georg@riseup.net
2014-Oct-08 20:57 UTC
[Puppet Users] puppetlabs-corosync: cs_property doesn't get set
[I'm quite new to puppet, so not sure if this is an appropriate question
for this list.]
Hi all,
I'm running Debian wheezy, puppet version 2.7.23. I wanted to have a
look at the corosync module provided by puppetlabs [1].
My code for the test looks quite simple:
class { 'corosync':
enable_secauth => true,
authkey => '/var/lib/puppet/ssl/certs/ca.pem',
bind_address => $ipaddress_eth0,
multicast_address => '239.1.1.2',
}
corosync::service { 'pacemaker':
version => '0',
notify => Service['corosync'],
}
include corosync::reprobe
cs_property { 'stonith-enabled': value => 'false', }
cs_property { 'no-quorum-policy': value => 'ignore', }
cs_property { 'default-resource-stickiness': value => '100',
}
After the agent run, corosync and pacemaker gets installed, and the
cluster gets set up. However, despite values for cs_property being set
(incorrect?), these values aren't being recognized, which leads to
ressources being unuseable. The cluster reports:
crm_verify -LV
crm_verify[12971]: 2014/10/08_22:53:36 WARN: cluster_status: We do not
have quorum - fencing and resource management disabled
crm_verify[12971]: 2014/10/08_22:53:36 ERROR: unpack_resources: Resource
start-up disabled since no STONITH resources have been defined
crm_verify[12971]: 2014/10/08_22:53:36 ERROR: unpack_resources: Either
configure some or disable STONITH with the stonith-enabled option
crm_verify[12971]: 2014/10/08_22:53:36 ERROR: unpack_resources: NOTE:
Clusters with shared data need STONITH to ensure data integrity
Errors found during check: config not valid
I thought that the agent has maybe to run twice, which didn't lead to
success however.
I'm quite clueless and appreciate any help!
Thanks,
Georg
[1] https://forge.puppetlabs.com/puppetlabs/corosync