d465e25f4b
Detect if crm_diff is fixed and if it is favour the crm_diff resource comparison. Also introduce a 'update_settle_secs' parameter for resources that can be used to override the amount of time (s) to wait for the cluster to settle after updating a resource. Tested this as follows: 1) Normal deploys work fine (the code never gets triggered anyway) 2) Redeploy with no changes -> pcmk resources are untouched (the code does not get triggered anyway) 3) Redeploy by adding the following hiera keys: ExtraConfig: pacemaker::resource::bundle::deep_compare: true pacemaker::resource::ip::deep_compare: true pacemaker::resource::ocf::deep_compare: true -> pcmk resources are untouched 4) Redeploy with keys at 3) but with changes that trigger a restart -> pcmk resources are restarted and cluster waits for the restart before continuing Change-Id: I62d5662327333e17e1c32a029695b3d3a904ca10 |
||
---|---|---|
.. | ||
puppet | ||
serverspec/type |