? With 'llist jobs' I get the filesetids
of completed jobs, but where can I check the current filesetid?

I tried to get it from the db.

bacula=> select * from fileset;
 filesetid |fileset|  md5   | createtime
---+---++-
 1 | SMTCZB0003| K98Oq//ZH2plCHkth6UKkC | 2007-02-20 22:25:00
 2 | VU0EM003 Full | yB/6TU1SCB+HyTpYeFwiGD | 2007-02-20 22:25:01
 3 | Catalog   | z4/v/7/uQ+xwMkF9v//4VD | 2007-02-25 00:10:00
 5 | VU0EM003 Full | I3oXN/QmSUZEs2/H7i+C4C | 2007-02-25 16:22:49
 7 | Test-Set  | 4H+m384yU9+n7//wR6/KjA | 2007-03-01 08:43:47
(5 Zeilen)

I changed the 'VU0EM003 Full' FileSet to 'VU0EM003' and issued reload in
bconsole before. But I can't find the modified FileSet name or id in the db.
Show fileset returns the correct modified name.

* show fileset
...
FileSet: name=VU0EM003
...


Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] change FileSet name, where's the new filesetid?

[Ralf Gross]

Ralf Gross schrieb:
> [resent, because of sf problems with the mailing lists]
> [...hey, the list is finally working again ;) ] 
> 
> just for confirmation: if I change the name of a FileSet (not the file
> list), the next job will be a full backup? 

The backup was upgraded to full, that's what I expected.
 
> Where can I see the current FileSet ID? With 'llist jobs' I get the filesetids
> of completed jobs, but where can I check the current filesetid?
> 
> I tried to get it from the db.
> 
> bacula=> select * from fileset;
>  filesetid |fileset|  md5   | createtime
> ---+---++-
>  1 | SMTCZB0003| K98Oq//ZH2plCHkth6UKkC | 2007-02-20 22:25:00
>  2 | VU0EM003 Full | yB/6TU1SCB+HyTpYeFwiGD | 2007-02-20 22:25:01
>  3 | Catalog   | z4/v/7/uQ+xwMkF9v//4VD | 2007-02-25 00:10:00
>  5 | VU0EM003 Full | I3oXN/QmSUZEs2/H7i+C4C | 2007-02-25 16:22:49
>  7 | Test-Set  | 4H+m384yU9+n7//wR6/KjA | 2007-03-01 08:43:47
> (5 Zeilen)
> 
> I changed the 'VU0EM003 Full' FileSet to 'VU0EM003' and issued reload in
> bconsole before. But I can't find the modified FileSet name or id in the db.
> Show fileset returns the correct modified name.

Looking at the psql output after the next backup happend, now there is a new
fileset in the list. VU0EM003, which is the one I changed before. The
createtime is the time the backup started. 

bacula=> select * from fileset;
 filesetid |fileset|  md5   | createtime
---+---++-
 1 | SMTCZB0003| K98Oq//ZH2plCHkth6UKkC | 2007-02-20 22:25:00
 2 | VU0EM003 Full | yB/6TU1SCB+HyTpYeFwiGD | 2007-02-20 22:25:01
 3 | Catalog   | z4/v/7/uQ+xwMkF9v//4VD | 2007-02-25 00:10:00
 5 | VU0EM003 Full | I3oXN/QmSUZEs2/H7i+C4C | 2007-02-25 16:22:49
 7 | Test-Set  | 4H+m384yU9+n7//wR6/KjA | 2007-03-01 08:43:47
 9 | VU0EM003  | I3oXN/QmSUZEs2/H7i+C4C | 2007-06-12 00:05:01
(6 Zeilen)


I'm still a bit confused why the new fileset wasn't created at the time I
changed the config and reloaded the daemon.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] exabyte autochanger

[Ralf Gross]

Maria McKinley schrieb:
> Falk Sauer wrote:
> > please make shure that your changer device has the correct permissions eg.:
> > 
> >  crw-rw root disk /dev/sg0
> > 
> > and for the potentially next problem ...
> > by your tapedrive device i'm unshure, i think this should /dev/nst0, i 
> > don't 
> > know how its correct on exabyte tapes.
> > 
> > Normally the /dev/st* device makes a automatic rewind after write, 
> > the /dev/nst* make no auto rewind. Bacula needs imho a non auto rewinding 
> > device. You dosn't write wich OS you use, here are little differences 
> > between 
> > the OSes.
> 
> My permissions are:
> 
> crw--- 1 root root 21,   0 2005-02-25 22:38 sg0
> 
> so, maybe that is my problem. Can I just change this, like any file, 
> with chown (assuming that the disk part is important) and chmod?

udev might override the permissions again. I would create an udev rule
to set the right permissions (check if your system uses udev).

You could try something like that:

/etc/udev/rules.d/010-local.rules 

KERNEL=="st*",  GROUP="tape", MODE="0660"
KERNEL=="nst*", GROUP="tape", MODE="0660"

/etc/init.d/udev restart (or reload...)

The bacula user has to be member of group tape.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Double backup

[Ralf Gross]

Lucio Crusca schrieb:
> I'm a newbie, and, believing it was a newbie question, a few days ago I wrote:
> http://sourceforge.net/mailarchive/forum.php?thread_name=200706132220.36816.lucio%40sulweb.org&forum_name=bacula-users
> 
> However I've received no reply... ok, I realize this is a mailing list, not a 
> help desk service, but just out of curiousity: it's that no one 
> knows what to reply, or it's that the reply would be RTFM or the like but 
> everyone thinks my question isn't even worth that kind of reply?
> 
> Sorry for insisting, but I have no clue about how to search that in the docs.

AFAIK there is no direct way to send the data to two storage daemons.
This was discussed a couple of times during the last months. Maybe the
new migration feature could be helpful.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Double backup

[Ralf Gross]

Lucio Crusca schrieb:
> Ralf Gross wrote:
> > AFAIK there is no direct way to send the data to two storage daemons.
> > This was discussed a couple of times during the last months. Maybe the
> > new migration feature could be helpful.
> 
> Thanks for the reply. Can you point me to those messages in the archives? The 
> SF seach server seems to be out of order at this time, and I wouldn't know 
> what to search anyway.

Try search the gmane archive.

http://news.gmane.org/gmane.comp.bacula.user

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Modifying Fileset changes incr to Full

[Ralf Gross]

Michel Meyers schrieb:
> Jean-François Leroux wrote:
> > Well, sorry, I must have an outdated doc. All I had was :
> > 'If this directive is set to *yes*, any changes you make to the FileSet
> > Include or Exclude lists will be ignored and not cause Bacula to
> > immediately perform a Full backup'
> 
> The sentence before that reads:
> 'Normally, if you modify the FileSet Include or Exclude lists, the next
> backup will be forced to a Full so that Bacula can guarantee that any
> additions or deletions are properly backed up. If this directive is set
> to yes, any changes you make to the FileSet Include or Exclude lists
> will be ignored and not cause Bacula to immediately perform a Full backup.'
> 
> This implies that the scope of the option is the 'forcing a full backup
> on changes', which is correct. The 'any changes [...] will be ignored'
> bit could be misinterpreted if read out of context.
> 
> Maybe we should change the sentence as follows?
> 
> 'If this directive is set to yes, any changes you make to the FileSet
> Include or Exclude lists will not cause Bacula to immediately perform a
> Full backup.'

...but include new files in the next regular backup. ?

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Modifying Fileset changes incr to Full

[Ralf Gross]

Michel Meyers schrieb:
> >> Maybe we should change the sentence as follows?
> >>
> >> 'If this directive is set to yes, any changes you make to the FileSet
> >> Include or Exclude lists will not cause Bacula to immediately perform a
> >> Full backup.'
> > 
> > ...but include new files in the next regular backup. ?
> 
> It should. The way I understood it is that Bacula first determines what
> job level to use. For that, it checks several things (schedule, is there
> a full backup that's valid, ...) and one of them is the MD5 hash of the
> FileSet definition. If that hash (stored at the time of the last
> backup?) doesn't match the current one, Bacula upgrades the backup to
> Full, unless 'Ignore Fileset Changes' is set. In the latter case, Bacula
> may (or may not) run an Incremental/Differential as usual while taking
> into account whatever you changed in the FileSet (ie if you added sth,
> it should see it as new and back it up, if you removed sth, it shouldn't
> back it up any more).

I think so too. Maybe I can test this tomorrow, so that we can add the
note about "...but include new files in the next regular backup" in
the docs.

Ralf

-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Slow FD

[Ralf Gross]

Gabriele Bulfon said:
> How do I know if it is half-duplex or full-duplex?

This link might help you:

http://www.sunmanagers.org/pipermail/summaries/2003-December/004745.html

Ralf


-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Issues concerning the production release of Bacula 1.39.x

[Ralf Gross]

Hi,

although I'm still using 1.38.5 and never tested 1.39.x I vote for #1.

I know the situation when one makes major changes right before going
on holiday. That's not a very pleasant feeling and it's hard to relax
in this time.

Ralf

-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Bacula gets tired.

[Ralf Gross]

Ian Leithhead schrieb:
> This seems odd.
> 
> Bacula job starts , connects to a remote client (no firewall, etc), and 
> storms away  transferring data very fast.
> 
> Then after some time or volume, it just stops. Packet sniffing indicates 
> very light (keep alive?) traffic, but no data transfer. 
> 
> Has anyone seen this before?

Are you spooling the data before writing to tape? 

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Error changing tapes in autochanger

[Ralf Gross]

Todd R Davis schrieb:
> I have recently moved our backup system from one machine running bacula
> 1.34 to another machine running 1.38.  All my database conversion seemed to
> go successful. Backups are working fine but when the tape fills and the
> autochanger needs to change tapes I get a message "bacula0-sd: 3995 Bad
> autochanger "unload slot 1, drive 0": ERR=Child exited with code 1"
> followed by a request to manually mount a different tape. When I manually
> mount the new tape the backup picks right up where it left off.  I read in
> the docs that it may be a permission issue??  I have checked the
> permissions to match our old system.  Any ideas?   Thanks.

I think there was a change in how changers are set up in the
bacula-sd.conf file in 1.38 (or before). 

example:

Autochanger {
  Name = Autochanger
  Device = AIT-2
  Changer Device = /dev/sg2
  Changer Command = "/opt/bacula/etc/mtx-changer %c %o %S %a %d"
  }

Device {
  Name = AIT-2
  Media Type = AIT-2
  Archive Device = /dev/nst0
  Autochanger = yes
  LabelMedia = no 
  AutomaticMount = yes
  AlwaysOpen = yes
}


Ralf

-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] bfill error - Attempt to WEOF on non-appendable Volume

[Ralf Gross]

Hi,

I'm testing bacula 1.39.28 with our new NEC-T40A changer (LTO3 Ultrium
drive). btape's test and fill commands were successful. Only bfill
finished with an error message.

+19-Nov 21:58 btape:
End of Volume "" at 425:8650 on device "NEC-T40A" (/dev/nst0). Write
of 64512 bytes got -1.
19-Nov 21:58 btape: Re-read of last block succeeded.

Write failed at block 6596150.
19-Nov 21:58 btape: Fatal Error at dev.c:1655 because:
dev.c:1654 Attempt to WEOF on non-appendable Volume
btape: btape.c:483 Bad status from weof. ERR=dev.c:1654 Attempt to
WEOF on non-appendable Volume


Is this something I have to worry about?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bfill error - Attempt to WEOF on non-appendable Volume

[Ralf Gross]

Ralf Gross schrieb:
> I'm testing bacula 1.39.28 with our new NEC-T40A changer (LTO3 Ultrium
> drive). btape's test and fill commands were successful. Only bfill
> finished with an error message.
> 
> +19-Nov 21:58 btape:
> End of Volume "" at 425:8650 on device "NEC-T40A" (/dev/nst0). Write
> of 64512 bytes got -1.
> 19-Nov 21:58 btape: Re-read of last block succeeded.
> 
> Write failed at block 6596150.
> 19-Nov 21:58 btape: Fatal Error at dev.c:1655 because:
> dev.c:1654 Attempt to WEOF on non-appendable Volume
> btape: btape.c:483 Bad status from weof. ERR=dev.c:1654 Attempt to
> WEOF on non-appendable Volume
> 
> 
> Is this something I have to worry about?

Any ideas on this? I just want to know if this is something I have to
take care about before I declare the T40A changer is working fine with
bacula.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bfill error - Attempt to WEOF on non-appendable Volume

[Ralf Gross]

Kern Sibbald said:
> On Wednesday 22 November 2006 08:12, Ralf Gross wrote:
>> Ralf Gross schrieb:
>> > I'm testing bacula 1.39.28 with our new NEC-T40A changer (LTO3
Ultrium
>> > drive). btape's test and fill commands were successful. Only bfill
finished with an error message.
>> >
>> > +19-Nov 21:58 btape:
End of Volume "" at 425:8650 on device "NEC-T40A" (/dev/nst0). Write
of 64512 bytes got -1.
>> > 19-Nov 21:58 btape: Re-read of last block succeeded.
>> >
>> > Write failed at block 6596150.
>> > 19-Nov 21:58 btape: Fatal Error at dev.c:1655 because:
>> > dev.c:1654 Attempt to WEOF on non-appendable Volume
>> > btape: btape.c:483 Bad status from weof. ERR=dev.c:1654 Attempt to
WEOF on non-appendable Volume
>> >
>> >
>> > Is this something I have to worry about?
>> Any ideas on this? I just want to know if this is something I have to
take care about before I declare the T40A changer is working fine with
bacula.
>
> Without knowing the details, this does not look too good.  It appears
more
> like a bug than a tape problem, but if I were you I would resolve the
problem
> before relying on the changer.  Possibly part of the problem is that you
have
> edited the output rather than presenting everything so I cannot see
everything.
>
> In any case, btape should not be getting this kind of error.  At a first
glance, it looks like btape attempted to write on the tape after
re-reading
> the last block, which is not what it normally does -- I suspect this is
because you edited the messages.  In any case, with the little snipit of
output it is impossible to get an idea of what is causing the problem.
>

Well, bfill doesn't give me much more output than the above.


--- bfill ---

*bfill
btape: btape.c:2536 Begin writing Bacula blocks of 64512 bytes.
+++
[bacula is filling the tape]
22-Nov 15:05 btape:
End of Volume "" at 427:8518 on device "LTO3" (/dev/nst0). Write of 64512
bytes got -1.
22-Nov 15:05 btape: Re-read of last block succeeded.

Write failed at block 6596018.
22-Nov 15:05 btape: Fatal Error at dev.c:1655 because:
dev.c:1654 Attempt to WEOF on non-appendable Volume
btape: btape.c:483 Bad status from weof. ERR=dev.c:1654 Attempt to WEOF on
non-appendable Volume



Here is btape's output:


--- test ---

./btape /dev/nst0
Tape block granularity is 1024 bytes.
btape: butil.c:272 Using device: "/dev/nst0" for writing.
22-Nov 13:08 btape: 3301 Issuing autochanger "loaded? drive 0" command.
22-Nov 13:08 btape: 3302 Autochanger "loaded? drive 0", result is Slot 1.
22-Nov 13:08 btape: 3301 Issuing autochanger "loaded? drive 0" command.
22-Nov 13:08 btape: 3302 Autochanger "loaded? drive 0", result is Slot 1.
btape: btape.c:356 open device "LTO3" (/dev/nst0): OK
*test

=== Write, rewind, and re-read test ===

I'm going to write 1000 records and an EOF
then write 1000 records and an EOF, then rewind,
and re-read the data to verify that it is correct.

This is an *essential* feature ...

btape: btape.c:813 Wrote 1000 blocks of 64412 bytes.
btape: btape.c:487 Wrote 1 EOF to "LTO3" (/dev/nst0)
btape: btape.c:829 Wrote 1000 blocks of 64412 bytes.
btape: btape.c:487 Wrote 1 EOF to "LTO3" (/dev/nst0)
btape: btape.c:838 Rewind OK.
1000 blocks re-read correctly.
Got EOF on tape.
1000 blocks re-read correctly.
=== Test Succeeded. End Write, rewind, and re-read test ===


=== Write, rewind, and position test ===

I'm going to write 1000 records and an EOF
then write 1000 records and an EOF, then rewind,
and position to a few blocks and verify that it is correct.

This is an *essential* feature ...

btape: btape.c:925 Wrote 1000 blocks of 64412 bytes.
btape: btape.c:487 Wrote 1 EOF to "LTO3" (/dev/nst0)
btape: btape.c:941 Wrote 1000 blocks of 64412 bytes.
btape: btape.c:487 Wrote 1 EOF to "LTO3" (/dev/nst0)
btape: btape.c:950 Rewind OK.
Reposition to file:block 0:4
Block 5 re-read correctly.
Reposition to file:block 0:200
Block 201 re-read correctly.
Reposition to file:block 0:999
Block 1000 re-read correctly.
Reposition to file:block 1:0
Block 1001 re-read correctly.
Reposition to file:block 1:600
Block 1601 re-read correctly.
Reposition to file:block 1:999
Block 2000 re-read correctly.
=== Test Succeeded. End Write, rewind, and re-read test ===



=== Append files test ===

This test is essential to Bacula.

I'm going to write one record  in file 0,
   two records in file 1,
 and three records in file 2

btape: btape.c:457 Rewound "LTO3" (/dev/nst0)
btape: btape.c:1559 W

Re: [Bacula-users] bfill error - Attempt to WEOF on non-appendable Volume

[Ralf Gross]

Kern Sibbald said:
> Well, I have no idea what bfill is.  It must be some silly hold over from
> long ago.   I just deleted it from the source code.
>
> If "fill" works, you are in good shape. Please ignore the bfill junk ...

Ok, very good.

I did a quick test against 1.38.11 where bfill seems to work.

*weof
btape: btape.c:469 Wrote 1 EOF to "LTO3" (/dev/nst0)

Ralf


-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Autochanger advice

[Ralf Gross]

christian zimmermann said:
> mount
> The defined Storage resources are:
>  1: File
>  2: Overland
> Select Storage resource (1-2): 2
> Enter autochanger drive[0]: 0
> 3301 Issuing autochanger "loaded drive 0" command.
> 3302 Autochanger "loaded drive 0", result: nothing loaded.
> 3301 Issuing autochanger "loaded drive 0" command.
> 3302 Autochanger "loaded drive 0", result: nothing loaded.
> 3301 Issuing autochanger "loaded drive 0" command.
> 3302 Autochanger "loaded drive 0", result: nothing loaded.
> 3902 Cannot mount Volume on Storage Device "Drive-1" (/dev/st0) because:
> Couldn't rewind device "Drive-1" (/dev/st0): ERR=dev.c:678 Rewind error on
> "Drive-1" (/dev/st0). ERR=No medium found.
>
> 3905 Device "Drive-1" (/dev/st0) open but no Bacula volume is mounted.
> If this is not a blank tape, try unmounting and remounting the Volume.
>
>
> I can mount the tape with mtx and label it without a problem I can even
> umount the tape with the bconsole. But I´m not able to mount it.
>
> Can someone point me in the right direction?
> I think I have a config error but not sure.

Did you run the btape tests?

Maybe you have to enabel the offline option in the mtx-changer script.

# enable the following line if you need to eject the cartridge
 mt -f $device offline
 sleep 10

Ralf



-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] find out the volume name in drive

[Ralf Gross]

Manuel Staechele said:
> how can i get the name of the volume which is in the drive without
> mounting the volume?
> is there a command in the bconsole etc.?

You can use the mtx-changer command for that.

# /opt/bacula/etc/mtx-changer /dev/NEC-T40A list 0 /dev/nst0 0
1:06D120L3
2:06D121L3
3:06D122L3
4:06D123L3
5:06D124L3
6:06D125L3
7:06D126L3
8:06D127L3
9:06D128L3
10:06D129L3
11:06D130L3
12:06D131L3
13:06D132L3
15:06D134L3
16:06D135L3
17:06D136L3
18:06D137L3
19:06D138L3
20:06D139L3
21:06D140L3
22:06D141L3
23:06D142L3
24:06D143L3
14:06D133L3

In this example volume 06D133L3 from slot 14 is loaded.

Ralf


-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] [Bacula-devel] Bacula 2.0.0 released to Source Forge

[Ralf Gross]

Kern Sibbald said:
> On Thursday 04 January 2007 14:23, James Harper wrote:
>> > - Data encryption is now implemented and is documented in
>> >   a new chapter of the manual at:
>> >   http://www.bacula.org/dev-manual/Bacula_Data_Encryption.html
>>
>> I think that link should be:
>> http://www.bacula.org/dev-manual/Data_Encryption.html
>> Bacula_Data_Encryption.html doesn't seem to exist.
>
> A shame -- too late now.

What about creating a symlink to Data_Encryption.html or a apache
rewrite/redirect?

Ralf




-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] compile bacula 2.0.1 with python support (debian)

[Ralf Gross]

Hi,

I'm trying to compile bacula on debian etch with python support. Debian's
python2.5-dev package is installed an I added 

--with-python=/usr/lib/python2.5/

to my configure options.
 
At the end of configure I get this status output:
Python support:  yes -L/usr/include/python2.5//config -lpython -lutil -lrt


make fails at some points with an error


/usr/bin/g++  -O -L../lib -L../findlib -o bacula-fd filed.o authenticate.o 
acl.o backup.o estimate.o filed_conf.o heartbeat.o job.o pythonfd.o restore.o 
status.o verify.o verify_vol.o \
   -lacl -lz   -lfind -lbac -lm 
-L/usr/include/python2.5//config -lpython -lutil -lrt  -lpthread -ldl  \
   -lwrap  -L/usr/lib//lib -lssl -lcrypto
/usr/bin/ld: cannot find -lpython
collect2: ld returned 1 exit status
make[1]: *** [bacula-fd] Fehler 1
make[1]: Leaving directory `/home/rg/bacula-2.0.1/src/filed'


I assume this is because there is no /usr/include/python2.5/config


ls -l /usr/include/python2.5/
insgesamt 572
-rw-r--r-- 1 root root 38971 2006-12-09 15:33 abstract.h
-rw-r--r-- 1 root root  1099 2006-12-09 15:33 asdl.h
-rw-r--r-- 1 root root   230 2006-12-09 15:33 ast.h
-rw-r--r-- 1 root root   792 2006-12-09 15:33 bitset.h
-rw-r--r-- 1 root root   914 2006-12-09 15:33 boolobject.h
-rw-r--r-- 1 root root   924 2006-12-09 15:33 bufferobject.h
-rw-r--r-- 1 root root   653 2006-12-09 15:33 cellobject.h
-rw-r--r-- 1 root root  4944 2006-12-09 15:33 ceval.h
-rw-r--r-- 1 root root  2893 2006-12-09 15:33 classobject.h
-rw-r--r-- 1 root root  1763 2006-12-09 15:33 cobject.h
-rw-r--r-- 1 root root  5014 2006-12-09 15:33 codecs.h
-rw-r--r-- 1 root root  3277 2006-12-09 15:33 code.h
-rw-r--r-- 1 root root  1058 2006-12-09 15:33 compile.h
-rw-r--r-- 1 root root  1476 2006-12-09 15:33 complexobject.h
-rw-r--r-- 1 root root  2005 2006-12-09 15:33 cStringIO.h
-rw-r--r-- 1 root root  8288 2006-12-09 15:33 datetime.h
[...]


I found a config directory in /usr/lib/python2.5/

ls -l /usr/lib/python2.5/config/
insgesamt 5296
-rw-r--r-- 1 root root2079 2006-12-09 15:33 config.c
-rw-r--r-- 1 root root1446 2006-12-09 15:33 config.c.in
-rwxr-xr-x 1 root root7122 2006-12-09 15:33 install-sh
[...]


Debian seems to put some files on different locations. I already tried some
compiler options, but couldn't get bacula to compile with python support yet.
Any idea what I have to change?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] compile bacula 2.0.1 with python support (debian)

[Ralf Gross]

Martin Simmons schrieb:
> >>>>> On Tue, 16 Jan 2007 11:21:00 +0100, Ralf Gross said:
> > 
> > Hi,
> > 
> > I'm trying to compile bacula on debian etch with python support. Debian's
> > python2.5-dev package is installed an I added 
> > 
> > --with-python=/usr/lib/python2.5/
> > 
> > to my configure options.
> 
> Did you try
> 
> --with-python
> 
> i.e. without the directory?  That works on Debian 3.1 (python2.3).

Hey, that was easy ;)

Python support:  yes -L/usr/lib64/python2.4/config -lpython2.4 -lutil -lrt

This is working now.

Thanks, Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] compile bacula 2.0.1 with python support (debian)

[Ralf Gross]

Martin Simmons schrieb:
> > > > I'm trying to compile bacula on debian etch with python support. 
> > > > Debian's
> > > > python2.5-dev package is installed an I added 
> > > > 
> > > > --with-python=/usr/lib/python2.5/
> > > > 
> > > > to my configure options.
> > > 
> > > Did you try
> > > 
> > > --with-python
> > > 
> > > i.e. without the directory?  That works on Debian 3.1 (python2.3).
> > 
> > Hey, that was easy ;)
> > 
> > Python support:  yes -L/usr/lib64/python2.4/config -lpython2.4 -lutil -lrt
> > 
> > This is working now.
> 
> Great.  Can you suggest any improvement to documentation that would have
> helped you?

Well, I think it was my own mistake. Nothing to blame the documentation for.

./configure --help | grep pyth
 --with-python=DIR   Include Python support.  DIR is the Python base

Thus I thought I had to add the path to python. Reading the whole description
makes it much clearer.

--with-python=DIR   Include Python support.  DIR is the Python base
install directory, default is to search through
a number of common places for the Python files.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] FileSet Options - does the order of directives matter ?

[Ralf Gross]

Markus Falb schrieb:
> ...snippel
>  Include {
>  Options {
>  signature=SHA1
>  verify = p
> snappel...
> 
> i get:
> 17-Jan 03:29 winch-dir: Verifying against JobId=354 Job=IDSarabisch. 
> 2007-01-17_02.44.23
> 17-Jan 03:29 winch-dir: Start Verify JobId=375 Level=Catalog  
> Job=IDSarabisch.2007-01-17_03.29.48
> 17-Jan 03:29 winch-dir: Bacula 2.0.0 (04Jan07): 17-Jan-2007 03:29:55
> 
> (no differences found)
> 
> when i change my Fileset and put the verify before the signature
> 
> ...snippel
> Include {
>  Options {
>  verify = p
>  signature=SHA1
>   snappel...
> 
> i get:
> 17-Jan 03:31 winch-dir: Verifying against JobId=354 Job=IDSarabisch. 
> 2007-01-17_02.44.23
> 17-Jan 03:31 winch-dir: Start Verify JobId=376 Level=Catalog  
> Job=IDSarabisch.2007-01-17_03.31.18
> 17-Jan 03:31 winch-dir: File: /root/test/
> 17-Jan 03:31 winch-dir:   st_mode  differ. Cat: 41c5 File: 41ed
> 17-Jan 03:31 winch-dir: IDSarabisch.2007-01-17_03.31.18 Error: Bacula  
> 2.0.0 (04Jan07): 17-Jan-2007 03:31:25

There have been some issues with the order of options in 2.0.0. Look
for a thread called: [Bacula-users] strange software compression
problem.

I think 2.0.1 solved the problem. From the relaese notes:

kes  Correct the Options scanner in the FD to correctly handle  
 SHA1 option, which was eating the next option.  Reported by
 Sebastien Guilbaud.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] bweb - print() on closed filehandle GEN0 at...

[Ralf Gross]

Hi,

I'm trying to setup bweb 2.0.1. I changed some config settings (paths, db), and
I can access http://myserver/cgi-bin/bweb/bweb.pl, but I get this error message
when accessing the client status or other functions:

[Wed Jan 17 09:25:26 2007] [error] [client x.60.5.110] print() on closed
filehandle GEN0 at /usr/lib/perl/5.8/IO/Handle.pm line 147., referer:
http://vu0em005/cgi-bin/bweb/bweb.pl?action=client

I get this error with the debian package and with bweb from the 2.0.1 tar
archive.

A second error message ist about png. I think I have installed all the
necessary debian packages...

[Wed Jan 17 09:27:55 2007] [error] [client x.60.5.111] Can't call method "png"
on an undefined value at /usr/lib/cgi-bin/bweb/bgraph.pl line 488., referer:
http://vu0em005/cgi-bin/bweb/bweb.pl?

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bweb - print() on closed filehandle GEN0 at...

[Ralf Gross]

Kern Sibbald schrieb:
> I've now installed bweb on a second machine, and have never run into a 
> problem 
> like you describe.  Perhaps Eric will have some idea on this, but one piece 
> of advice that I just got from Eric regarding a problem I had is to always 
> look in the Apache error_log when things go wrong. That often points to the 
> problem, though you might need to open one of the cgi-bin/bweb/xxx.pl files 
> to get down to more details.

I've done a fresh install from bacula-gui-2.0.1.tar.gz.

Some notes:

* hard coded path to bweb.conf in install_bweb (/etc/bacula)

I changed the config file path to /opt/bacula/etc, but after
executing install_bweb the wrong path was assigned to $config_file
in /usr/local/share/perl/5.8.8/Bweb.pm. The result is an error when
you try to http://your-server/bweb

* /var/www/bweb not created from the install_bweb script

# Root directory of web files
#
WEB_DIR=/var/www
  
If WEB_DIR should be the web root aka /var/www ...

if [ ! -d ${WEB_DIR} ] ; then
 mkdir ${WEB_DIR}/bweb
fi

...then this won't create the bweb directory. I changed ist to:

if [ ! -d ${WEB_DIR}/bweb ] ; then
 mkdir ${WEB_DIR}/bweb
fi

If WEB_DIR should be the location of the bweb web files, then
WEB_DIR should point to /var/www/bweb. Maybe the description is just a
bit misleading and it's clearer when you read the related section in
the INSTALL file.


Ok, now I'm at the same point as before. I can access
http://myserver/bweb and get the list of my 2 clients. But as soon as
I try to get the state of a client or the 'Defined Jobs' I get errors
in my apache error.log.

[Wed Jan 17 16:23:40 2007] [error] [client x.60.5.110] print() on
closed filehandle GEN0 at /usr/lib/perl/5.8/IO/Handle.pm line 147.,
referer: http://vu0em005/cgi-bin/bweb/bweb.pl?action=client

[Wed Jan 17 16:23:17 2007] [error] [client x.60.5.111] print() on
closed filehandle GEN0 at /usr/lib/perl/5.8/IO/Handle.pm line 147.,
referer: http://vu0em005/cgi-bin/bweb/bweb.pl?action=job


Since I see the client names, I assume that bweb has access to the db
and that this problem is related to bconsole (which www-data can
execute).


The second problem are the png graphs.

[Wed Jan 17 16:20:56 2007] [error] [client x.60.5.110] Can't call
method "png" on an undefined value at /usr/lib/cgi-bin/bweb/bgraph.pl
line 220., referer: http://vu0em005/cgi-bin/bweb/bweb.pl?action=graph


I've installed the debian packages that are mentioned in the INSTALL
file.

libgd-graph-perl libhtml-template-perl libexpect-perl
libdbd-mysql-perl libdbd-pg-perl libdbi-perl libtime-modules-perl

And some png libs.

libpng12-0
libpng12-dev
libpng3


Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bweb - print() on closed filehandle GEN0 at...]

[Ralf Gross]

Eric Bollengier schrieb:
> 
> > Kern Sibbald schrieb:
> > > I've now installed bweb on a second machine, and have never run into a
> > > problem like you describe.  Perhaps Eric will have some idea on this, but
> > > one piece of advice that I just got from Eric regarding a problem I had
> > > is to always look in the Apache error_log when things go wrong. That
> > > often points to the problem, though you might need to open one of the
> > > cgi-bin/bweb/xxx.pl files to get down to more details.
> >
> > I've done a fresh install from bacula-gui-2.0.1.tar.gz.
> >
> > Some notes:
> >
> > * hard coded path to bweb.conf in install_bweb (/etc/bacula)
> 
> I will move all bweb.conf references to Bweb.pm, and the install
> script will do a sed 's//' to adapt it.
> 
> > Ok, now I'm at the same point as before. I can access
> > http://myserver/bweb and get the list of my 2 clients. But as soon as
> > I try to get the state of a client or the 'Defined Jobs' I get errors
> > in my apache error.log.
> 
> You can set debug to "on" in configuration, we will see sql queries and
> debug messages to error.log.

I don't see any additional error message in the error log. But there is now some
additional info on the webpage.

http://vu0em005/cgi-bin/bweb/bweb.pl?action=client

SELECT Name   AS name,
   Uname  AS uname,
   AutoPrune AS autoprune,
   FileRetention AS fileretention,
   JobRetention  AS jobretention
FROM Client 


Clients
NameSelect  DescAuto Prune  File Retention  Job Retention
Bangsparc-sun-solaris2.8,solaris,5.81   30 days 180 days
VU0EM0042.0.1 (12Jan07) x86_64-unknown-linux-gnu,debian,4.0 
1   30 days 180 days
VU0EM0052.0.1 (12Jan07) x86_64-unknown-linux-gnu,debian,4.0 
1   30 days 180 days
Actions  
Last jobs   Current jobs   Status   Stats  


If I then select VU0EM005 add click on 'Status', I immediately get this error
again:

http://vu0em005/cgi-bin/bweb/bweb.pl?client=VU0EM005&action.x=23&action.y=19&action=client_status

[Fri Jan 19 17:35:01 2007] [error] [client 53.60.5.110] print() on closed
filehandle GEN0 at /usr/lib/perl/5.8/IO/Handle.pm line 147., referer:
http://vu0em005/cgi-bin/bweb/bweb.pl?action=client

Then it takes some time and a nearly empty page is diplayed.


Client status : VU0EM005

  

> I think it's a problem with bconsole (in Bconsole.pm) It's look like
> wrong bconsole option or something like.
> 
> Can you give me your bweb.conf ?

R1 = bless( {
  'graph_font' => '/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf',
  'name' => undef,
  'config_file' => '/opt/bacula/etc/bweb.conf',
  'bconsole' => '/opt/bacula/etc/bconsole -n -c /opt/bacula/etc/bconsole.conf',
  'ach_list' => {
 'S1_L80' => bless( {
 'info' => {
 'drive' => 0,
 'io' => 0,
 'slot' => 0
   },
 'name' => 'S1_L80',
 'bweb' => undef,
 'device' => '/dev/changer',
 'drive' => [],
 'debug' => 0,
 'label' => {},
 'precmd' => 'sudo',
 'io' => [],
 'mtxcmd' => '/usr/sbin/mtx',
 'drive_name' => [
   'S1_L80_SDLT0',
   'S1_L80_SDLT1'
 ],
 'slot' => []
   }, 'Bweb::Autochanger' )
   },
  'password' => 'asd01',
  'template_dir' => '/usr/share/bweb/tpl',
  'dbi' => 'DBI:mysql:database=bacula',
  'error' => '',
  'debug' => 'on',
  'user' => 'bacula',
  'email_media' => '[EMAIL PROTECTED]'
}, 'Bweb::Config' );

 
If I add bweb.conf with the web config editor, I always get an error about
fv_write_path.

An error has occurred :

bad parameter : fv_write_path = []

 
> > The second problem are the png graphs.
> >
> > [Wed Jan 17 16:20:56 2007] [error] [client x.60.5.110] Can't call
> > method "png" on an undefined value at /usr/lib/cgi-bin/bweb/bgraph.pl
> > line 220., referer: http://vu0em005/cgi-bin/bweb/bweb.pl?action=graph
> 
> With the debug=on option, you will see SQL queries on error.log
> You execute them into a mysql session, to see if you get data.
 
This is the debug output from bweb's start page
(http://vu0em005/cgi-bin/bweb/bweb.pl?)

, referer: http://vu0em005/cgi-bin/bweb/bweb.pl?
 SELECT, referer: http://vu0em005/cgi-bin/bweb/bweb.pl?
 UNIX_TIMESTAMP( DATE_FORMAT(StartTime, '%Y-%m-%d') ) AS A,, referer: 
http://vu0em005/cgi-bin/bweb/bweb.pl?
 sum(JobBytes)  AS nb, referer: 
http://vu0em005/cgi-bin/bweb/bweb.pl?
 FROM Job, Client, FileSet, referer: http://vu0em005/cgi-bin/bweb/bweb.pl?
 WHERE Job.ClientId = Client.ClientId, referer: 
http://vu0em005/cgi-bin/bweb/bweb.pl?
  AND Job.FileSetId = FileSet.FileSetId, referer: 
http://vu0em005/cgi-bin/bweb/bweb.pl?
  AND Job.Type = 'B', referer: http://vu0em005/cgi-bin/bweb/bweb.pl?
  , referer: http://vu0em005/cgi-bin/bweb/bweb.pl?
  , referer: http://vu0em005/cgi-bin/bweb/bweb.pl?
  , referer: http://

Re: [Bacula-users] bweb - print() on closed filehandle GEN0 at...]

[Ralf Gross]

Eric Bollengier schrieb:
> > If I then select VU0EM005 add click on 'Status', I immediately get this
> > error again:
> >
> > http://vu0em005/cgi-bin/bweb/bweb.pl?client=VU0EM005&action.x=23&action.y=1
> >9&action=client_status
> >
> > [Fri Jan 19 17:35:01 2007] [error] [client 53.60.5.110] print() on closed
> > filehandle GEN0 at /usr/lib/perl/5.8/IO/Handle.pm line 147., referer:
> > http://vu0em005/cgi-bin/bweb/bweb.pl?action=client
> >
> > Then it takes some time and a nearly empty page is diplayed.
> 
> This page use bconsole to get informations. Can you try this in a terminal 
> (xterm) ? (with www-data)
> 
> [EMAIL PROTECTED]:~$ /opt/bacula/etc/bconsole -n -c 
> /opt/bacula/etc/bconsole.conf
> * st client=VU0EM005

I've done this for an other client, this one has already been backed up. The
output on the client status page is the same (empty). 

[EMAIL PROTECTED]:~$ /opt/bacula/etc/bconsole -n -c 
/opt/bacula/etc/bconsole.conf
Connecting to Director VU0EM005:9101
1000 OK: VU0EM005 Version: 2.0.1 (12 January 2007)
Enter a period to cancel a command.
*st client=VU0EM004
Using Catalog "MyCatalog"
Connecting to Client VU0EM004 at VU0EM004-1:9102

VU0EM004 Version: 2.0.1 (12 January 2007)  x86_64-unknown-linux-gnu debian 4.0
Daemon started 17-Jan-07 09:52, 3 Jobs run since started.
 Heap: bytes=25,204,944 max_bytes=25,205,083 bufs=135,840 max_bufs=135,841
 Sizeof: boffset_t=8 size_t=8 debug=105 trace=0

Running Jobs:
JobId 4 Job VU0EM004.2007-01-19_23.40.51 is running.
Backup Job started: 19-Jan-07 23:40
Files=475,722 Bytes=25,236,423,357 Bytes/sec=27,401,111
Files Examined=475,722
Processing file: 
/data/BackupPC/pc/hogwards/0/f%2fhome%2fhogwards%2fknfe%2fdata2/fincoming/fvp0848_0445/fmac0080d90b02ed_v256_ARTWarnton-06-10-30-11-08-52.tgz
SDReadSeqNo=5 fd=5
Director connected at: 19-Jan-07 23:56


Terminated Jobs:
 JobId  LevelFiles  Bytes   Status   FinishedName 
==
 1  Full  5,221358.5 M  Cancel   19-Jan-07 23:18 VU0EM004
 2  Full146,0092.082 G  OK   19-Jan-07 23:29 VU0EM004
 3  Full146,0102.082 G  OK   19-Jan-07 23:39 VU0EM004

*

 
> > If I add bweb.conf with the web config editor, I always get an error about
> > fv_write_path.
> >
> > An error has occurred :
> >
> > bad parameter : fv_write_path = []
> 
> You can set it to /tmp
> 
> > > > The second problem are the png graphs.
> > Maybe the png error only occurres because there have not been any backups
> > yet?
> 
> Yes, to get graph, you must have data.

Understood ;) I didn't know what I should see there, only stats or other
(static) png's too.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bweb - print() on closed filehandle GEN0 at...]

[Ralf Gross]

Eric Bollengier schrieb:
> Can you put this into a test.pl file and execute it with the
> www-data user ? (chmod 755 test.pl ; ./test.pl)
> 
> ---8<---8<---8<---
> #!/usr/bin/perl -w
> use strict;
> use Bconsole;
> my $c = new Bconsole(pref => {
> bconsole => '/opt/bacula/etc/bconsole -n -c /opt/bacula/etc/bconsole.conf',
>   },
>  debug => 1);
> 
> print "fileset : ", join(',', $c->list_fileset()), "\n";
> print "job : ", join(',', $c->list_job()), "\n";
> print "storage : ", join(',', $c->list_storage()), "\n";
> 
> 
> print "status client :\n";
> $c->send_cmd("status client=VU0EM004");
> ---8<---8<---8<---

I've attached the debug output as file.
 
> It could be also a good idea to open a bug on bugs.bacula.org.

Ok, I'll open a bug report.

Ralf


debug.gz
Description: Binary data
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] bweb - print() on closed filehandle GEN0 at...]

[Ralf Gross]

Eric Bollengier schrieb:
> Can you retry the script using
> 
> /opt/bacula/sbin/bconsole -n -c /opt/bacula/etc/bconsole.conf
> instead of
> /opt/bacula/etc/bconsole -n -c /opt/bacula/etc/bconsole.conf 
> 

This works! Ok, this seems to be completely my fault. I'm always using
bconsole in the bacula/etc directory. But it's clear that bweb needs
the binary, not the script. bweb is now working as expected.

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] confused about job that ...is waiting for an appendable Volume

[Ralf Gross]

Hi,

today I started a backup job and had only one tape in the job's pool. After
some time bacula requested a new tape and I did label and mount a second one
(I'm still testing..)

Then I did a 'status dir' and got this message for a long time.

JobId Level   Name Status
==
13FullVU0EM004.2007-01-21_11.25.00 is waiting for an
   appendable Volume

But this couldn't be true because with dstat I saw that bacula was already
despooling (disk reads). Even after the spool file was removed the director
showed the above message. The log file shows that everything was working as
expected, only the director seems to have old status. Is there anything I'm
missing? 

21-Jan 11:25 VU0EM005: Spooling data ...
21-Jan 13:17 VU0EM005: Job write elapsed time = 01:52:00, Transfer rate = 34.54 
M bytes/second
21-Jan 13:17 VU0EM005: Committing spooled data to Volume "06D120L3". Despooling 
232,384,116,464 bytes ...
21-Jan 14:07 VU0EM005: End of Volume "06D120L3" at 426:9758 on device "LTO3" 
(/dev/nst0). Write of 64512 bytes got -1.
21-Jan 14:08 VU0EM005: Re-read of last block succeeded.
21-Jan 14:08 VU0EM005: End of medium on Volume "06D120L3" Bytes=425,958,220,800 
Blocks=6,602,774 at 21-Jan-2007 14:08.
21-Jan 14:08 VU0EM005: Job VU0EM004.2007-01-21_11.25.00 waiting. Cannot find 
any appendable volumes.
Please use the "label"  command to create a new Volume for:
Storage:  "LTO3" (/dev/nst0)
Media type:   LTO3
Pool: Default
21-Jan 14:11 VU0EM005: Volume "06D120L3" previously written, moving to end of 
data.
21-Jan 14:11 VU0EM005: Ready to append to end of Volume "06D120L3" at file=427.
21-Jan 14:11 VU0EM005: New volume "06D120L3" mounted on device "LTO3" 
(/dev/nst0) at 21-Jan-2007 14:11.
21-Jan 14:11 VU0EM005: End of Volume "06D120L3" at 427:1 on device "LTO3" 
(/dev/nst0). Write of 64512 bytes got -1.
21-Jan 14:11 VU0EM005: Re-read of last block succeeded.
21-Jan 14:11 VU0EM005: End of medium on Volume "06D120L3" Bytes=425,958,285,312 
Blocks=6,602,775 at 21-Jan-2007 14:11.
21-Jan 14:11 VU0EM005: Job VU0EM004.2007-01-21_11.25.00 waiting. Cannot find 
any appendable volumes.
Please use the "label"  command to create a new Volume for:
Storage:  "LTO3" (/dev/nst0)
Media type:   LTO3
Pool: Default
21-Jan 14:21 VU0EM005: Job VU0EM004.2007-01-21_11.25.00 waiting. Cannot find 
any appendable volumes.
Please use the "label"  command to create a new Volume for:
Storage:  "LTO3" (/dev/nst0)
Media type:   LTO3
Pool: Default
21-Jan 14:22 VU0EM005: Wrote label to prelabeled Volume "06D121L3" on device 
"LTO3" (/dev/nst0)
21-Jan 14:22 VU0EM005: New volume "06D121L3" mounted on device "LTO3" 
(/dev/nst0) at 21-Jan-2007 14:22.
21-Jan 14:32 VU0EM005: Despooling elapsed time = 01:00:50, Transfer rate = 
63.66 M bytes/second
21-Jan 14:38 VU0EM005: Sending spooled attrs to the Director. Despooling 
574,423,009 bytes ...
21-Jan 14:53 VU0EM005: Bacula 2.0.1 (12Jan07): 21-Jan-2007 14:53:29
  JobId:  13
  Job:VU0EM004.2007-01-21_11.25.00
  Backup Level:   Full
  Client: "VU0EM004" 2.0.1 (12Jan07) 
x86_64-unknown-linux-gnu,debian,4.0
  FileSet:"VU0EM004" 2007-01-19 23:40:51
  Pool:   "Default" (From Job resource)
  Storage:"NEC-T40A" (From Job resource)
  Scheduled time: 21-Jan-2007 11:24:55
  Start time: 21-Jan-2007 11:25:04
  End time:   21-Jan-2007 14:53:29
  Elapsed time:   3 hours 28 mins 25 secs
  Priority:   10
  FD Files Written:   1,618,365
  SD Files Written:   1,618,365
  FD Bytes Written:   231,736,482,663 (231.7 GB)
  SD Bytes Written:   232,136,134,130 (232.1 GB)
  Rate:   18531.5 KB/s
  Software Compression:   None
  VSS:no
  Encryption: no
  Volume name(s): 06D120L3|06D121L3
  Volume Session Id:  1
  Volume Session Time:1169374518
  Last Volume Bytes:  38,723,779,584 (38.72 GB)
  Non-fatal FD errors:0
  SD Errors:  0
  FD termination status:  OK
  SD termination status:  OK
  Termination:Backup OK



-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] BWeb / Bacula Web

[Ralf Gross]

Gavin Conway schrieb:
> Michel Meyers wrote:
> >>> I'm trying to locate the bacula-web component. It's listed as a feature
> >>> in bacula version 2.0 but it's not showing in the package nor is it
> >>> referenced or linked as a download anywhere on bacula.org .
> >>>
> >>> Any assistance is greatly appreciated
> >>>
> >> bacula-web is in the gui downnload.
> >>
> >> http://sourceforge.net/project/showfiles.php?group_id=50727&package_id=215994
> > 
> > Also note that bacula-web and bweb are two different things (bweb is the
> > much newer web-interface).
> > 
> > Greetings,
> > Michel
> 
> Thanks for that, it's the bweb variant I'm looking for then (bacula-web, 
> nice but dim). Again, any pointers to where to find bweb would be really 
> appreciated.

Both are part of the bacula-gui package.

http://sourceforge.net/project/showfiles.php?group_id=50727&package_id=215994

# ls -l ~rg/bacula-gui-2.0.1
insgesamt 216
drwxr-xr-x 2 root root   4096 2007-01-16 16:32 autoconf
-rw-r--r-- 1 root root  14452 2007-01-16 16:32 bacula-gui.spec
-rw-r--r-- 1 root root  14456 2007-01-01 15:49 bacula-gui.spec.in
drwxr-xr-x 9 root root   4096 2007-01-13 09:02 bacula-web
drwxr-xr-x 2 root root   4096 2007-01-16 16:32 bimagemgr
drwxr-xr-x 2 root root   4096 2007-01-13 09:02 brestore
drwxr-xr-x 9 root root   4096 2007-01-17 15:45 bweb
-rw-r--r-- 1 root root   5967 2007-01-16 16:32 config.log
-rw-r--r-- 1 root root160 2007-01-16 16:32 config.out
-rwxr-xr-x 1 root root  26308 2007-01-16 16:32 config.status
-rwxr-xr-x 1 root root 100265 2006-05-06 23:26 configure
drwxr-xr-x 2 root root   4096 2007-01-13 09:02 debian
-rw-r--r-- 1 root root   4715 2007-01-16 16:32 Makefile
-rwxr-xr-x 1 root root   3631 2006-08-20 19:05 Makefile.in
-rw-r--r-- 1 root root   1431 2005-10-28 22:52 README

Ralf

-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] Comm error with SD. bad response to Append Data.

[Ralf Gross]

Hi,

I'm having trouble with the backup of one of my clients since last week.


29-Jan 17:09 bacula-dir: Start Backup JobId 217,
Job=SMTCZB0003.2007-01-29_17.09.43

29-Jan 17:09 bacula-dir: Recycled volume "MKS3-2006-02-08"

29-Jan 17:09 bacula-sd: 3301 Issuing autochanger "loaded drive 0" command.

29-Jan 17:09 bacula-sd: 3302 Autochanger "loaded drive 0", result is Slot 12.

29-Jan 17:09 bacula-sd: 3301 Issuing autochanger "loaded drive 0" command.

29-Jan 17:09 bacula-sd: 3302 Autochanger "loaded drive 0", result is Slot 12.

29-Jan 17:10 bacula-sd: Recycled volume "MKS3-2006-02-08" on device
"AIT-2" (/dev/nst0), all previous data lost.

29-Jan 17:10 bacula-sd: Spooling data ...

SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into
/var/lib/nfs/rpc_pipefs

SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /dev

29-Jan 18:21 bacula-sd: User specified spool size reached.

29-Jan 18:21 bacula-sd: Writing spooled data to Volume. Despooling
37,580,970,446 bytes ...

29-Jan 19:32 bacula-sd: Spooling data again ...

SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /sys

SMTCZB0003-fd:  Filesystem change prohibited. Will not descend into /boot

29-Jan 20:37 bacula-sd: Committing spooled data to Volume
"MKS3-2006-02-08". Despooling 34,969,384,736 bytes ...

29-Jan 21:48 bacula-sd: Sending spooled attrs to the Director. Despooling
120,303,549 bytes ...


So far, so good. But 3 hours later the backup failed.



30-Jan 00:37 SMTCZB0003-fd: SMTCZB0003.2007-01-29_17.09.43 Fatal error:
job.c:1599 Comm error with SD. bad response to Append Data. ERR=Die
Verbindung wurde vom Kommunikationspartner zurückgesetzt

30-Jan 00:37 bacula-dir: SMTCZB0003.2007-01-29_17.09.43 Error: Bacula
1.38.5 (18Jan06): 30-Jan-2007 00:37:28

  JobId:  217

  Job:SMTCZB0003.2007-01-29_17.09.43

  Backup Level:   Full

  Client: "SMTCZB0003-fd"
i686-pc-linux-gnu,redhat,Enterprise release

  FileSet:"SMTCZB0003" 2006-06-25 08:00:02

  Pool:   "MKS"

  Storage:"Autochanger"

  Scheduled time: 29-Jan-2007 17:09:37

  Start time: 29-Jan-2007 17:09:45

  End time:   30-Jan-2007 00:37:28

  Priority:   10

  FD Files Written:   368,570

  SD Files Written:   368,570

  FD Bytes Written:   72,404,687,954

  SD Bytes Written:   72,472,374,146

  Rate:   2695.3 KB/s

  Software Compression:   None

  Volume name(s): MKS3-2006-02-08

  Volume Session Id:  1

  Volume Session Time:1170086393

  Last Volume Bytes:  72,536,797,943

  Non-fatal FD errors:0

  SD Errors:  0

  FD termination status:  Error

  SD termination status:  OK

  Termination:*** Backup Error ***

I can see that cpu load of the bacula server was high from beginning of
the backup until 22:45. Then it dropped nearly to zero. The last
successful backup took 50-60 min. for spooling the attributs to the
director. Thus the time from 21:48 to 22:45 would fit that window.

I found nothing interesting in the postgres log and it seems that bacula
finished at 22:46.

2007-01-29 22:38:29 [20740] LOG:  connection authorized: user=postgres
database=bacula
2007-01-29 22:41:34 [20742] LOG:  recycled transaction log file
"00150093"
2007-01-29 22:41:34 [20742] LOG:  recycled transaction log file
"00150094"
2007-01-29 22:43:32 [20745] LOG:  connection received: host=[local] port=
2007-01-29 22:43:32 [20745] LOG:  connection authorized: user=postgres
database=template1
2007-01-29 22:43:32 [20746] LOG:  connection received: host=[local] port=
2007-01-29 22:43:32 [20746] LOG:  connection authorized: user=postgres
database=bacula
2007-01-29 22:46:36 [20747] LOG:  recycled transaction log file
"00150095"
2007-01-29 22:46:36 [20747] LOG:  recycled transaction log file
"00150096"
2007-01-29 22:48:32 [20748] LOG:  connection received: host=[local] port=
2007-01-29 22:48:32 [20748] LOG:  connection authorized: user=postgres
database=template1
2007-01-29 22:48:32 [20749] LOG:  connection received: host=[local] port=
2007-01-29 22:48:32 [20749] LOG:  connection authorized: user=postgres
database=bacula
2007-01-29 22:53:32 [20753] LOG:  connection received: host=[local] port=
2007-01-29 22:53:32 [20753] LOG:  connection authorized: user=postgres
database=template1
2007-01-29 22:53:32 [20754] LOG:  connection received: host=[local] port=
2007-01-29 22:53:32 [20754] LOG:  connection authorized: user=postgres
database=bacula
2007-01-29 22:58:32 [20756] LOG:  connection received: host=[local] port=


I already tried an other tape from the pool, set the heartbeat interval to
60 and restarted the backup several times - same result. The host is
behind a firewall, but I'm not sure if the error message is an evidence of
a fw problem.

30-Jan 00:37 SMTCZB0003-fd: SMTCZB0003.2007-01-29_17.09.43 Fatal

Re: [Bacula-users] Comm error with SD. bad response to Append Data.

[Ralf Gross]

Ralf Gross schrieb:
> 30-Jan 00:37 SMTCZB0003-fd: SMTCZB0003.2007-01-29_17.09.43 Fatal error:
> job.c:1599 Comm error with SD. bad response to Append Data. ERR=Die
> Verbindung wurde vom Kommunikationspartner zur?ckgesetzt
> 
> (translation ;) ERR=The connection was reset by the communication partner)
> 
> 
> Backups of other hosts succeeded. Backups of this host were successful for
> the last 8 months.
> 
> Any idea what to look for? FD/SD Connect Timeout?

After setting the FD/SD timeout to 70 minutes the backup
succeeded.

Ralf

-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

[Bacula-users] settinge SD address in FD config?

[Ralf Gross]

Hi,

I'm setting up a new bacula server (bacula-dir, bacula-sd) which is
located at our data center. This server is connected to the client lan and
the backup lan with 2 network cards. The backup lan is a subnet that is
not accessible from the client lan. It's only used to backup server that
are located in the data center.

I want to backup some server that are located in the b-lan (S-FD) and some
clients in the c-lan (C-FD). This is working fine, however data from S-FD
in the b-lan are always sent to the c-lan address of the SD.

Setting the SD's address in bacula-sd.conf and bacula-dir.conf to the
b-lan address is working for the S-FD's, but not for C-FD because they
can't reach the b-lan.

 server
 __   ||
|  |  backup lan  | bacula-dir | bacula server
| S-FD |--| bacula-sd  |
|__|eth1  eth1||
   |eth0|eth0   __
   ||  |  |
   ||__| C-FD | client
client lan eth0|__|


What I'm missing is an option in the FD to set the SD's address. Maybe
depending on the DIR that started the backup in case there are more than
one DIR and SD.

Workaround at the moment:

* setting SD's addresse to the address in c-lan (eth0), so that it is
reachable from c-lan FD's.

* adding an extra route on S-FD that routes all traffic to the SD over the
b-lan (ip route add bacula-sd/32 dev eth1).


Ralf


-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Re: [Bacula-users] Comm error with SD. bad response to Append Data.

[Ralf Gross]

Erich Prinz schrieb:
> >> 30-Jan 00:37 SMTCZB0003-fd: SMTCZB0003.2007-01-29_17.09.43 Fatal  
> >> error:
> >> job.c:1599 Comm error with SD. bad response to Append Data. ERR=Die
> >> Verbindung wurde vom Kommunikationspartner zur?ckgesetzt
> >>
> >> (translation ;) ERR=The connection was reset by the communication  
> >> partner)
> >>
> >>
> >> Backups of other hosts succeeded. Backups of this host were  
> >> successful for
> >> the last 8 months.
> >>
> >> Any idea what to look for? FD/SD Connect Timeout?
> >
> > After setting the FD/SD timeout to 70 minutes the backup
> > succeeded.
>
> That's one slow link.

You mean the connection between FD and SD? While the SD is spooling
data I see that the transfer rate is ~60MiB/s.

The SD usually needs ~50 min. to send the attributes to the director.

29-Jan 20:37 bacula-sd: Committing spooled data to Volume
"MKS3-2006-02-08". Despooling 34,969,384,736 bytes ...
29-Jan 21:48 bacula-sd: Sending spooled attrs to the Director.
Despooling 120,303,549 bytes ...
30-Jan 00:37 SMTCZB0003-fd: SMTCZB0003.2007-01-29_17.09.43 Fatal
error: job.c:1599 Comm error with SD. bad response to Append Data.
ERR=Die Verbindung wurde vom Kommunikationspartner zur?ckgesetzt
30-Jan 00:37 bacula-dir: SMTCZB0003.2007-01-29_17.09.43 Error: Bacula
1.38.5 (18Jan06): 


With FD/SD timeout set to 70 minutes the backup was ok, it took only
53 minutes to send the attributes.

31-Jan 13:15 bacula-sd: Sending spooled attrs to the Director.
Despooling 120,355,627 bytes ...
31-Jan 14:08 bacula-dir: Bacula 1.38.5 (18Jan06): 31-Jan-2007 14:08:27

The default SD/FD timeout is 30 minutes, so this should have happend
before. Or is there any other 60 minutes timeout I'm missing?

Although setting the SD/FD timeout to 70 min. solved the problem, I'm
not sure what exactly happend.

Ralf

-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users