mariadb_recovery playbooks doesn't work as expected
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
kolla-ansible |
Won't Fix
|
Undecided
|
Unassigned |
Bug Description
My kolla-asible version is queens(6.0.0). In a multi-node deployment, if I stop mariadb container on all nodes, then run "kolla-ansible -i multinode mariadb_recovery", I will get some errors like this:
TASK [mariadb : Registering MariaDB seqno variable] *******
fatal: [controller-1]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: list object has no element 0\n\nThe error appears to have been in '/usr/share/
line 46 of /usr/share/
shell: "tail -n 200 /tmp/mariadb_
where $7 should be $8, as I checked one of the mariadb.log, it looks like this:
[root@controller-1 mariadb]# cat mariadb.log |grep Recovered | tail -1
2018-05-08 4:11:29 140380250507456 [Note] WSREP: Recovered position: e4f12a6f-
If I change $7->$8, then mariadb_recovery run successfully.
Queens is no longer supported.