-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Well, I've figured out what's happening here, but not why or how to
correct it elegantly.
My problem is I have made a change (as seen below) to keep the tape
drive open. I don't have operators -- at least temporarily -- that can
be bothered with/given access to the console. My solution for this is
one unmount on Monday morning. This should allow the operator to remove
the tape sometime on Monday and insert the new tape before Monday night.
However, apparently "AutomaticMount" does not apply to situations where
a "user" (or crontab) has unmounted the tape drive? As a result, the
night's backups are BLOCKED as a result of user unmount.
What is the solution here? A separate script to mount the drive in the
evening before the backups? Isn't there any way to unmount a tape on
demand but not in such a fashion that an AutomaticMount cannot overcome it?
The next problem -- and this one I think is easily solvable with some
config changes -- is when things get into this state, there is no
continuing with the backups. I run a mount and the process hangs. I
think this is because there is no spare channel to the SD. At around the
same time, this happened, which IMO further supports that idea:
04-Dec 23:11 helios-dir: Max configured use duration exceeded. Marking
Volume "combined_BW2" as Used.
04-Dec 23:12 helios-dir: Pruned 6 Jobs on Volume "combined_BW3" from
catalog.
04-Dec 23:12 helios-dir: Error: bnet.c:426 Write error sending 77 bytes
to client:130.219.34.102:9101: ERR=Broken pipe
....my understanding is the director wanted to tell the SD about the
changes to the jobs, but was unable since the SD was busy. Is there some
way that I can allow a second communication channel to the SD, if this
is indeed the problem, without allowing for other weird things to happen
(like intermingling backups or anything like that -- essentially I
don't want any other things to be able to happen at the same time aside
from control commands).
Ryan Novosielski wrote:
> I'm a little stumped about this one. The only thing I've changed
> recently (and we're talking a couple of weeks ago) is the way I deal
> with incremental backup media. I had previously been closing the tape
> device after each backup, but really, when a tape is used all week, this
> is unnecessary and bad for the drive. So I am currently running like this:
>
> Device {
> Name = helios_DDS #
> Media Type = DDS-4
> Archive Device = /dev/rmt/0cbn
> AutomaticMount = yes; # when device opened, read it
> AlwaysOpen = yes;
> Volume Poll Interval = 30 minutes;
> Close on Poll = yes;
> RemovableMedia = yes;
> RandomAccess = no;
> Spool Directory = /usr/local/bacula/var/spool;
> }
>
> I'm not sure that any of these are a problem (though I suppose it's
> possible that I should not have Close on Poll in here or Volume Poll
> Interval). I believe the change I made was "AlwaysOpen", which had been
> "No." As far as tape changes, I do the following:
>
> 1 8 * * 1 echo "umount storage=helios_DDS" |
> /usr/local/bacula/sbin/bconsole -c /usr/local/bacula/etc/bconsole.conf
>
> .....to make sure that the tape is removable when my staff need to remove
> it, sometime before Monday night.
>
> However, this is where I've ended up the last two Monday nights:
>
> Running Jobs:
> Console connected at 28-Nov-06 01:13
> JobId Level Name Status
> ======================================================================
> 1717 Increme NJMSCatalyst.2006-11-27_23.00.00 is waiting on Storage
> helios_DDS
> 1718 Increme CFMX-dev.2006-11-27_23.00.01 is waiting on Storage
> helios_DDS
> 1719 Increme sopris-WEBMAIL.2006-11-27_23.00.02 is waiting on Storage
> helios_DDS
> 1720 Increme BigBrother-Display.2006-11-27_23.00.08 is waiting on
> Storage helios_DDS
> 1721 Increme SVNrepos.2006-11-27_23.00.09 is waiting on Storage
> helios_DDS
> 1722 Increme VistaBKsave.2006-11-27_23.00.10 is waiting execution
> 1723 Increme BigBrother-NET.2006-11-27_23.00.11 is waiting execution
> 1724 Full Catalog.2006-11-27_23.55.00 is waiting execution
> ====
>
> .....I don't know what "waiting" means in this context, or what I'm
> supposed to do to free up this situation (short of stopping Bacula or
> cancelling the jobs). Is there any way to bust up the log-jam here, or
> am I hosed? How can I avoid this mess in the future.
- -------------------------------------------------------------------------
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
- --
---- _ _ _ _ ___ _ _ _
|Y#| | | |\/| | \ |\ | | |Ryan Novosielski - Systems Programmer III
|$&| |__| | | |__/ | \| _| |[EMAIL PROTECTED] - 973/972.0922 (2-0922)
\__/ Univ. of Med. and Dent.|IST/AST - NJMS Medical Science Bldg - C630
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFFdZrhmb+gadEcsb4RAtpFAKCc13u6WGVthkEyNzLvlqff9QvELQCfY+ky
/zJBxrKpxNjotcF2oU/5kc8=
=n9JI
-----END PGP SIGNATURE-----
-------------------------------------------------------------------------
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
