[Bacula-users] question about multiple file pools
Hi,
We have an old bacula setup with a single tape drive as part of an
autochanger, which holds three different pools for incremental,
differential and full backups. On restore, the correct tape is correctly
extracted from the database then mounted from the correct pool on our
single tape drive.
I'd like to replicate such functionnality on hard disks and a newer
release of bacula (2.4.4-1 under Debian Lenny, AMD64 architecture,
PostgreSQL backend).
During backups, all my pools and volumes are used as needed.
But during restore, what happens is that only volumes configured as
being part of the pool defined for the "RestoreFiles" job are
automatically mounted, and bacula waits for us to manually mount the
other volumes, which don't seem to be possible since they are file
volumes (i.e. always 'mounted')
Here's the message I've got in the logs :
myclient-dir JobId 29: Start Restore Job RestoreFiles.2009-11-16_10.45.10.03
myclient-dir JobId 29: Using Device "FileStorageFull"
myclient-sd JobId 29: Ready to read from volume "full0007" on device
"FileStorageFull" (/backups/full/).
myclient-sd JobId 29: Forward spacing Volume "full0007" to file:block 0:202.
myclient-sd JobId 29: End of Volume at file 0 on device "FileStorageFull"
(/backups/full/), Volume "full0007"
myclient-sd JobId 29: Warning: acquire.c:221 Read open device
"FileStorageFull" (/backups/full/) Volume "differential0009" failed:
ERR=dev.c:490 Could not open: /backups/full/differential0009, ERR=No
such file or directory
myclient-sd JobId 29: Please mount Volume "differential0009" for:
Job: RestoreFiles.2009-11-16_10.45.10.03
Storage: "FileStorageFull" (/backups/full/)
Pool: FullPool
Media type: File
As you can see bacula searches for volume differential0009 in
/backups/full instead of in /backups/differential, but my configuration
is as follows :
--- extracted from bacula-sd.conf ---
Device {
Name = FileStorageFull
Media Type = File
Archive Device = /backups/full/
LabelMedia = yes;
Random Access = Yes;
AutomaticMount = yes;
RemovableMedia = no;
AlwaysOpen = no;
}
Device {
Name = FileStorageDifferential
Media Type = File
Archive Device = /backups/differential/
LabelMedia = yes;
Random Access = Yes;
AutomaticMount = yes;
RemovableMedia = no;
AlwaysOpen = no;
}
Device {
Name = FileStorageIncremental
Media Type = File
Archive Device = /backups/incremental/
LabelMedia = yes;
Random Access = Yes;
AutomaticMount = yes;
RemovableMedia = no;
AlwaysOpen = no;
}
--- end of bacula-sd.conf ---
--- extracted from bacula-dir.conf ---
Storage {
Name = FileFull
Address = myclient.example.org
SDPort = 9103
Password = "yyy"
Device = FileStorageFull
Media Type = File
}
Storage {
Name = FileDifferential
Address = myclient.example.org
SDPort = 9103
Password = "yyy"
Device = FileStorageDifferential
Media Type = File
}
Storage {
Name = FileIncremental
Address = myclient.example.org
SDPort = 9103
Password = "yyy"
Device = FileStorageIncremental
Media Type = File
}
Pool {
Name = FullPool
Pool Type = Backup
Recycle = yes
AutoPrune = yes
Volume Retention = 70 days
Maximum Volume Bytes = 33554432
LabelFormat = "full"
Storage = FileFull
Catalog Files = yes
}
Pool {
Name = DifferentialPool
Pool Type = Backup
Recycle = yes
AutoPrune = yes
Volume Retention = 60 days
Maximum Volume Bytes = 33554432
LabelFormat = "differential"
Storage = FileDifferential
Catalog Files = yes
}
Pool {
Name = IncrementalPool
Pool Type = Backup
Recycle = yes
AutoPrune = yes
Volume Retention = 16 days
Maximum Volume Bytes = 33554432
LabelFormat = "incremental"
Storage = FileIncremental
Catalog Files = yes
}
Job {
Name = "RestoreFiles"
Type = Restore
Client = MyClient-fd
Messages = Standard
Where = /backups/restore
FileSet="FullUnixSet"
Pool = FullPool
}
Client {
Name = MyClient-fd
Address = myclient.example.org
FDPort = 9102
Catalog = MyCatalog
Password = "xxx"
File Retention = 45 days
Job Retention = 60 days
AutoPrune = yes
}
--- end of bacula-dir.conf ---
A solution would be for the "RestoreFiles" job to not have to specify
the pool, but this doesn't seem to be possible. Any reason for this ?
Anything wrong in my configuration ?
How to solve this big problem ?
Another small question :
As you can see above, I want to have very small volumes, but LOTS of
them. When labelling them automatically, bacula only uses 4 digits, what
will happen when it will reach ?
Thanks in advance for your help.
--
Jérôme Alet - - Centre de Ressources Informatiques
Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX
Tél : +687 266754 Fax : +687 254829
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report des
Re: [Bacula-users] question about multiple file pools
Hello, On Mon, Nov 16, 2009 at 02:35:12PM +1100, Jerome Alet wrote: > > We have an old bacula setup with a single tape drive as part of an > autochanger, which holds three different pools for incremental, > differential and full backups. On restore, the correct tape is correctly > extracted from the database then mounted from the correct pool on our > single tape drive. > > I'd like to replicate such functionnality on hard disks and a newer > release of bacula (2.4.4-1 under Debian Lenny, AMD64 architecture, > PostgreSQL backend). > > During backups, all my pools and volumes are used as needed. > > But during restore, what happens is that only volumes configured as > being part of the pool defined for the "RestoreFiles" job are > automatically mounted, and bacula waits for us to manually mount the > other volumes, which don't seem to be possible since they are file > volumes (i.e. always 'mounted') OK, I've fixed this particular problem by creating multiple Media Types. This was written in the documentation, but not about this particular subject, about multiple concurrent accesses instead. Now what about my other question (probably an easy one, I suppose I'll have the answer anyway in a month or two...) > Another small question : > > As you can see above, I want to have very small volumes, but LOTS of > them. When labelling them automatically, bacula only uses 4 digits, what > will happen when it will reach ? Could someone tell me what happens then ? Does it automatically label the next volume as, for example, "full-1", or does it start over at "full-0001" and overwrite an existing volume, or does it fail miserably because "full-0001" should not be overwritten (if it's still full, as it will probably be) ? Thanks. -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] question about multiple file pools
Hi again, On Tue, Nov 17, 2009 at 02:09:03AM +0100, Arno Lehmann wrote: > > 17.11.2009 00:49, Jerome Alet wrote: > > > > On Mon, Nov 16, 2009 at 02:35:12PM +1100, Jerome Alet wrote: > >> We have an old bacula setup with a single tape drive as part of an > >> autochanger, which holds three different pools for incremental, > >> differential and full backups. On restore, the correct tape is correctly > >> extracted from the database then mounted from the correct pool on our > >> single tape drive. > >> > >> I'd like to replicate such functionnality on hard disks and a newer > >> release of bacula (2.4.4-1 under Debian Lenny, AMD64 architecture, > >> PostgreSQL backend). > >> > >> During backups, all my pools and volumes are used as needed. > >> > >> But during restore, what happens is that only volumes configured as > >> being part of the pool defined for the "RestoreFiles" job are > >> automatically mounted, and bacula waits for us to manually mount the > >> other volumes, which don't seem to be possible since they are file > >> volumes (i.e. always 'mounted') > > > > OK, I've fixed this particular problem by creating multiple Media Types. > > Great... I was about to suggest that :-) > > > This was written in the documentation, but not about this particular > > subject, about multiple concurrent accesses instead. > > At least you found it - but where did you initially look? (Just so we > can, perhaps, improve the manual) I think I've searched this a lot in resources definitions, especially pools, but the answer is (somewhat hidden) there : http://www.bacula.org/en/rel-manual/Basic_Volume_Management.html#ConcurrentDiskJobs Maybe this could be clarified but I don't really know where nor how, since I'm a bacula newbie (our existing setup was put in place before I landed here) > >> As you can see above, I want to have very small volumes, but LOTS of > >> them. When labelling them automatically, bacula only uses 4 digits, what > >> will happen when it will reach ? > > > > Could someone tell me what happens then ? > > > > Does it automatically label the next volume as, for example, > > "full-1", > > That's what should happen, if I read the code correctly. Things will > break if you end up with more than 19 digits. > > > or does it start over at "full-0001" and overwrite an > > existing volume, or does it fail miserably because "full-0001" should > > not be overwritten (if it's still full, as it will probably be) ? > > But - why do you want so many volumes that you expect to have 10,000 > in a few months? Sounds a bit hard to manage... I knew the question would come ! :-) I'm currently doing some testing and I like to play with things to see if the solution is robust. When in production my volumes will certainely be bigger than 32 MB as they are now. Thanks a lot for your answers which clarify my thoughts. bye -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Strange problems with Bacula which mostly works
Hi, I'm using bacula 2.4.4 to save to file volumes, using gzip compression, with both Windows and GNU/Linux clients. The database backend is PostgreSQL 8.3.8 Backups and restore both seem to work just fine. However, in the "Media" panel in bat, the "Files" column always contain 0, even when I use volumes that can hold a lot of files. If I use "list files jobid=##" all files are correctly listed, and in the "Joblist" panel I can see the number of files saved for each job. Now if I try to purge files for a particular client, the answer is always "No Files found for client myclient.example.org to purge from MyCatalog catalog." Finally, the "Version browser" panel always come empty, and doesn't even contain any of the buttons nor drop down lists, nor the four sub-windows. What could be wrong ? I've named both my clients, jobs and filesets as each client's FQDN, could this be the problem ? Thanks for your time. -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Does someone have bacula-fd for SunOS 5.8 ?
Hi, I've got an old Sun box that I'd like to integrate into our Bacula setup. oursunbox:/# uname -a SunOS oursunbox 5.8 Generic_108528-13 sun4u sparc SUNW,UltraAX-i2 Since this machine doesn't even have any C compiler installed, I'd like to know if someone with a similar OS and a bacula client installed on it could send the necessary files to me to have bacula-fd run. Thanks in advance -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Does someone have bacula-fd for SunOS 5.8 ?
Hi, On Thu, Nov 19, 2009 at 05:15:39PM -0500, Ryan Novosielski wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Don't know what version you're looking for in particular. Sorry I forgot this. I'm using 2.4.4 on the director, so I suppose any older release will do. > At least > Bacula 2.2.8 is on Blastwave or OpenCSW. I should really figure out how > to and then release a set of current ones. I'm sure it's not a difficult > task. My Sun/Solaris skills are a bit rusty nowadays, I didn't know where to search for this. I'll look into these websites ASAP. Thanks a lot for your help. -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Director connection failure -- FD and SD on same machine
Good evening,
>
> From: Arno Lehmann
> Sent: Tue Nov 24 19:16:07 NCT 2009
> To:
> Subject: Re: [Bacula-users] Director connection failure -- FD and SD on same
> machine
> >
> > I am having an issue in that the director cannot communicate with a
> > particular file daemon. The file daemon in question is running on the
> > same machine as the storage daemon.
Could you double check the configuration of the TCP wrapper
(/etc/hosts.{allow,deny}) ?
let's say your /etc/hosts.deny file contains "ALL : ALL" then in your
/etc/hosts.allow file you need to put something like "YourFDName: 127.0.0.1" to
allow the director running on localhost to connect to the local FD.
hth
bye
Jerome Alet
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] [GENERAL] Catastrophic changes to PostgreSQL 8.4
On Thu, Dec 03, 2009 at 10:54:07AM +0800, Craig Ringer wrote: > > Anyway, it'd be nice if Bacula would convert file names to utf-8 at the > file daemon, using the encoding of the client, for storage in a utf-8 > database. +1 for me. this is the way to go. I understand people with an existing backup history won't be very happy with this unless you provide them the appropriate tools or instructions to convert their database's content, though. bye -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Dell PV LT2000 supported in Bacula ?
Hi, > > From: Ken Barclay > Sent: Wed Jan 13 16:56:54 NCT 2010 > To: 'Ralf Brinkmann' , 'Bekker' > > Subject: Re: [Bacula-users] Dell PV LT2000 supported in Bacula ? > > > Regarding Barcode Labels, this is too good not to share. > > Free Generator of Barcode Tape Labels for Ultrium LTO Backup Libraries. > (Nothing to install) > > http://tapelabels.librelogiciel.com/ Thanks Ken for your comment about my website ! I appreciate it very much. It seems we use the same software then ;-) bye Jerome Alet -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Dell PV LT2000 supported in Bacula ?
Hi, > > From: Jon Schewe > Sent: Wed Jan 13 23:32:14 NCT 2010 > To: Ken Barclay > Subject: Re: [Bacula-users] Dell PV LT2000 supported in Bacula ? > > > Any reason this is limited to LTO? I would think I could use it for my > DLT tapes too. The only reason at the time I wrote this is that I only had the complete specification for LTO labels, not DLT ones. The second reason right now is that I don't have much time to work on this anymore, so only small modifications will be implemented (like a list of color schemes for alpha characters, for example) bye Jerome Alet -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Dell PV LT2000 supported in Bacula ?
On Wed, Jan 13, 2010 at 01:53:25PM -0600, Jon Schewe wrote: > On 1/13/10 1:49 PM, jerome alet wrote: > > > > > Then my guess is that it'll work just fine. I've used generic barcode > creators in the past and they've been OK too, just wondering if there > was something special in there with sizing or something like that. I think there could be something "special" : the LTO labels specification is very precise wrt sizes, spacing, and error margins, so I suppose the DLT labels specification can be different. bye -- Jérôme Alet - - Centre de Ressources Informatiques Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 266754 Fax : +687 254829 -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] How to get free hard disk space back ?
Hi,
We've got a Bacula 5.0.2 setup here under Debian (standard packages).
We only use hard disk storage pools, and have created 3 of them on a
single partition : FileFull, FileDifferential, and FileIncremental.
The problem is, after adding more and more servers to our backup
strategy over the years, we went out of free hard disk space, so I've
updated the configuration and database to expire jobs, files and volumes
after 45 days instead of 70 before.
I've also changed the configuration of the catalog backup job to be
something like this :
RunBeforeJob = "/etc/bacula/scripts/make_catalog_backup bacula bacula xxx
localhost"
RunAfterJob = "/etc/bacula/scripts/delete_catalog_backup"
Write Bootstrap = "/backups/bootstraps/BackupCatalog-%n.bsr"
Priority = 11 # run after main backup
RunScript {
RunsWhen=After
RunsOnClient=No
Console = "purge volume action=all allpools storage=FileFull"
Console = "purge volume action=all allpools storage=FileDifferential"
Console = "purge volume action=all allpools storage=FileIncremental"
}
And all storage pools are defined with "Action On Purge = Truncate"
But of course the volume that are purged remain on disk, and so our hard
disk is still full.
The problem is that the three storage pools reside on the same
partition, so for example you do an incremental backup, it will
correctly truncate one of the "dead" volumes and so on until the
incremental backup is done. At that time, our hard disk is still full.
But now a full backup takes place, and unfortunately there's no more any
volume from the Full storage pool to expire.
Of course Bacula won't reuse a "dead" volume from a different storage
pool, of which there are plenty, so now I can successfully do
incremental backups, since dead volumes are truncated when needed in
this pool, but my full backups don't work because there's no more volume
to recycle in the Full pool.
So I'd like to know the way to physically delete all dead volumes from
the disk with rm, regaining almost 50% of our total hard disk space
which is occuped by volumes containing expired datas.
I suppose there's some SQL query to launch to retrieve the list of them
all, and then another query to delete all matching records from the
database, but there are several tables involved and I don't want to make
an error.
So please could you tell me how to proceed now to :
1 - Extract all the names of the "expired" volumes.
2 - Clean my database from any and all "expired" volumes from all pools.
3 - With the list obtained in 1 I think I'll be able to rm them ;-)
Thanks so much in advance.
--
Jérôme Alet - - Direction du Système d'Information
Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX
Tél : +687 290081 Fax : +687 254829
--
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] How to get free hard disk space back ?
On Thu, Apr 19, 2012 at 10:11:20AM -0400, Josh Fisher wrote: > > Changing the "action on purge" for the pool does not change it for the > old volumes that were created before the change was made. You have to > individually update the action on purge for each of those old volumes. Sorry I didn't mention it but I did this with : UPDATE media SET actiononpurge=1; But still no volume is truncated excepted at recycle time, when possible (in our case not possible anymore for our Full storage pool). Since all storage pools are on the same disk partition, despite the Differential and Incremental storage pools having hundreds (they are small) of "dead" volumes, Full backups still can't happen because there's not a single recyclable volume in the Full storage pool. Any other idea ? Thanks in advance. -- Jérôme Alet - - Direction du Système d'Information Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 290081 Fax : +687 254829 -- For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] How to get free hard disk space back ?
On Thu, Apr 19, 2012 at 05:14:35PM -0400, John Drescher wrote:
>
> How big of a limit have you set on volume size? I usually set this to
> 10GB or so. Or do you have use volume once or some other way to limit
> volume size / use duration?
My volume size is limited to 2 GB.
My SD is configured like this (similar config for other devices) :
Device {
Name = FileStorageFull
Media Type = FileFullMT
Archive Device = /backups/full/
LabelMedia = yes; # lets Bacula label unlabeled media
Random Access = Yes;
AutomaticMount = yes; # when device opened, read it
RemovableMedia = no;
AlwaysOpen = no;
Maximum Block Size = 64512;
Maximum Network Buffer Size = 64512;
}
My Storages (similar config for other storages) :
Storage {
Name = FileFull
Address = sss.univ-nc.nc
SDPort = 9103
Password = "sss"
Device = FileStorageFull
Media Type = FileFullMT
Maximum Concurrent Jobs = 30
Heartbeat Interval = 60
}
My Pools (similar config for other pools) :
Pool {
Name = FullPool
Pool Type = Backup
Recycle = yes
AutoPrune = yes
Action On Purge = Truncate
Volume Retention = 45 days
Maximum Volume Bytes = 2 GB
LabelFormat = "full-"
Storage = FileFull
Catalog Files = yes
}
And my clients (similar for other clients) :
Client {
Name = "yyy.univ-nc.nc"
Address = yyy.univ-nc.nc
FDPort = 9102
Catalog = UNCCatalog
Password = "yyy" # password for FileDaemon
File Retention = 45 days
Job Retention = 45 days
AutoPrune = yes # Prune expired Jobs/Files
Heartbeat Interval = 60
}
Anything wrong ?
bye
--
Jérôme Alet - - Direction du Système d'Information
Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX
Tél : +687 290081 Fax : +687 254829
--
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] How to get free hard disk space back ?
Hi, On Thu, Apr 19, 2012 at 05:30:41PM -0400, Phil Stracchino wrote: > > I gather you are creating a new volume for every job, and never reusing > the old ones. So write a simple shell script that will connect to the > bacula DB, select VolumeName from Media where VolStatus = 'Purged', save > the list of volume names, and delete them from the disk. No, old volumes get recycled correctly. But what I've got is this : bacula=# select distinct volstatus from media; volstatus --- Error Full Append (3 lignes) So there's no volume marked as purged, but old volumes are recylcled correctly anyway as can be seen below : 20-avril 02:11 sauvageon-sd JobId 47093: Recycled volume "incremental-0671" on device "FileStorageIncremental" (/backups/incremental/), all previous data lost. 20-avril 02:11 sauvageon-sd JobId 47093: New volume "incremental-0671" mounted on device "FileStorageIncremental" (/backups/incremental/) at 20-avril-2012 02:11. 20-avril 02:33 sauvageon-sd JobId 47093: User defined maximum volume capacity 2,000,000,000 exceeded on device "FileStorageIncremental" (/backups/incremental/). 20-avril 02:33 sauvageon-sd JobId 47093: End of medium on Volume "incremental-0671" Bytes=1,999,936,621 Blocks=31,001 at 20-avril-2012 02:33. 20-avril 02:33 sauvageon-dir JobId 47093: There are no more Jobs associated with Volume "incremental-6625". Marking it purged. 20-avril 02:33 sauvageon-dir JobId 47093: All records pruned from Volume "incremental-6625"; marking it "Purged" 20-avril 02:33 sauvageon-dir JobId 47093: Recycled volume "incremental-6625" 20-avril 02:33 sauvageon-sd JobId 47093: Recycled volume "incremental-6625" on device "FileStorageIncremental" (/backups/incremental/), all previous data lost. 20-avril 02:33 sauvageon-sd JobId 47093: New volume "incremental-6625" mounted on device "FileStorageIncremental" (/backups/incremental/) at 20-avril-2012 02:33. 20-avril 02:37 sauvageon-sd JobId 47093: Job write elapsed time = 02:07:04, Transfer rate = 1.483 M Bytes/second 20-avril 02:38 sauvageon-dir JobId 47093: Bacula sauvageon-dir 5.0.2 (28Apr10): 20-avril-2012 02:38:57 Build OS: x86_64-pc-linux-gnu debian squeeze/sid JobId: 47093 Job:neo.univ-nc.nc.2012-04-20_00.30.00_27 Backup Level: Incremental, since=2012-04-19 00:30:05 Client: "neo.univ-nc.nc" i386-pc-linux-gnu,debian,3.1 FileSet:"neo.univ-nc.nc" 2011-06-23 00:30:00 Pool: "IncrementalPool" (From Job IncPool override) Catalog:"UNCCatalog" (From Client resource) Storage:"FileIncremental" (From Pool resource) Scheduled time: 20-avril-2012 00:30:00 Start time: 20-avril-2012 00:30:07 End time: 20-avril-2012 02:38:57 Elapsed time: 2 hours 8 mins 50 secs Priority: 10 FD Files Written: 42,320 SD Files Written: 42,320 FD Bytes Written: 11,301,484,945 (11.30 GB) SD Bytes Written: 11,307,093,934 (11.30 GB) Rate: 1462.0 KB/s Software Compression: 29.3 % VSS:no Encryption: no Accurate: no Volume name(s): incremental-0652|incremental-0653|incremental-0654|incremental-0281|incremental-0655|incremental-0656|incremental-0657|incremental-0658|incremental-0659|incremental-0660|incremental-0661|incremental-0662|incremental-0663|incremental-0664|incremental-0666|incremental-0667|incremental-0668|incremental-0669|incremental-0670|incremental-0671|incremental-6625 Volume Session Id: 2607 Volume Session Time:1330974874 Last Volume Bytes: 1,376,792,193 (1.376 GB) Non-fatal FD errors:0 SD Errors: 0 FD termination status: OK SD termination status: OK Termination:Backup OK 20-avril 02:38 sauvageon-dir JobId 47093: Begin pruning Jobs older than 1 month 15 days . 20-avril 02:39 sauvageon-dir JobId 47093: Pruned 1 Job for client neo.univ-nc.nc from catalog. 20-avril 02:39 sauvageon-dir JobId 47093: Begin pruning Jobs. 20-avril 02:39 sauvageon-dir JobId 47093: No Files found to prune. 20-avril 02:39 sauvageon-dir JobId 47093: End auto prune. The problem is only a single volume is marked as purged at a time, then it is immediately recycled. So from what I'm reading from your answers I think all my "old" volumes should have been marked as "Purged" already, but this is not the case, and I don't know why. Of course when Bacula will set their status as "Purged" in the DB, I'll know how to clean them all :-) So what's wrong with my setup ? (BTW my database backend is PostgreSQL, could this be the problem ?) bye -- Jérôme Alet - - Direction du Système d'Information Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 290081 Fax : +687 254829 ---
[Bacula-users] For those of you who might be interested
Hi there, Just in case someone would be interested, here's the link to my Free online generator of LTO tape labels : http://tapelabels.librelogiciel.com/ Any comment is very welcome. I'd particularly like to know if a standalone offline release would be of interest, and if yes then why ? TIA. -- Jérôme Alet - - Direction du Système d'Information Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 290081 Fax : +687 254829 -- Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] barcode printing choices (Was: For those of you who might be interested)
Hi, On Wed, Apr 25, 2012 at 05:51:52PM -0400, mark.berg...@uphs.upenn.edu wrote: > > Unfortunately, our new tape library (Dell ML6010 -- rebranded Adic i500) > is much more sensitive to the barcodes than our 2 previous libraries. > The new library cannot read the old labels generated through your site > and printed on Avery 6577 labels. I'm now using a derivitave of the > "make_tape_label" script[1] to print new barcodes (on the same Avery > stock) to replace the existing labels, and am purchasing pre-labeled > tapes for all new media. > > I see this as an issue with the barcode reader on the ML6010--your > labels were fine with 3 previous barcode readers, and Dell tech support > acknowledged that other people have had similar issues with the ML6010 being > very sensitive as to the label format, size, and placement on the tape > cartridge. As far as I know my generator fully respects the specification made available by IBM, but doesn't add any way to add some variability to account for buggy tape libraries. In fact I didn't know this sort of problem could happen, stupidly thinking that libraries than can't read labels made to the specification wouldn't be allowed to be approved for selling. I'll see how this could be done in a future release. Thanks for your very interesting feedback. -- Jérôme Alet - - Direction du Système d'Information Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 290081 Fax : +687 254829 -- Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ ___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] How to get free hard disk space back ?
Hi, On Sat, Apr 28, 2012 at 09:16:00AM +0100, Dermot Beirne wrote: > Glad to hear it was useful. > Would you mind sharing your other solution as it might be useful to me in > future. Attached to this message you'll find the shell script I'm now using. Beware to adapt it to your needs : I use PostgreSQL and all my storage pools' names contain the word "File". Hoping this helps. Thanks to all who helped me with this problem. -- Jérôme Alet - - Direction du Système d'Information Université de la Nouvelle-Calédonie - BPR4 - 98851 NOUMEA CEDEX Tél : +687 290081 Fax : +687 254829 truncatevolumes.sh Description: Bourne shell script -- Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/___ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
