Hi Giuseppe,
I would like to append some comments here about your "volume use duration"
configuration. As the directive name says, this is used to leave the volume
available for Bacula "during" your backup jobs (at least the ones you want
the data goes to that volume). Maybe you are having volume retention time
troubleshooting here. If Bacula is using 000529L4 instead of 000506L4, this
is because both volumes have a status (purged, append, recycle) that
permits bacula to choose which one is the best one (guided by its volume
recycling algorithm). So, it seems the 000529L4 has a last written time
older than the 000506L4.
I recommend you to put your "volume use duration" in hours, like Josh
Fisher said, and just the time that you need the volume be available
(append) for receiving data. And take a look at your volume retention
period. If you have 10 volumes and want just one of them be available at a
time, set a volume use duration of 9 days (you can set in hours if you are
sure about the last written time of your volumes).
Best regards,
Ana
On Fri, Oct 17, 2014 at 10:11 AM, Josh Fisher <jfis...@pvct.com> wrote:
> On 10/17/2014 7:05 AM, Kern Sibbald wrote:
>
> Hello,
>
> I have not seen your original post so I respond only to comments you
> mention below without understanding the full issue. Yes, there can be
> typically a few seconds delay between when a volume is chosen for use and
> when it expires. The use duration is only checked when the volume is first
> selected and not at any other time during running the job. This can mean
> that a volume will be used for more time than specified in the use
> duration. It would not be easy to fix this, and any simple fix might cause
> performance problems so you must be aware of this restriction.
>
>
> Generally, schedules are starting jobs at the same time of day. It follows
> that the time period between successive runs will tend to be very nearly an
> integer number of days. If the use duration is also an integer number of
> days, then it can be expected that a volume's use duration will expire at
> very nearly the exact time that a job using a volume from that pool is
> scheduled to run. This is alleviated by configuring a use duration that is
> not an integer multiple of days. For example, instead of using a use
> duration of "7 days", set use duration to "156 hours" (6 days 12 hours).
> The hysteresis makes it far less likely that a job will be launched at
> exactly the same time that a volume's use duration is expiring.
>
>
>
> By the way, there were quite a few race conditions concerning use
> durations and other time periods that affect whether or not a volume can
> continue to be used. Many of these race conditions were fixed in 7.0.5.
> If you are having problems with volumes, I strongly recommend upgrading.
>
> Best regards,
> Kern
>
> On 14-10-16 07:28 PM, Giuseppe Vitillaro wrote:
>
> Apologies for appending to my own post, but I found
> another occurence of the same pattern, I didn't notice,
> in another job of some months ago, again attached to
> this message. The same volumes are involved, and these,
> for what I can see from my logs, are the only occurences
> of the pattern.
>
> It looks like that a volume exceeding the "use duration"
> once the job has choosed the backup volume, create this
> problem.
>
> May I guess the routine "find_next_volume_for_append()"
> in the source next_vol.c of the director could have
> a bug, at least in the version 5.2.13 (19Jan13) I'm using?
>
> In all the other cases I've verified, almost one year
> of correctly running jobs, the "use duration"
> have not been checked just after the beginning of the
> job, once the director has already decided the volume
> to use, but at a different time.
>
> I may be on the wrong track, but if I'm right bacula
> may overwrite a non empty, not purged volume.
>
> Something, I guess, all the bacula users would hate,
> isn't?
>
> G. Vitillaro.
>
>
> I can't understand why the bacula director, 5.2.13 under a stable updated
> gentoo, recycled one volume "000506L4", the correct one in the slot 9 of my
> changer, but instead the storage daemon recycled the current volume mounted
> in drive1, from slot4, volume "000529L4" and without actually doing any
> pruning on the volume, pruning that was correctly performed for the
> "000506L4" volume.
>
> Note that the volume "000506L4" exceeded "Max configure use duration" just
> before the records for volume "000506L4" had been pruned.
>
> After this I've just a sequence, complete log attached to this message, of
>
> Recycled volume "000506L4" Recycled volume "000529L4" on device "lto1"
> (/dev/nlto1), all previous data lost.
>
> This configuration of bacula had been running for quite a while without a
> glitch, almost 10 months now and I had not any hardware error on my library
> at the time of this job, at least for what I can see.
>
> I can't understand what is going on.
>
> Any clue?
>
> Regards, G. Vitillaro.
>
>
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for $9/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.http://p.sf.net/sfu/Zoho
>
>
>
> _______________________________________________
> Bacula-users mailing
> listBacula-users@lists.sourceforge.nethttps://lists.sourceforge.net/lists/listinfo/bacula-users
>
>
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for $9/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.http://p.sf.net/sfu/Zoho
>
>
>
> _______________________________________________
> Bacula-users mailing
> listBacula-users@lists.sourceforge.nethttps://lists.sourceforge.net/lists/listinfo/bacula-users
>
>
>
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for $9/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://p.sf.net/sfu/Zoho
> _______________________________________________
> Bacula-users mailing list
> Bacula-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bacula-users
>
>
------------------------------------------------------------------------------
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for $9/Month.
Get alerted through email, SMS, voice calls or mobile push notifications.
Take corrective actions from your mobile device.
http://p.sf.net/sfu/Zoho
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
