Jobs don't remain on schedule

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Jobs don't remain on schedule

Craig Shiroma
Hello All,

I have a few jobs that are on the below schedule:

Schedule {
  Name = "WeeklyFullOnFriRunFirst"
  Run = Level=Full fri at 17:58
  Run = Level=Incremental sat-thu at 18:00
}

The jobs followed the schedule until last night which was a Thursday.  An incremental was suppose to run, but a Full was run instead.  Would anyone know what could cause this to happen?  Previous full backup did run on a Friday at 17:58.

The log for these jobs show similar text shown below:
bacula-dir JobId 413490:
bacula-dir JobId 413490: No prior or suitable Full backup found in catalog. Doing FULL backup.
bacula-dir JobId 413490: Start Backup JobId 413490, Job=servername.2017-03-30_18.00.06_05
bacula-dir JobId 413490: Max configured use duration=82,800 sec. exceeded. Marking Volume "servername-258519-vol" as Used.
bacula-dir JobId 413490: Pruning oldest volume "servername-263826-vol"
bacula-dir JobId 413490: There are no more Jobs associated with Volume "servername-263826-vol". Marking it purged.
bacula-dir JobId 413490: All records pruned from Volume "servername-263826-vol"; marking it "Purged"
bacula-dir JobId 413490: Using Device "servername" to write.
bacula-sd JobId 413490: Recycled volume "servername-263826-vol" on file device "servername" (/backup01/servername/), all previous data lost.

Seems like something is out of sync with the schedule that is causing the Full to run on the wrong day.

Best regards,
Craig

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Jobs don't remain on schedule

Bill Arlofski-4
On 03/31/2017 12:49 PM, Craig Shiroma wrote:

> Hello All,
>
> I have a few jobs that are on the below schedule:
>
> Schedule {
>   Name = "WeeklyFullOnFriRunFirst"
>   Run = Level=Full fri at 17:58
>   Run = Level=Incremental sat-thu at 18:00
> }
>
> The jobs followed the schedule until last night which was a Thursday.  An
> incremental was suppose to run, but a Full was run instead.  Would anyone know
> what could cause this to happen?  Previous full backup did run on a Friday at
> 17:58.
>
> The log for these jobs show similar text shown below:
> bacula-dir JobId 413490:
> bacula-dir JobId 413490: No prior or suitable Full backup found in catalog.
> Doing FULL backup.
> bacula-dir JobId 413490: Start Backup JobId 413490,
> Job=servername.2017-03-30_18.00.06_05
> bacula-dir JobId 413490: Max configured use duration=82,800 sec. exceeded.
> Marking Volume "servername-258519-vol" as Used.
> bacula-dir JobId 413490: Pruning oldest volume "servername-263826-vol"
> bacula-dir JobId 413490: There are no more Jobs associated with Volume
> "servername-263826-vol". Marking it purged.
> bacula-dir JobId 413490: All records pruned from Volume
> "servername-263826-vol"; marking it "Purged"
> bacula-dir JobId 413490: Using Device "servername" to write.
> bacula-sd JobId 413490: Recycled volume "servername-263826-vol" on file device
> "servername" (/backup01/servername/), all previous data lost.
>
> Seems like something is out of sync with the schedule that is causing the Full
> to run on the wrong day.
>
> Best regards,
> Craig

Hi Craig,

I see a couple possibilities here.

The first one might be that there is an issue with your retention period(s)
for the Full backup.  Probably not the case though.

The second, and I think more probable answer is that the fileset has been
edited recently, and when that occurs, Bacula automatically upgrades the job
to Full.

Take a look at the job's summary.  If you have a semi-recent version there
will be a date stamp in the summary next to the fileset.  This date is the
time the Fileset was last edited.

There is a Fileset setting called "IgnoreFilesetChanges = yes"  which can help
mitigate this second case, but there are some warnings to think about when
using this option.

Also, I have a feature request that has been accepted in the Enterprise
version that the reason why a job was upgraded to a Full is listed in the logs
too.   I do not know when this will be ported to the community version though.


Best regards,

Bill


--
Bill Arlofski
http://www.revpol.com/bacula
-- Not responsible for anything below this line --

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Jobs don't remain on schedule

Craig Shiroma
Hi Bill,

Hmmm, I know I didn't touch the fileset(s) and I have IgnoreFilesetChanges set to "yes" on these filesets.  The FileSet date on one of the jobs is "2015-10-10 18:00:12".

Here are some configs for one of the problem jobs:
Pool {
    Name = server_shortname
    Storage = server_shortname
    Pool Type = Backup
    Recycle = yes
    Recycle Oldest Volume = yes
    AutoPrune = no
    Volume Retention = 15 days
    Maximum Volumes = 16
    Volume Use Duration = 23h
    Label Format = "${Client}-${JobId}-vol"
}

Client {
    Name = server_shortname
    Address = server_fqdn
    FD Port = 9102
    Password = "some-password"
    Catalog = MyCatalog
    File Retention = 15 days
    Job Retention =  15 days
    AutoPrune = no
}

Job {
    Name = server_shortname
    Client = server_shortname
    Enabled = yes
    Type = Backup
    Schedule = WeeklyFullOnFriRunFirst
    FileSet = Windows_E_NoCompression_FileSet
    Max Full Interval = 6 days
    Pool = server_shortname
    Messages = Standard
    Write Bootstrap = "/var/spool/bacula/server_shortname.bsr"
    Priority = 10
    Reschedule On Error = yes
    Reschedule Interval = 1 hour
    Reschedule Times = 3
    Cancel Lower Level Duplicates = yes
    Allow Duplicate Jobs = no
    Spool Attributes = yes
    Accurate = yes
}

Thanks for the help!
Craig

On Fri, Mar 31, 2017 at 9:00 AM, Bill Arlofski <[hidden email]> wrote:
On 03/31/2017 12:49 PM, Craig Shiroma wrote:
> Hello All,
>
> I have a few jobs that are on the below schedule:
>
> Schedule {
>   Name = "WeeklyFullOnFriRunFirst"
>   Run = Level=Full fri at 17:58
>   Run = Level=Incremental sat-thu at 18:00
> }
>
> The jobs followed the schedule until last night which was a Thursday.  An
> incremental was suppose to run, but a Full was run instead.  Would anyone know
> what could cause this to happen?  Previous full backup did run on a Friday at
> 17:58.
>
> The log for these jobs show similar text shown below:
> bacula-dir JobId 413490:
> bacula-dir JobId 413490: No prior or suitable Full backup found in catalog.
> Doing FULL backup.
> bacula-dir JobId 413490: Start Backup JobId 413490,
> Job=servername.2017-03-30_18.00.06_05
> bacula-dir JobId 413490: Max configured use duration=82,800 sec. exceeded.
> Marking Volume "servername-258519-vol" as Used.
> bacula-dir JobId 413490: Pruning oldest volume "servername-263826-vol"
> bacula-dir JobId 413490: There are no more Jobs associated with Volume
> "servername-263826-vol". Marking it purged.
> bacula-dir JobId 413490: All records pruned from Volume
> "servername-263826-vol"; marking it "Purged"
> bacula-dir JobId 413490: Using Device "servername" to write.
> bacula-sd JobId 413490: Recycled volume "servername-263826-vol" on file device
> "servername" (/backup01/servername/), all previous data lost.
>
> Seems like something is out of sync with the schedule that is causing the Full
> to run on the wrong day.
>
> Best regards,
> Craig

Hi Craig,

I see a couple possibilities here.

The first one might be that there is an issue with your retention period(s)
for the Full backup.  Probably not the case though.

The second, and I think more probable answer is that the fileset has been
edited recently, and when that occurs, Bacula automatically upgrades the job
to Full.

Take a look at the job's summary.  If you have a semi-recent version there
will be a date stamp in the summary next to the fileset.  This date is the
time the Fileset was last edited.

There is a Fileset setting called "IgnoreFilesetChanges = yes"  which can help
mitigate this second case, but there are some warnings to think about when
using this option.

Also, I have a feature request that has been accepted in the Enterprise
version that the reason why a job was upgraded to a Full is listed in the logs
too.   I do not know when this will be ported to the community version though.


Best regards,

Bill


--
Bill Arlofski
http://www.revpol.com/bacula
-- Not responsible for anything below this line --

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Jobs don't remain on schedule

Bill Arlofski-4
On 03/31/2017 02:39 PM, Craig Shiroma wrote:

> Hi Bill,
>
> Hmmm, I know I didn't touch the fileset(s) and I have IgnoreFilesetChanges set
> to "yes" on these filesets.  The FileSet date on one of the jobs is
> "2015-10-10 18:00:12".
>
> Here are some configs for one of the problem jobs:
> Pool {
>     Name = server_shortname
>     Storage = server_shortname
>     Pool Type = Backup
>     Recycle = yes
>     Recycle Oldest Volume = yes
>     AutoPrune = no
>     Volume Retention = 15 days
>     Maximum Volumes = 16
>     Volume Use Duration = 23h
>     Label Format = "${Client}-${JobId}-vol"
> }
>
> Client {
>     Name = server_shortname
>     Address = server_fqdn
>     FD Port = 9102
>     Password = "some-password"
>     Catalog = MyCatalog
>     File Retention = 15 days
>     Job Retention =  15 days
>     AutoPrune = no
> }
>
> Job {
>     Name = server_shortname
>     Client = server_shortname
>     Enabled = yes
>     Type = Backup
>     Schedule = WeeklyFullOnFriRunFirst
>     FileSet = Windows_E_NoCompression_FileSet
>     Max Full Interval = 6 days
>     Pool = server_shortname
>     Messages = Standard
>     Write Bootstrap = "/var/spool/bacula/server_shortname.bsr"
>     Priority = 10
>     Reschedule On Error = yes
>     Reschedule Interval = 1 hour
>     Reschedule Times = 3
>     Cancel Lower Level Duplicates = yes
>     Allow Duplicate Jobs = no
>     Spool Attributes = yes
>     Accurate = yes
> }
>
> Thanks for the help!
> Craig

Hi Craig,

To me it looks possible that the "Max Full Interval" in your job resource
might be the culprit here. It seems a bit short to me if you are running a
Full weekly.

Take a look at the output of:

* list job=server_shortname    (substitute the correct job here)

See if a recent previous one has failed, or was cancelled, etc. Actually, even
check to see when the last Full was run. If it was more than 6 days, then this
"Max Full Interval" will trigger the automatic upgrade to Full.

Have a great weekend!

Bill


--
Bill Arlofski
http://www.revpol.com/bacula
-- Not responsible for anything below this line --

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Jobs don't remain on schedule

Kern Sibbald
In reply to this post by Craig Shiroma

Hello Craig,

Given the first message below:

    No prior or suitable Full backup found in catalog. Doing FULL backup.

I would agree with Bill's first point.  It appears that your Full backup was pruned, so Bacula had to do another one.

Best regards,

Kern


On 03/31/2017 08:49 PM, Craig Shiroma wrote:
Hello All,

I have a few jobs that are on the below schedule:

Schedule {
  Name = "WeeklyFullOnFriRunFirst"
  Run = Level=Full fri at 17:58
  Run = Level=Incremental sat-thu at 18:00
}

The jobs followed the schedule until last night which was a Thursday.  An incremental was suppose to run, but a Full was run instead.  Would anyone know what could cause this to happen?  Previous full backup did run on a Friday at 17:58.

The log for these jobs show similar text shown below:
bacula-dir JobId 413490:
bacula-dir JobId 413490: No prior or suitable Full backup found in catalog. Doing FULL backup.
bacula-dir JobId 413490: Start Backup JobId 413490, Job=servername.2017-03-30_18.00.06_05
bacula-dir JobId 413490: Max configured use duration=82,800 sec. exceeded. Marking Volume "servername-258519-vol" as Used.
bacula-dir JobId 413490: Pruning oldest volume "servername-263826-vol"
bacula-dir JobId 413490: There are no more Jobs associated with Volume "servername-263826-vol". Marking it purged.
bacula-dir JobId 413490: All records pruned from Volume "servername-263826-vol"; marking it "Purged"
bacula-dir JobId 413490: Using Device "servername" to write.
bacula-sd JobId 413490: Recycled volume "servername-263826-vol" on file device "servername" (/backup01/servername/), all previous data lost.

Seems like something is out of sync with the schedule that is causing the Full to run on the wrong day.

Best regards,
Craig


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot


_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Jobs don't remain on schedule

Craig Shiroma
Hi Bill and Kern,

Yes, it looks like Bill was correct.  The jobs giving the problem were long running jobs (12+ hours) so I think it just so happened that occasionally (didn't happen every week), the full got pruned due to timing.

I increased "Max Full Interval" to 7 and things have been okay for 2 weeks now.

Thank you for the help guys!

-craig


On Sat, Apr 1, 2017 at 5:21 AM, Kern Sibbald <[hidden email]> wrote:

Hello Craig,

Given the first message below:

    No prior or suitable Full backup found in catalog. Doing FULL backup.

I would agree with Bill's first point.  It appears that your Full backup was pruned, so Bacula had to do another one.

Best regards,

Kern


On 03/31/2017 08:49 PM, Craig Shiroma wrote:
Hello All,

I have a few jobs that are on the below schedule:

Schedule {
  Name = "WeeklyFullOnFriRunFirst"
  Run = Level=Full fri at 17:58
  Run = Level=Incremental sat-thu at 18:00
}

The jobs followed the schedule until last night which was a Thursday.  An incremental was suppose to run, but a Full was run instead.  Would anyone know what could cause this to happen?  Previous full backup did run on a Friday at 17:58.

The log for these jobs show similar text shown below:
bacula-dir JobId 413490:
bacula-dir JobId 413490: No prior or suitable Full backup found in catalog. Doing FULL backup.
bacula-dir JobId 413490: Start Backup JobId 413490, Job=servername.2017-03-30_18.00.06_05
bacula-dir JobId 413490: Max configured use duration=82,800 sec. exceeded. Marking Volume "servername-258519-vol" as Used.
bacula-dir JobId 413490: Pruning oldest volume "servername-263826-vol"
bacula-dir JobId 413490: There are no more Jobs associated with Volume "servername-263826-vol". Marking it purged.
bacula-dir JobId 413490: All records pruned from Volume "servername-263826-vol"; marking it "Purged"
bacula-dir JobId 413490: Using Device "servername" to write.
bacula-sd JobId 413490: Recycled volume "servername-263826-vol" on file device "servername" (/backup01/servername/), all previous data lost.

Seems like something is out of sync with the schedule that is causing the Full to run on the wrong day.

Best regards,
Craig


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot


_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/bacula-users
Loading...