[ClusterLabs] split brain cluster

Richard Korsten richard at rkorsten.nl
Mon Nov 16 08:43:17 EST 2015


Hello Cluster guru's.

I'm having a bit of trouble with a cluster of ours. After an outage of 1
node it went into a split brain situation where both nodes aren't talking
to each other. Both say the other node is offline. I've tried to get them
both up and running again by stopping and starting the cluster services on
both nodes, one at a time. with out luck.

I've been trying to reproduce the problem with a set of test servers but i
can't seem to get it in the same state.

Because of this i'm looking for some help because i'm not that known with
pacemaker/corosync.

this is the output of the command pcs status:
Cluster name: MXloadbalancer
Last updated: Mon Nov 16 10:18:44 2015
Last change: Fri Nov 6 15:35:22 2015
Stack: corosync
Current DC: bckilb01 (1) - partition WITHOUT quorum
Version: 1.1.12-a14efad
2 Nodes configured
3 Resources configured

Online: [ bckilb01 ]
OFFLINE: [ bckilb02 ]

Full list of resources:
 haproxy (systemd:haproxy): Stopped

Resource Group: MXVIP
ip-192.168.250.200 (ocf::heartbeat:IPaddr2): Stopped
ip-192.168.250.201 (ocf::heartbeat:IPaddr2): Stopped

PCSD Status:
bckilb01: Online
bckilb02: Online

Daemon Status:
corosync: active/enabled
pacemaker: active/enabled
pcsd: active/enabled


And the config:
totem {
version: 2
secauth: off
cluster_name: MXloadbalancer
transport: udpu }

nodelist {
node { ring0_addr: bckilb01 nodeid: 1 }
node { ring0_addr: bckilb02 nodeid: 2 } }
quorum { provider: corosync_votequorum two_node: 1 }
logging { to_syslog: yes }

If any has an idea about how to get them working together again please let
me know.

Greetings Richard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clusterlabs.org/pipermail/users/attachments/20151116/1afabc2b/attachment-0002.html>


More information about the Users mailing list