[DRBD-user] Strange secondary behavior

Chamara Peris dimsyboy at gmail.com
Thu Oct 22 01:36:26 CEST 2009

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Hi All,

I've setup a DRBD two node cluster with heartbeat. But I am having trouble
with my secondary node. No matter how many times I do  a sync it still says
"secondary inconsistent". However if I stop the heartbeat service from the
primary the secondary take over the resource.

But If I pull the network plug from the Primary then seconday cannot take
the resources and complain about "at least 1 uptodate disk required"


1/ Please help me to get secondary disk to consistent state.
2/ Simulate network failure won't start secondary (pulling the cable out).
How ever stoping heartbeat gracefully on primary kicks the secondary into
action.


Please find the attached configs for the files.

Thanks heaps,
CP


ha.cf:

logfile /var/log/ha-log
logfacility     local0
keepalive 2
deadtime 30
udpport 694
bcast  eth0
ucast eth0 192.168.0.50
auto_failback off
node primary
node backup
respawn hacluster /usr/lib/heartbeat/ipfail
apiauth ipfail gid=haclient uid=hacluster
respawn hacluster /usr/lib/heartbeat/dopd
apiauth dopd gid=haclient uid=hacluste


drbd.conf:

global {
        usage-count no;
        dialog-refresh 5;
}
common {
        protocol C;
}
resource drbd0 {
  disk {
        fencing resource-only;
 #       on-io-error detach;
  }
  handlers {

        split-brain "/usr/lib/drbd/notify-split-brain.sh root";
        outdate-peer "/usr/lib/heartbeat/drbd-peer-outdater -t 5";
  }
  syncer {
        rate 40M;
        al-extents 257;
   }
  on primary {
        device  /dev/drbd0;
        disk    /dev/sdb1;
        address 192.168.0.51:7789;
        meta-disk       internal;
        }
  on backup {
        device  /dev/drbd0;
        disk    /dev/sdb1;
        address 192.168.0.50:7789;
        meta-disk       internal;
}
net {
        max-buffers 2048;
        ko-count 4;
#       on-disconnect reconnect;
        after-sb-0pri discard-zero-changes;
        after-sb-1pri discard-secondary;
        after-sb-2pri disconnect;
}
startup {
wfc-timeout 0; # drbd init script will wait infinitely on resources.
degr-wfc-timeout 120; # 2 minutes.
}
---------------------------------------------------------------------------------------------------------
primary:/etc/init.d # cat /proc/drbd
version: 8.2.7 (api:88/proto:86-88)
GIT-hash: a1b440e8b3011a1318d8bff1bb7edc763ef995b0 build by lmb at hermes,
2009-02-20 13:35:59
 0: cs:Connected st:Primary/Secondary ds:UpToDate/Inconsistent C r---
    ns:340620 nr:944 dw:12004 dr:1705567 al:21 bm:115 lo:0 pe:0 ua:0 ap:0
ep:1 wo:b oos:96


backup:/etc/init.d # cat /proc/drbd
version: 8.2.7 (api:88/proto:86-88)
GIT-hash: 61b7f4c2fc34fe3d2acf7be6bcc1fc2684708a7d build by phil at fat-tyre,
2008-11-12 16:47:11
 0: cs:Connected st:Secondary/Primary ds:Inconsistent/UpToDate C r---
    ns:944 nr:3844 dw:7360 dr:953 al:13 bm:27 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b
oos:96
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20091022/ca0f475d/attachment.htm>


More information about the drbd-user mailing list