[CENTREON-HA][24.10] My two database nodes are slaves.

Question

Hello,My two database nodes are now slaves after an unsuccessful attempt to switch from slave to master.I am looking for a way to switch my primary node back to MASTER status. Currently, it is the active one operating in the HA setup. Previous command # pcs resource move ms_mysql    Error: to move promotable clone resources you must use the promotable clone id (ms_mysql-clone)    Error: Errors have occurred, therefore pcs is unable to continue # pcs resource move ms_mysql-clone    Warning: Creating location constraint 'cli-ban-ms_mysql-clone-on-BDD_SECONDARY_NODE' with a score of -INFINITY for resource ms_mysql-clone on BDD_SECONDARY_NODE.            This will prevent ms_mysql-clone from running on BDD_SECONDARY_NODE until the constraint is removed            This will be the case even if BDD_SECONDARY_NODE is the last node in the cluster Result:[root@BDD_SECONDARY_NODE]# /usr/share/centreon-ha/bin/mysql-check-status.shConnection SLAVE Status 'BDD_PRIMARY_NODE' [OK]Connection SLAVE Status 'BDD_SECONDARY_NODE' [OK]Slave Thread Status [KO]Error reports:    No slave (maybe because we cannot check a server).Position Status [SKIP]Error reports:    Skip because we can't identify a unique slave.

fgbetokpanou · Answer

Hi ​@benjamin.biguet, thank you for the detailed output. Because this involves the state of your database replication on a production cluster, we’d recommend following the official guides rather than running ad-hoc commands: Troubleshooting HA and the Operating guide.Please note that the pcs resource move command you used leaves a location constraint on the resource (as shown in your warning message), which can prevent it from running on that node until it is removed.If you have a Centreon support contract, we’d advise opening a ticket so that our support team can look at your cluster safely. Otherwise, please let us know your version, and we’ll forward your case to our team.

Sign up

Login to the community