Hi all,
I'm currently backing up mostly remote hosts. Generally our connectivity
is good. Sometimes it's not... especially at night I find that our
backup window coincides with someone's maintenance window.
How does bacula deal with this? It didn't look like it retried the job,
which wouldn't really bother me on a daily incremental run, but on a full
run that needs to happen overnight or over the weekend when no one is in
the office, putting the job off until the next run and hoping that someone
will remove the daily tape and put in a tape from the full backup pool
could be problematic.
What's the best way to deal with this?
And on a slightly different topic:
Currently I have one last full job that needs to run before tomorrow.
I've manually scheduled it after it failed two nights in a row, but it
never seems to actually kick off. I get the following messages:
15-Jul 22:48 devel2-dir: No prior Full backup Job record found.
15-Jul 22:48 devel2-dir: No prior or suitable Full backup found in
catalog. Doing FULL backup.
15-Jul 22:48 devel2-dir: Start Backup JobId 165, Job=b02.2007-07-15_22.48.33
15-Jul 22:48 devel2-sd: Volume "Monthly0001" previously written, moving to
end of data.
And then it sits there, for hours. I see no traffic to the client during
this time, no sign of the spool file being created, and no real db
activity. I can't look at the tape status since the SD seems to block
"mt" from talking to the drive. This is an LTO-2 drive, so I wouldn't
expect it to take hours to find the last file on the tape...
If I look at the status of the director, I see the job listed as
"running":
Running Jobs:
JobId Level Name Status
======================================================================
165 Full b02.2007-07-15_22.48.33 is running
====
However if I do "status storage" it sits there and then eventually returns
this error:
----
Connecting to Storage daemon Tape at devel2:9103
Failed to connect to Storage daemon Tape.
====
You have messages.
*mess
15-Jul 22:44 devel2-dir: *Console*.2007-07-15_22.15.30 Fatal error:
authenticate.c:113 Director unable to authenticate with Storage daemon on
"devel2:9103". Possible causes:
Passwords or names not the same or
Maximum Concurrent Jobs exceeded on the SD or SD networking messed up
(restart daemon).
Please see http://www.bacula.org/rel-manual/faq.html#AuthorizationErrors
for help.
----
I do not run concurrent jobs, the SD is set to take up to 30 though, and
the password is correct.
After restarting the SD, this is in the message queue:
----
15-Jul 22:44 devel2-dir: *Console*.2007-07-15_22.15.30 Error: bnet.c:439
Write error sending 23 bytes to Storage daemon:devel2.biglist.com:9103:
ERR=Broken pipe
15-Jul 22:44 b02-fd: b02.2007-07-15_22.18.52 Fatal error: job.c:1749 Comm
error with SD. bad response to Append Data. ERR=Resource temporarily
unavailable
----
Any ideas?
Thanks for your patience,
Charles
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
