Hello,
czw., 7 mar 2024 o 22:58 Ismael Matos <iabma...@gmail.com> napisał(a):
> Hello all,
>
> After a while and investing a little time, I've managed to run
> successfully the Postgres database backups using the BPIPE plugin. At least
> I think it worked.
>
> I've got 2 questions:
>
> 1 - The BPIPE Plugin is at the start of the "include" list in the
> "fileset" definition, but it runs at the end of the backup job., after all
> other regular file backups So, the log files from the pg_dump for each
> database are not backed up with the database backup files. Is there a way
> to get the BPIPE Plugin to run first?
>
Have you tried something like this?
FileSet {
Include {
Plugin ="bpipe: ..."
}
Include {
File = "/..."
}
}
2 - I've checked the log from the pg_dumps and they seem to run OK, BUT the
> there is no information about the size of the files in the bacula backup
> (-1 is the file length, when checking with Bacularis). Is this expected?
>
Yes, unfortunately the way how Bacula handles backup requires a plugin to
prepare a metadata for the file it wants to backup (save) before it could
save real data. It is not a problem for regular file data you can get from
any standard filesystem, as you can simply execute a stat(2) function to
get all standard metadata in advance. It won't work with the bpipe plugin
as the size of the data is unknown until all data is read. This is why
bpipe generated backup files got '-1' as a size.
best regards
--
Radosław Korzeniewski
rados...@korzeniewski.net
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
