[MCOL-5052] Primary mismatch after restart : MXS vs MCS Created: 2022-04-13 Updated: 2022-04-14 Resolved: 2022-04-14 |
|
| Status: | Closed |
| Project: | MariaDB ColumnStore |
| Component/s: | cmapi |
| Affects Version/s: | None |
| Fix Version/s: | cmapi-1.6.3 |
| Type: | Task | Priority: | Major |
| Reporter: | alexey vorovich (Inactive) | Assignee: | Alan Mologorsky |
| Resolution: | Not a Bug | Votes: | 0 |
| Labels: | None | ||
| Description |
|
After the primary (node 1) in a cluster is killed and then restarted , we have a discrepancy between MXS showing node 2 as primary and MCS showing node 1 as primary . The details are in https://mariadbcorp.atlassian.net/wiki/spaces/~695761903/pages/1629783457/CMAPI+primary+mismatch+observation and in https://jira.mariadb.org/browse/MCOL-4939 The problem goes back to mariadb/skysql-enterprise-columnstore:sky-5.5.2-1-GA from last June |
| Comments |
| Comment by alexey vorovich (Inactive) [ 2022-04-13 ] |
|
More on the subject . It turns out that after some wait (1-2 min) the MCS switches the primary to match MXS. Unless there are objections , I will cancel the Jira tomorrow |