Results 1 to 8 of 8

Thread: Daily backup errors

  1. #1
    socate is offline Senior Member
    Join Date
    Dec 2007
    Location
    Romania
    Posts
    50
    Rep Power
    7

    Default Daily backup errors

    Hi all,

    I'm using ZCS 5.0.2 and I'm using the Open Source backup procedure. Since few day ago, all the time I receive this e-mail:
    ------------
    Date: 13-03-2008, Thursday, 00:00 Hours
    Stopping Zimbra: Success
    Rsync Zimbra: Some Problem Occurred. Please check.
    Starting Zimbra: Success
    Zimbra was off-line for: 2912 seconds
    Rotating Backup: Removed /bkp/11-02-2008-Monday-00-00-1202680801
    The backup took: 2967 seconds
    --------------

    Where I see the errors? Where is the log?

    I see that this error it's connected to "zmremrsyncstat" function from backup script.

    Anybody can give me an advice?

    Thank you!

  2. #2
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,201
    Rep Power
    56

    Default

    I would make a guess at a problem removing an old backup or creating a new directory.
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  3. #3
    uxbod's Avatar
    uxbod is offline Moderator
    Join Date
    Nov 2006
    Location
    UK
    Posts
    8,016
    Rep Power
    24

    Default

    would you run the following commands so that we can check the permissions :-

    Code:
    ls -ld /bkp
    ls -lR /bkp
    Thanks.

  4. #4
    socate is offline Senior Member
    Join Date
    Dec 2007
    Location
    Romania
    Posts
    50
    Rep Power
    7

    Default

    I'm not sure...

    The new folder it's created...the old folder is removed...

    I try to follow the system log (on Open SuSe 10.3) but I can't find anything....

    Not all of CRON backup jobs are errors...for example yesterday was OK; it's not a folder permision issue...

    -----------------
    Date: 12-03-2008, Wednesday, 00:00 Hours Stopping Zimbra: Success Rsync Zimbra: Successfully created /bkp/12-03-2008-Wednesday-00-00-1205272801
    Starting Zimbra: Success
    Zimbra was off-line for: 1980 seconds
    Rotating Backup: Removed /bkp/10-02-2008-Sunday-00-00-1202594402
    The backup took: 2030 seconds
    ------------------

    What error can be?

  5. #5
    Bill Brock is offline Outstanding Member
    Join Date
    May 2007
    Location
    Oklahoma
    Posts
    703
    Rep Power
    8

    Default

    If you run the script manually, I believe it will show you the errors as rsync runs. I think.

  6. #6
    uxbod's Avatar
    uxbod is offline Moderator
    Join Date
    Nov 2006
    Location
    UK
    Posts
    8,016
    Rep Power
    24

    Default

    Looking at the Perl module for RSYNC I believe that it writes stdin/stdout to a array, so you would have to modify the code (using Dumper) to print out the contents.

  7. #7
    dwmtractor's Avatar
    dwmtractor is offline Moderator
    Join Date
    Jul 2007
    Location
    San Jose, CA
    Posts
    1,027
    Rep Power
    9

    Default

    Quote Originally Posted by socate View Post
    I'm not sure...

    The new folder it's created...the old folder is removed...

    I try to follow the system log (on Open SuSe 10.3) but I can't find anything....

    Not all of CRON backup jobs are errors...for example yesterday was OK; it's not a folder permision issue...

    -----------------
    Date: 12-03-2008, Wednesday, 00:00 Hours Stopping Zimbra: Success Rsync Zimbra: Successfully created /bkp/12-03-2008-Wednesday-00-00-1205272801
    Starting Zimbra: Success
    Zimbra was off-line for: 1980 seconds
    Rotating Backup: Removed /bkp/10-02-2008-Sunday-00-00-1202594402
    The backup took: 2030 seconds
    ------------------

    What error can be?
    How large is your backup file? Half an hour offline seems a little excessive to me. How many users/how much data/what hardware specs?

    I ask because my backup takes far less time than this. Total of 33 active accounts, roughly 3.5 or 4 GB total data and programs compressed to 2.2GB of backup file, and according to my zimbra.log file, the actual downtime on my server was about three minutes from the "stopping Zimbra" entry to the "Status OK" one. It takes more time to tar and compress the archive, and to ftp it off to another machine, but the actual down and backup time is very quick. I use the simplest and least feature-rich of all the backup scripts on the wiki, but I have successfully restored a backup using the resultant file so I know it works.

    Dan

  8. #8
    dwmtractor's Avatar
    dwmtractor is offline Moderator
    Join Date
    Jul 2007
    Location
    San Jose, CA
    Posts
    1,027
    Rep Power
    9

    Default

    Oh, and I should have added, your "some problems occurred" could very well be totally harmless. Given the syntax of the error you are describing, I presume you are using the perl, non-LVM backup method here. I don't know perl, but as I read the script, it appears to me that the script does a data comparison of the source and destination folders after the rsync, and sends you the "problems" messages if it finds a difference.

    The "problem" you may encounter is that, though the script does a pid for existing processes after stopping Zimbra, it doesn't appear to wait after stopping them (the finall kill -9) before launching into rsync. If rsync were to happen to grab a temporary file from the source folder that was in the process of being deleted, but wasn't gone yet, it would go ahead and sync that file to your destination directory, but by the time of the compare, the file would be gone from the source directory, thus creating an exception (the converse problem also happens sometimes, in that rsync makes a list of files to synchronize before copying them, and then by the time it tries to get the file it's gone, resulting in rsync itself returning an error).

    You can verify this, or observe other problems, by doing one of two things:
    1. Run the script live from the command line and watch what happens; or
    2. Create a literal log of the whole process by adding something like this string to the end of the cron command that fires off your backup:
    Code:
    > /temp/zbackup.log 2>&1
    Anyway, I am guessing that if you would introduce a few seconds (say 10) of sleep time between the end of the killing of processes, and the starting of rsync, this error would go away.

    How to do this, if in fact my guess is a good one, I leave to people who actually know how to write perl!

    Cheers,

    Dan

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Replies: 658
    Last Post: 04-04-2014, 09:01 AM
  2. Daily Full Backup Math Problems
    By DougWare in forum Administrators
    Replies: 9
    Last Post: 03-18-2008, 08:14 AM
  3. Replies: 0
    Last Post: 01-25-2008, 09:05 AM
  4. FYI: ZCS NE backup to fuse/sshfs mount, worked.
    By jagipson in forum Administrators
    Replies: 0
    Last Post: 09-28-2007, 06:37 AM
  5. NE Backup best practises
    By Klug in forum Administrators
    Replies: 3
    Last Post: 05-29-2006, 08:32 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •