On 12/23/2010 7:35 AM, Dermot Beirne wrote:
> Hello,
> I recently switched from copy jobs to migrate jobs functionality, and
> I am concerned about retention.
>
> I am running Bacula 5.0.3 on Ubuntu 9.04
>
> I have 4 disk pools (daily, weekly, monthly, yearly) and a matching 4
> tape pools.
>
> The disk pool retentions are set to:
>
> Daily-Disk-Pool: File, Job, Volume all 2 days
> Weekly-Disk-Pool: File, Job, Volume all 2 days
> Monthly-Disk-Pool: File and Job 3 days. Volume 4 days
> Yearly-Disk-Pool: File and Job 3 days. Volume 4 days
>
> The tape pool retentions are set to:
>
> Daily-Tape-Pool: File and Job are 12 days. Volume is 13 days
> Weekly-Tape-Pool: File and Job are 32 days. Volume is 33 days
> Monthly-Tape-Pool: File, Job and Volume all 1 year
> Yearly-Tape-Pool: File and Job 1 year. Volume 10 years
>
> All client definitions have a job retention setting of 1 year. My
> understanding is that this needs to be specified, but is overridden by
> the pool settings. No other retentions periods are set in
> bacula-dir.conf
>
> If i run "list volume pool = Weekly-Tape-Pool, I get the following:
>
> +---------+------------+-----------+---------+-----------------+----------+--------------+---------+------+-----------+-----------+---------------------+
> | MediaId | VolumeName | VolStatus | Enabled | VolBytes |
> VolFiles | VolRetention | Recycle | Slot | InChanger | MediaType |
> LastWritten |
> +---------+------------+-----------+---------+-----------------+----------+--------------+---------+------+-----------+-----------+---------------------+
> | 295 | 000023L4 | Append | 1 | 10,332,628,992 |
> 11 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-10-26 11:47:11 |
> | 306 | 000025L4 | Append | 1 | 373,078,056,960 |
> 374 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-10-12 15:00:42 |
> | 308 | 000015L4 | Append | 1 | 624,046,058,496 |
> 628 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-11-29 14:27:15 |
> | 314 | 000018L4 | Append | 1 | 784,165,358,592 |
> 790 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-10-18 21:44:12 |
> | 791 | 000038L4 | Used | 1 | 755,243,596,800 |
> 759 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-10-25 19:32:43 |
> | 806 | 000053L4 | Append | 1 | 549,454,574,592 |
> 553 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-12-20 12:50:22 |
> | 807 | 000054L4 | Append | 1 | 856,033,339,392 |
> 860 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-11-01 17:31:47 |
> | 895 | 000067L4 | Used | 1 | 643,130,385,408 |
> 648 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-09-20 14:08:35 |
> | 896 | 000068L4 | Append | 1 | 528,352,441,344 |
> 533 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-12-13 12:22:48 |
> | 905 | 000076L4 | Used | 1 | 502,886,651,904 |
> 505 | 2,851,200 | 1 | 0 | 0 | LTO-4 |
> 2010-09-27 12:12:58 |
> +---------+------------+-----------+---------+-----------------+----------+--------------+---------+------+-----------+-----------+---------------------+
>
> I hope this is legible, but MediaID 896 (as an example) was last
> written to on the 13/12/10, just 10 days ago. This should have a
> retention period for file/jobs/volume in the region of 32 days as
> indicated previously.
According to the above, Media ID 896 has a retention period of 2,851,200
seconds (or 33 days).
>
> Yet if I run query in bconsole I get the following for this mediaid:
>
> 13: List Jobs stored on a selected MediaId
> 14: List Jobs stored for a given Volume name
> 15: List Volumes Bacula thinks are in changer
> 16: List Volumes likely to need replacement from age or errors
> Choose a query (1-16): 13
> Enter MediaId: 896
> No results to list.
>
>
> I get the same for multiple other volumes in other pools which should
> be listing files and jobs based on the pools retention.
>
> When I do an llist, the output for that media id (if this is any use)
> is as follows:
>
> MediaId: 896
> VolumeName: 000068L4
> Slot: 0
> PoolId: 8
> MediaType: LTO-4
> FirstWritten: 2010-12-13 05:02:53
> LastWritten: 2010-12-13 12:22:48
> LabelDate: 2010-12-13 05:02:53
> VolJobs: 7
> VolFiles: 533
> VolBlocks: 8,189,986
> VolMounts: 2
> VolBytes: 528,352,441,344
> VolErrors: 0
> VolWrites: 10,757,372
> VolCapacityBytes: 0
> VolStatus: Append
> Enabled: 1
> Recycle: 1
> VolRetention: 2,851,200
> VolUseDuration: 82,800
> MaxVolJobs: 0
> MaxVolFiles: 0
> MaxVolBytes: 0
> InChanger: 0
> EndFile: 532
> EndBlock: 0
> VolParts: 0
> LabelType: 0
> StorageId: 4
> DeviceId: 0
> LocationId: 0
> RecycleCount: 1
> InitialWrite: 0000-00-00 00:00:00
> ScratchPoolId: 0
> RecyclePoolId: 0
> Comment: NULL
>
>
> My understanding with migration is that the retention periods of the
> migrated job/file/volume become those of the pool it is being migrated
> to, i.e. the relevant tape pool.
Aren't Job and File retention specific to a client, not a Volume?
What I do, or have tired, and have not tested is putting Job and File
retention to a value equal to the maximum Pool retention value.
Thus, in your case, your maximum Volume retention is 10 years. I
suggest you want Job and File retention set at 10 years.
Would that help you?
> This does not appear to be working or something else is very wrong.
>
> I have run the update volumes command to ensure all volumes have
> inherited their retention from their pool settings.
>
> Can anyone explain what is wrong here, as it looks like I'll have to
> use bscan to restore anything from these volumes?
>
> Thank you, and Happy Christmas.
--
Dan Langille - http://langille.org/
------------------------------------------------------------------------------
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 AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|