From 0a53ba43cc8a93f47868ffbbeb7faea325529c81 Mon Sep 17 00:00:00 2001 From: Jiri Stransky Date: Tue, 5 Apr 2016 19:15:39 +0200 Subject: Increase corosync token timeout This might prevent dropping members from corosync cluster on high load environments. Symptoms of this problem happening can sometimes be found in corosync log: dub 05 17:23:45 overcloud-controller-0 corosync[14152]: [MAIN ] Corosync main process was not scheduled for 3691.8391 ms (threshold is 1320.0000 ms). Consider token timeout increase. The default in the Puppet manifest is 1 second, which matches the corosync default, and we override it with hiera to 10 seconds. Change-Id: I5ea850ada657e5eecafa3e8b28613a0ac48e78f3 --- puppet/hieradata/controller.yaml | 1 + puppet/manifests/overcloud_controller_pacemaker.pp | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'puppet') diff --git a/puppet/hieradata/controller.yaml b/puppet/hieradata/controller.yaml index 5e87793a..8766263a 100644 --- a/puppet/hieradata/controller.yaml +++ b/puppet/hieradata/controller.yaml @@ -122,6 +122,7 @@ pacemaker::corosync::cluster_name: 'tripleo_cluster' pacemaker::corosync::manage_fw: false pacemaker::resource_defaults::defaults: resource-stickiness: { value: INFINITY } +corosync_token_timeout: 10000 # horizon horizon::cache_backend: django.core.cache.backends.memcached.MemcachedCache diff --git a/puppet/manifests/overcloud_controller_pacemaker.pp b/puppet/manifests/overcloud_controller_pacemaker.pp index f1092321..54176231 100644 --- a/puppet/manifests/overcloud_controller_pacemaker.pp +++ b/puppet/manifests/overcloud_controller_pacemaker.pp @@ -82,9 +82,9 @@ if hiera('step') >= 1 { $pacemaker_cluster_members = downcase(regsubst(hiera('controller_node_names'), ',', ' ', 'G')) $corosync_ipv6 = str2bool(hiera('corosync_ipv6', false)) if $corosync_ipv6 { - $cluster_setup_extras = { '--ipv6' => '' } + $cluster_setup_extras = { '--token' => hiera('corosync_token_timeout', 1000), '--ipv6' => '' } } else { - $cluster_setup_extras = {} + $cluster_setup_extras = { '--token' => hiera('corosync_token_timeout', 1000) } } class { '::pacemaker': hacluster_pwd => hiera('hacluster_pwd'), -- cgit 1.2.3-korg