Results 1 to 4 of 4

Thread: "Broken pipe" when chg-disk moves to second (virtual) tape

  1. #1

    Default "Broken pipe" when chg-disk moves to second (virtual) tape

    I'm using chg-disk with virtual tapes. Backups work fine until a second tape is required.

    Here was today's example:

    DailySet1 AMANDA MAIL REPORT FOR November 11, 2008

    FAILURE AND STRANGE DUMP SUMMARY:
    thinkpad.mydomain /home lev 1 STRANGE
    coolmass.mydomain //amd64/amd_digphoto_submissions lev 1 FAILED ["data write: Broken pipe"]
    coolmass.mydomain //amd64/amd_digphoto_submissions lev 1 FAILED [dump to tape failed]

    USAGE BY TAPE:
    Label Time Size % Nb
    DailySet114 1:45 78663M 85.4 19
    DailySet115 0:00 0M 0.0 0

    from log file:

    SUCCESS dumper coolmass.mydomain /cmt_b 20081111 0 [sec 4698.881 kb 69425610 kps 14774.9 orig-kb 69425610]
    SUCCESS taper coolmass.mydomain /cmt_b 20081111 0 [sec 4699.005 kb 69425610 kps 14774.5 {wr: writers 2169552 rdwait 4484.937 wrwait 177.569 filemark 0.098}]
    INFO taper tape DailySet114 kb 94371808 fm 20 writing file: No space left on device
    START taper datestamp 20081111 label DailySet115 tape 1
    INFO taper retrying coolmass.mydomain://amd64/amd_digphoto_submissions.1 on new tape due to: [writing file: No space left on device]
    FAIL dumper coolmass.mydomain //amd64/amd_digphoto_submissions 20081111 1 ["data write: Broken pipe"]

    (Every time this happens to a DLE, it is included and works on the next day's backup)

    It looks to me that my problem is the "Broken pipe" when connecting to the new tape.

    How do I troubleshoot to find out what is breaking the pipe?

    Hosts backedup are both the OpenBSD server and ubuntu and winxp desktops. The "broken pipe" failure seems to occur every time a second tape is needed on a backup.

    Thanks for any ideas

    Thanks

  2. #2
    Join Date
    Mar 2007
    Location
    Chicago, IL
    Posts
    688

    Default

    Are you actually running out of space for your vtapes?

    Are you using a holding disk?

  3. #3

    Default

    Dustin,

    Thanks for the quick response.
    I don't have a holding disk configured. The backups are going directly to a dedicated disk and it still has 160+ gig available.

    Sorry about posting on the 2.6.0 forum, thought I was on the general Installation and configuration.

  4. #4
    Join Date
    Mar 2007
    Location
    Chicago, IL
    Posts
    688

    Default

    You really should use a holding disk -- without a holding disk, you can only back up one DLE at a time. Also, with no holding disk you'll need to set up a fallback split buffer -- see fallback_diskbuffer and fallback_splitsize in amanda.conf. As your configuration is now, tape spanning is disabled, and Amanda is killing the dump of the DLE once it runs out of tape space.

Posting Permissions

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