Dear All,
I am running Version: 5.0.3 (04 August 2010) i386-pc-solaris2.10
solaris 5.10 Bacula.
I have a situation where backups are running and the client status
reports the backups as OK, but they jobs do not appear when I query
"List all backups for a Client". I know that the jobs aren't making it
into the Catalog because I have a Nagios check that runs against the
Catalog DB for Client Job status. The backup ran as the client
reports, but the check and sql query don't find it in the Catalog.
This seems to happen during when my very large backups are running
with a status of "Dir inserting Attributes". However, I don't get any
error in the job status from Bacula and am not getting any MySQL, DIR,
SD, or FD errors. The "very large" job backs up 239,897,443 files.
This just might be more than MySQL can handle. I'm just wondering why
I'm not getting any errors from Bacula about it, or where to find them
that I'm not picking up on them.
Also, right now my Director is hung. The very large job is "Dir
inserting Attributes" and there is a job with "fatal error". Anytime
I've ever had "fatal error", I've had to restart the director. It just
won't recover, cancel the job, process other queued jobs, nothing.
Trying to cancel the job results in the message that its marked to be
cancelled, but bconsole then never returns to a prompt.
Console connected at 18-Aug-11 11:52
JobId Level Name Status
======================================================================
95559 Full Advisen_NFS2_Documents_Tape.2011-08-17_23.00.00_50 Dir
inserting Attributes
95560 Full Advisen_NFS2_AMX_Tape.2011-08-17_23.00.00_51 is waiting
on max Storage jobs
95561 Full Advisen_NFS2_SEC_Tape.2011-08-17_23.00.00_52 is waiting
on max Storage jobs
95565 Full Mentora_NAS_Weekly_Tape.2011-08-18_00.00.00_57 is
waiting on max Storage jobs
95593 Full bop-prod-dw-ts_Daily_Disk.2011-08-18_02.30.00_25 has a
fatal error
95608 Increme bop-prod-bm06_Daily_Disk.2011-08-18_03.00.00_40 is
waiting on Storage Shopbop_Files
95609 Increme bop-prod-web01_Daily_Disk.2011-08-18_03.00.00_41 is
waiting on Storage Shopbop_Files
95610 Increme bop-prod-adm2_Daily_Disk.2011-08-18_03.00.00_42 is
waiting on Storage Shopbop_Files
95614 Increme bop-prod-app06_Daily_Disk.2011-08-18_04.00.00_46 is
waiting on Storage Shopbop_Files
====
Select Job:
1: JobId=95559 Job=Advisen_NFS2_Documents_Tape.2011-08-17_23.00.00_50
2: JobId=95560 Job=Advisen_NFS2_AMX_Tape.2011-08-17_23.00.00_51
3: JobId=95561 Job=Advisen_NFS2_SEC_Tape.2011-08-17_23.00.00_52
4: JobId=95565 Job=Mentora_NAS_Weekly_Tape.2011-08-18_00.00.00_57
5: JobId=95593 Job=bop-prod-dw-ts_Daily_Disk.2011-08-18_02.30.00_25
6: JobId=95608 Job=bop-prod-bm06_Daily_Disk.2011-08-18_03.00.00_40
7: JobId=95609 Job=bop-prod-web01_Daily_Disk.2011-08-18_03.00.00_41
8: JobId=95610 Job=bop-prod-adm2_Daily_Disk.2011-08-18_03.00.00_42
9: JobId=95614 Job=bop-prod-app06_Daily_Disk.2011-08-18_04.00.00_46
Choose Job to cancel (1-9): 5
2001 Job bop-prod-dw-ts_Daily_Disk.2011-08-18_02.30.00_25 marked to be canceled.
Any help with diagnosing and solving this issue is very appreciated.
I'm almost wondering if I've reached the end of the capabilities of
Bacula (probably not Bacula's fault, but just too much for MySQL or
PostgreSQL to bear with not any other viable backend) I tuned for 15
million rows originally, don't know if I can tune for what is probably
a billion rows or more.
Yours,
Shon
------------------------------------------------------------------------------
Get a FREE DOWNLOAD! and learn more about uberSVN rich system,
user administration capabilities and model configuration. Take
the hassle out of deploying and managing Subversion and the
tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
