Re: [Bacula-users] Verify and admin job retention
Op 29/12/2010 21:00, Martin Simmons schreef:
On Wed, 29 Dec 2010 13:32:28 +0100, Jeremy Maes said:
When checking some things at our clients' sites where we are testing
bacula I found out that
our retention periods (and auto-pruning) don't seem to work for verify
jobs and admin jobs.
We are using general 30d File and Job retention periods everywhere, and
6d, 20d or 81d volume retention periods for our volumes in the pool
configs, depending on the type (daily, weekly, monthly)
Yes, I had problems with that too so I run the sql command
delete from job
where type in ('V', 'D')
and starttime now()-interval '2 months'
every week to remove them.
I think one problem is that Volume Retention doesn't remove them because they
are not associated with any particular volumes, so they hang around for much
longer than expected.
__Martin
Ah, shows me again that I should start thinking more in terms of SQL
statements when it comes to that kind of thing.
Thanks for the code, will definitely try it out and implement it as a
weekly job.
Will also look into the existing bugs and report it if necessary as I
don't think this is desireable behaviour.
Kind regards,
Jeremy
DISCLAIMER
http://www.schaubroeck.be/maildisclaimer.htm
--
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
[Bacula-users] Verify and admin job retention
Hey
When checking some things at our clients' sites where we are testing
bacula I found out that
our retention periods (and auto-pruning) don't seem to work for verify
jobs and admin jobs.
We are using general 30d File and Job retention periods everywhere, and
6d, 20d or 81d volume retention periods for our volumes in the pool
configs, depending on the type (daily, weekly, monthly)
All actual backup jobs (normal backup and catalog backup) are being
pruned correctly according to the Retention periods we set.
Daily backups disappear after 6 days due to the volume retention, weekly
or monthly backups get stored longer because their retention
is set differently in their respective pool configs. None of this seems
to matter for our verify and admin jobs however. All those jobs seem
to still be in the catalog from the day we started testing about 4-5
months ago.
Ex: |15 | Verifyyoda | 2010-08-25 21:50:58 | V | O | 553 | 0 | T |
|19 | Verifyyoda | 2010-08-26 21:56:32 | V | O | 2,273 | 0
| T |
|39 | Verifyyoda | 2010-09-01 01:54:06 | V | O | 285,617 | 0
| T |
Manually running the prune command (and confirming the job retention of
30 days) seems to ignore all verify and admin jobs aswell and will
quit with No Jobs found to prune.
I know I will probably have to tweak the retention times a little
further in the future, but I was wondering which of them should
influence the pruning of verify/admin jobs?
Are they dependent on the job retention? (seems logical to me) Or are
they in some way bound to the volume retention? (would seem weird, and
even then I shouldn't
be seeing jobs over 81d old)
Info: All backups being run with bacula 5.0.1 DIR, SD and FD on RHEL4
update 8 systems. Most are being written to tape, the exception being
one test where we are
using disk backups.
Relevant example configs:
Job {
Name = Verifyyoda
Type = Verify
VerifyJob = Backupyoda
Level = VolumeToCatalog
Pool = Default
FileSet = Full Linux Set
Client = yoda-fd
Schedule = WeeklyCycleVerify
Write Bootstrap = /var/lib/bacula/Vyoda.bsr
Storage = FileStorage
Messages = Standard
Priority = 11
RunScript {
RunsWhen = Before
FailJobOnError = Yes
Command = rm /tmp/%c
}
}
Client {
Name = yoda-fd
Address = yoda
FDPort = 9102
Catalog = MyCatalog
Password = # password for FileDaemon
File Retention = 30 days# 30 days
Job Retention = 30 days
AutoPrune = yes # Prune expired Jobs/Files
}
Pool {
Name = MondayPool
Pool Type = Backup
Recycle = yes
AutoPrune = yes # Prune expired volumes
Volume Retention = 6 days
Label Format = Maandag-
VolumeUseDuration = 1 days
}
Any ideas or insights would be greatly appreciated.
Kind regards,
Jeremy Maes
mailto:j...@schaubroeck.be
DISCLAIMER
http://www.schaubroeck.be/maildisclaimer.htm
--
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
Re: [Bacula-users] Verify and admin job retention
On Wed, 29 Dec 2010 13:32:28 +0100, Jeremy Maes said:
When checking some things at our clients' sites where we are testing
bacula I found out that
our retention periods (and auto-pruning) don't seem to work for verify
jobs and admin jobs.
We are using general 30d File and Job retention periods everywhere, and
6d, 20d or 81d volume retention periods for our volumes in the pool
configs, depending on the type (daily, weekly, monthly)
Yes, I had problems with that too so I run the sql command
delete from job
where type in ('V', 'D')
and starttime now()-interval '2 months'
every week to remove them.
I think one problem is that Volume Retention doesn't remove them because they
are not associated with any particular volumes, so they hang around for much
longer than expected.
__Martin
--
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
