I'm setting up a backup for some client, mostly windows devices, mostly
medical system (EEG, EMG, ...).

They are 'clients', eg they get powered on on need, they not are 'always on'
or powered on on every days.


My target was:
 - start backup at some hour, and be a bit 'aggressive' trying to retry
   backup (wait the power on of the system) until done
 - use VirtualFull, to do only incrementals and so put clients on minimal
   load
 - don't supose nothing on clients, eg 'client get powered on at least 3
   days a week'...


Saying that, i've done a pool like that:

 Pool {
        Name = FVG-SV-VWSSV1FilePool
        Pool Type = Backup
        NextPool = FVG-SV-VWSSV1FilePool
        Storage = SVPVE3FileMulti
        Maximum Volume Jobs = 6
        Recycle = yes
        AutoPrune = yes
        Action On Purge = Truncate
 }

Note that there's no 'Volume Retention = ', so no volume retention. ;-)
In every pool there are 3 volumes, simple file volumes.

Now job is:

 Job {
        Name = FVG-SV-VWSSV1
        JobDefs = DefaultJob
        Storage = SVPVE3FileMulti
        Pool = FVG-SV-VWSSV1FilePool
        Messages = StandardClient
        Schedule = WeeklyClient
        Reschedule On Error = yes
        Reschedule Interval = 30 minutes
        Reschedule Times = 8
        Max Run Sched Time = 8 hours
        Accurate = yes
        Backups To Keep = 10
        Delete Consolidated Jobs = yes
        Client = fvg-sv-vwssv1-fd
        FileSet = TS1000Std
        Write Bootstrap = "/var/lib/bacula/FVG-SV-VWSSV1.bsr"
 }

All scheduled with:

 Schedule {
        Name = WeeklyClient
        Run = Incremental mon-fri at 8:00
        Run = VirtualFull sun at 8:00
 }


In this way, all work as expected. Incremental runs, and weekly a
VirtualFull job 'consolidate' jobs, keeping 10 backups and purging other
jobs (so applying retention 'by number of jobs' and not 'by age of jobs').
Wonderful!


But there are two drawbacks:

1) because i need incremental backups to be 'aggressive', the VirtualFull
 one are aggressive too; so, if fail because there's no sufficient jobs to
consolidate, they are retried 8 time before to give it up. This clearly is
totally unuseful.

2) VirtualFull consolidate backups and remove consolidated jobs, but remove
 ONLY correctly terminated jobs:

 *list jobs client=fvg-sv-vwssv1-fd
 Using Catalog "BaculaLNF"
 
+--------+--------------------------+---------------------+------+-------+----------+-------------+-----------+
 | jobid  | name                     | starttime           | type | level | 
jobfiles | jobbytes    | jobstatus |
 
+--------+--------------------------+---------------------+------+-------+----------+-------------+-----------+
 | 16,614 | FVG-SV-VWSSV1            | 2024-02-01 08:40:46 | B    | I     |     
   0 |           0 | A         |
 | 16,660 | FVG-SV-VWSSV1            | 2024-02-02 16:01:11 | B    | I     |     
   0 |           0 | A         |
 | 16,767 | FVG-SV-VWSSV1            | 2024-02-05 08:00:02 | B    | F     |     
   0 |           0 | A         |
 | 16,807 | FVG-SV-VWSSV1            | 2024-02-06 08:35:16 | B    | I     |     
   0 |           0 | A         |
 | 16,847 | FVG-SV-VWSSV1            | 2024-02-07 14:06:41 | B    | I     |     
   0 |           0 | A         |
 | 16,887 | FVG-SV-VWSSV1            | 2024-02-08 16:01:15 | B    | I     |     
   0 |           0 | A         |
 | 16,927 | FVG-SV-VWSSV1            | 2024-02-09 15:22:07 | B    | I     |     
   0 |           0 | A         |
 | 17,033 | FVG-SV-VWSSV1            | 2024-02-12 16:01:13 | B    | F     |     
   0 |           0 | A         |
 | 17,073 | FVG-SV-VWSSV1            | 2024-02-13 13:46:29 | B    | I     |     
   0 |           0 | A         |
 | 17,430 | FVG-SV-VWSSV1            | 2024-02-22 11:14:40 | B    | F     |     
   0 |           0 | A         |
 | 17,435 | FVG-SV-VWSSV1            | 2024-02-22 11:17:53 | B    | F     |     
   0 |           0 | A         |
 | 17,436 | FVG-SV-VWSSV1            | 2024-02-22 11:31:27 | B    | F     |     
   0 |           0 | A         |
 | 17,434 | FVG-SV-VWSSV1            | 2024-02-22 11:32:21 | B    | F     |     
   0 |           0 | A         |
 | 17,449 | FVG-SV-VWSSV1            | 2024-02-22 11:33:38 | B    | F     |     
   9 | 154,380,338 | T         |
 | 17,439 | FVG-SV-VWSSV1            | 2024-02-22 11:34:31 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,440 | FVG-SV-VWSSV1            | 2024-02-22 11:39:22 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,441 | FVG-SV-VWSSV1            | 2024-02-22 11:40:30 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,442 | FVG-SV-VWSSV1            | 2024-02-22 11:48:43 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,443 | FVG-SV-VWSSV1            | 2024-02-22 11:58:40 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,444 | FVG-SV-VWSSV1            | 2024-02-22 12:01:24 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,445 | FVG-SV-VWSSV1            | 2024-02-22 12:07:45 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,446 | FVG-SV-VWSSV1            | 2024-02-22 12:15:21 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,447 | FVG-SV-VWSSV1            | 2024-02-22 12:17:43 | B    | I     |     
   2 | 127,566,004 | T         |
 | 17,448 | FVG-SV-VWSSV1            | 2024-02-22 12:19:58 | B    | I     |     
   2 | 127,566,004 | T         |
 
+--------+--------------------------+---------------------+------+-------+----------+-------------+-----------+

error, cancelled jobs... remain always there.


Both seems more nuisance that troubles, but... i hope there's a solution.


Thanks.

-- 
  Mi piaccion le fiabe
  raccontane altre                                      (F. Guccini)




_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to