Grastate.dat seqno = -1

Two out of three of my nodes in the cluster show a -1 for seqno in the /var/lib/mysql/grastate.dat file. The third one(50.116.45.43) has no grastate.dat file present. I did simulate a failure on one node(45.33.103.216) by stopping the mysql service on the node.

Should I be concerned by the -1 seqno value?

select hostgroup_id,hostname,port,status from runtime_mysql_servers;
±-------------±--------------±-----±--------+
| hostgroup_id | hostname | port | status |
±-------------±--------------±-----±--------+
| 10 | 45.33.103.216 | 3306 | SHUNNED |
| 11 | 45.33.103.216 | 3306 | ONLINE |
| 11 | 45.56.115.5 | 3306 | ONLINE |
| 12 | 45.33.103.216 | 3306 | ONLINE |
| 12 | 45.56.115.5 | 3306 | ONLINE |
| 10 | 50.116.45.43 | 3306 | ONLINE |
| 10 | 45.56.115.5 | 3306 | SHUNNED |
±-------------±--------------±-----±--------+

1 Like

@Mikem,
grastate.dat is only updated on clean shutdown, so while the node is running it will always show -1. If your 3rd node has no grastate.dat that tells me that PXC/Galera did not startup successfully.

1 Like