sap-linuxlab
sap_ha_pacemaker_cluster: corosync token timeout setting per platform
Cloud providers documented increased token timeouts to be configured in pacemaker clusters. The role should cover this and make sure that for other platforms the config (e.g. through ha_cluster role variable) remains optional.
- https://cloud.google.com/solutions/sap/docs/sap-hana-ha-config-rhel
- https://docs.aws.amazon.com/sap/latest/sap-hana/sap-hana-on-aws-cluster-configuration-1.html#sap-hana-on-aws-corosync-increase
- https://learn.microsoft.com/en-us/azure/virtual-machines/workloads/sap/high-availability-guide-rhel-pacemaker?tabs=msi#cluster-installation
@ja9fuchs is this issue still open?
I see different values for some cloud providers? for instance here
https://github.com/search?q=repo%3Asap-linuxlab%2Fcommunity.sap_install%20token&type=code
Variance across Infrastructure Platforms of every parameter (for Linux Pacemaker config, Fencing Agent config, Resource Agent config) has already been worked on - a rather large Google Sheet. It remains in the backlog, probably for the next release or the release after.
@sean-freeman if possible please share this spreadsheet with me. I've been working on ansible role to test cluster configuration https://galaxy.ansible.com/ui/repo/published/sap/sap_operations/content/role/pcs_config_checks/
At the moment this role only contains rudimentary checks.
This information is very valuable, I do not want to gather everything from documentation again, if this already been done.
Thank you
@kksat That document is not something understandable by anyone else than who compiled it, and it needs a once-through all of the documents to spot mistakes with fresh eyes after the holiday season - it's not something to share at this time. Better to wait a week or two.
I'm not going to look into that subdir, but would assume it has similar features to @rhmk work a while back > sap-linuxlab/community.sap_operations/roles/sap_hana_sr_takeover