[DRBD-user] DRBD 8.0.13 meta connection shut down by peer

[Amos Shapira]

Hello,

I'm using DRBD on CentOS 5.2 and started getting the following errors
in my logs:

Aug 20 17:09:42 afn2 kernel: drbd0: meta connection shut down by peer.
Aug 20 17:09:42 afn2 kernel: drbd0: peer( Primary -> Unknown ) conn(
Connected -> NetworkFailure ) pdsk( UpToDate -> DUnknown )
Aug 20 17:09:42 afn2 kernel: drbd0: asender terminated
Aug 20 17:09:42 afn2 kernel: drbd0: Terminating asender thread
Aug 20 17:09:42 afn2 kernel: drbd0: sock was shut down by peer
Aug 20 17:09:42 afn2 kernel: drbd0: short read expecting header on sock: r=0
Aug 20 17:09:42 afn2 kernel: drbd0: Writing meta data super block now.
Aug 20 17:09:42 afn2 kernel: drbd0: tl_clear()
Aug 20 17:09:42 afn2 kernel: drbd0: Connection closed
Aug 20 17:09:42 afn2 kernel: drbd0: conn( NetworkFailure -> Unconnected )
Aug 20 17:09:42 afn2 kernel: drbd0: receiver terminated
Aug 20 17:09:42 afn2 kernel: drbd0: Restarting receiver thread
Aug 20 17:09:42 afn2 kernel: drbd0: receiver (re)started
Aug 20 17:09:42 afn2 kernel: drbd0: conn( Unconnected -> WFConnection )
Aug 20 17:09:42 afn2 kernel: drbd0: Handshake successful: DRBD Network
Protocol version 86
Aug 20 17:09:42 afn2 kernel: drbd0: Peer authenticated using 20 bytes
of 'sha1' HMAC
Aug 20 17:09:42 afn2 kernel: drbd0: conn( WFConnection -> WFReportParams )
Aug 20 17:09:42 afn2 kernel: drbd0: Starting asender thread (from
drbd0_receiver [1436])
Aug 20 17:09:42 afn2 kernel: drbd0: peer( Unknown -> Primary ) conn(
WFReportParams -> WFBitMapT ) pdsk( DUnknown -> UpToDate )
Aug 20 17:09:42 afn2 kernel: drbd0: Writing meta data super block now.
Aug 20 17:09:42 afn2 kernel: drbd0: conn( WFBitMapT -> WFSyncUUID )
Aug 20 17:09:42 afn2 kernel: drbd0: conn( WFSyncUUID -> SyncTarget )
disk( UpToDate -> Inconsistent )
Aug 20 17:09:42 afn2 kernel: drbd0: Began resync as SyncTarget (will
sync 0 KB [0 bits set]).
Aug 20 17:09:42 afn2 kernel: drbd0: Resync done (total 1 sec; paused 0
sec; 0 K/sec)
Aug 20 17:09:42 afn2 kernel: drbd0: conn( SyncTarget -> Connected )
disk( Inconsistent -> UpToDate )
Aug 20 17:09:42 afn2 kernel: drbd0: Writing meta data super block now.

I haven't noticed them before. The secondary (which is a Xen guest I
just rebuilt, though the rebuild doesn't touch the DRBD partition)
seems to stay synchronised but I'm a bit worried about failing over to
it.

drbd package version is: 8.0.13-1.el5.centos, kernel is 2.6.18-92.1.10.el5xen

Here is the drbd.conf:

global {
  usage-count yes;
}
common {
  syncer {
    rate 100M;
  }
}
resource afn-data {
  protocol C;
  net {
    cram-hmac-alg sha1;
    shared-secret "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx";
  }
  startup {
    wfc-timeout 0;
    degr-wfc-timeout 120;
  }
  disk {
    on-io-error detach;
  }
  on afn1 {
    device /dev/drbd0;
    disk /dev/afn-prod-vg/afn-data-lv;
    address 74.54.177.447:7789;
    flexible-meta-disk /dev/afn-prod-vg/afn-meta-lv;
  }
  on afn2 {
    device /dev/drbd0;
    disk /dev/afn-prod-vg/afn-data-lv;
    address 74.54.16.232:7789;
    flexible-meta-disk /dev/afn-prod-vg/afn-meta-lv;
  }
}

Google'ing around for these messages usually comes up with
conversations about older drbd versions (0.7).

What's the issue here and how can I fix it?

Thanks,

--Amos