Would you mind to try to run the job with a debug level set to 150 +
timestamp. and attach the job log.
I was trying to fix that case, and just emit a warning when the database
has removed and changed a file during the backup.
Le lundi 18 septembre 2023 à 17:02:23 UTC+2, Dennis Benndorf a écrit
Hi Bruno,
its me once again. The next run with excluded pg_subtrans dir fails on
the base directory:
JobId 5116772: Fatal error: bareosfd: Traceback (most recent call
last):
File "/usr/lib64/bareos/plugins/BareosFdWrapper.py", line 61, in
start_backup_file
return bareos_fd_plugin_o
That Problem is a bit subtle...
You're using the Count parameter in your Device resource in the SD.
This will multiply your device, but the resulting devices will have
different names (i.e. there is a number added).
For a quick test, comment out the line "Count=X".
To make it work with count
Sorry in fact it is in 15 doc too.
Le lundi 18 septembre 2023 à 15:59:34 UTC+2, Bruno Friedmann
(bruno-at-bareos) a écrit :
> Hello Dennis,
>
> Nice you already tried the fresh PG16 meat, after checking the new
> documentation about subtrans we can see that upstream has added that
> directory
Hello Dennis,
Nice you already tried the fresh PG16 meat, after checking the new
documentation about subtrans we can see that upstream has added that
directory as shouldn't be saved.
```The contents of the directories pg_dynshmem/, pg_notify/, pg_serial/,
pg_snapshots/, pg_stat_tmp/, and pg_s
Hi Bruno,
we tested your new plugin with a quite large database with a running
application using it.
When doing a full backup we ran into:
JobId 5116734: Fatal error: bareosfd: Traceback (most recent call
last):
File "/usr/lib64/bareos/plugins/BareosFdWrapper.py", line 61, in
start_ba