Dear members
Request suggestion. Default value of pc.wait_prim_timeout is 30S. In a 3 node cluster, if I set it to 180S, will it help in a situation where bootstrap node starts after other 2 cluster members within 3mins
Thanx
Dear members
Request suggestion. Default value of pc.wait_prim_timeout is 30S. In a 3 node cluster, if I set it to 180S, will it help in a situation where bootstrap node starts after other 2 cluster members within 3mins
Thanx
Hi!
Defaults are usually safe values. 30 seconds should be good enough for most cases
If in your setup you need a larger timeout you should override the default in the configuration file through wsrep_provider options
Regards
If you already have 2 cluster members, then you should not have a bootstrap node. Bootstrapping creates a new cluster. If you have 2 nodes running, that is cluster A. Bootstrap node 1, and now you have cluster A and cluster B.
If this is an “all down” situation, then yes, increasing the timeout will allow for a longer startup.
Based on documentation, a node waits 30s for quorum before it continues start up sequence. By default pc.wait_for_timeout is set to 30s
Lets say, after full outage, SA or DBA didn’t follow correct startup sequence - bootstrap node was not the first to come up. A node with bootstrap set to 0 was started up . My intention is to increase duration to node wait from 30s to 180s
So in 180s, I expect bootstrap node to come up.
Unanswered | Unsolved | Solved
MySQL, InnoDB, MariaDB and MongoDB are trademarks of their respective owners.
Copyright © 2006 - 2024 Percona LLC. All rights reserved.