Restart After Unplanned Shutdown
We recommend that you always back up your database before performing any changes to the cluster, even if you do not plan to restore the backup data afterwards.
In case of unplanned shutdown, the seqno value in grastate.dat equals -1.
1. To identify the seqno of every node, it's also stored in the database itself and will be displayed executing adanuc dbid=<dbid> noplog options=autorestart_only.
2. See sample output for database 140, 141 and 142:
...
2022-03-15 14:59:51 %ADANUC-I-ADAADC, Cluster node is at GTID e12a0a0d-a384-11ec-a869-87246b1933e3:14
2022-03-15 14:59:52 %ADANUC-I-DBSHUT, Database 140, Shutdown completed
...
2022-03-15 15:03:15 %ADANUC-I-ADAADC, Cluster node is at GTID e12a0a0d-a384-11ec-a869-87246b1933e3:12
2022-03-15 15:03:16 %ADANUC-I-DBSHUT, Database 141, Shutdown completed
...
2022-03-15 15:04:05 %ADANUC-I-ADAADC, Cluster node is at GTID e12a0a0d-a384-11ec-a869-87246b1933e3:11
2022-03-15 15:04:06 %ADANUC-I-DBSHUT, Database 142, Shutdown completed
...
The GTID is the so called Global Transaction ID and contains the cluster-id and seqno.
GTID = <cluster-id>:<seqno>
3. After identifying the seqno of every node, you can follow the rules for restarting the bootstrap and secondary node(s) as described in steps 4 and 5 of
Restart After Planned Shutdown.