[DRBD-user] disconnecting hangs after ko-count failure

[Walter Haidinger]

> > How can I further diagnose this? Do you need more information, like the
> > drbd.conf setup?
> 
> Well how did you guess that? ;-)

Don't know, just a wild guess.. ;-)

drbd.conf:
global {
    minor-count 16;
    usage-count yes;
}
common {
  syncer { rate 10M; }
}
resource west {
  protocol C;
  handlers {
    pri-on-incon-degr "echo o > /proc/sysrq-trigger ; halt -f";
    pri-lost-after-sb "echo o > /proc/sysrq-trigger ; halt -f";
    local-io-error "echo o > /proc/sysrq-trigger ; halt -f";
    outdate-peer "/usr/sbin/drbd-peer-outdater";
  }
  startup {
    wfc-timeout 5;
    degr-wfc-timeout 5;
  }
  disk {
    on-io-error   detach;
    fencing dont-care;
  }
  net {
    timeout       100;    # < (connect-int|ping-int), (unit = 0.1 seconds)
    connect-int   30;    # (unit = 1 second)
    ping-int      60;    # (unit = 1 second)
    ping-timeout  50;    # (unit = 0.1 seconds)
    max-buffers         256;
    cram-hmac-alg "sha256";
    shared-secret "blah..."
    after-sb-0pri disconnect;
    after-sb-1pri disconnect;
    after-sb-2pri disconnect;
    rr-conflict disconnect;
  }
  syncer {
    rate 500K;  # _bytes_!
    al-extents 19;
  }
  on west.local {
    device     /dev/drbd0;
    disk       /dev/vg/drbd_offsite;
    address    192.168.99.2:7788;
    flexible-meta-disk  /dev/vg/drbd_meta0;
  }
  on east.local {
    device    /dev/drbd0;
    disk      /dev/vg/drbd_west;
    address   192.168.99.1:7788;
    flexible-meta-disk  /dev/vg/drbd_meta0;
  }
}
resource east {
  # -- cut ---
  # ... like resource west except for:
  syncer {
    rate 500K;  # _bytes_!
    after "west";
    al-extents 19;
  }
  on west.local {
    device     /dev/drbd1;
    disk       /dev/vg/drbd_east;
    address    192.168.99.2:7789;
    flexible-meta-disk  /dev/vg/drbd_meta1;
  }
  on east.local {
    device    /dev/drbd1;
    disk      /dev/vg/drbd_east;
    address   192.168.99.1:7789;
    flexible-meta-disk  /dev/vg/drbd_meta1;
  }
}

> Exact kernel version (uname -r) won't hurt either.

west: drbd-8.0.8, i686   openSuSE 10.3, vanilla kernel 2.6.22.15
east: drbd-8.0.8, x86_64 openSUSE 10.3, kernel 2.6.22.13-0.3-default 

Please note that I had the issue on east too while it ran 32-bit openSuSE 10.1 and drbd-8.0.7. Unfortunately upgrading to (read fresh install) 64-bit 10.3 did not solve the problem. West is also scheduled to be converted to x86_64 but I haven't had time yet. 
Also, 8.0.8 on x86_64 was compiled using a small include header patch found on the drbd mailing-list. 

Anything else you need or some tests I can run?

Regards, Walter
-- 
GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS.
Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail