Skip to content
This repository has been archived by the owner on Sep 30, 2024. It is now read-only.

Handle MariaDB behavior of dropping relay log entries on failure scenarios #1374

Merged
merged 5 commits into from
Jun 30, 2021

Conversation

shlomi-noach
Copy link
Collaborator

Fixes #1363

  • Avoid "stop replication nicely", hence tamper with IO_THREAD on MariaDB
  • Avoid starting replication on regrouped replicas leader in failover & promotion scenarios

The two changes together solve the issue of losing relay log entries in MariaDB on failover. The loss of entries (MariaDB specific behavior, explained in #1260 (comment)) causes prootion to proceed even if DelayMasterPromotionIfSQLThreadNotUpToDate is enabled.

@shlomi-noach shlomi-noach merged commit c738731 into master Jun 30, 2021
@shlomi-noach shlomi-noach deleted the recovery-mariadb-delay-replication branch June 30, 2021 10:06
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Orchestrator promotes a replica with lag on Mariadb
1 participant