Hi guys.

I have a peculiar case - to me at least - here.
Unless I tell cluster to move resource to a node (two-node cluster) with '--master' cluster keeps both nodes as "slaves"
As soon as I remove such a constraint both nodes start to log:
...
3442363:M 07 Jul 2022 20:05:31.561 # Setting secondary replication ID to b14b339d4b0162c3bcd73c9591b60fa491f475dd, valid up to offset: 435. New replication ID is d704c3a0815fdca98c22392341c7a05e2bb89429 3442363:M 07 Jul 2022 20:05:31.561 * MASTER MODE enabled (user request from 'id=66 addr=/run/redis/redis.sock:0 laddr=/run/redis/redis.sock:0 fd=8 name= age=0 idle=0 flags=U db=0 sub=0 psub=0 multi=-1 qbuf=34 qbuf-free=40920 argv-mem=12 obl=0 oll=0 omem=0 tot-mem=61476 events=r cmd=slaveof user=default redir=-1') 3442363:M 07 Jul 2022 20:05:32.269 * Replica 10.0.1.7:6379 asks for synchronization 3442363:M 07 Jul 2022 20:05:32.269 * Partial resynchronization request from 10.0.1.7:6379 accepted. Sending 0 bytes of backlog starting from offset 435. 3442363:S 07 Jul 2022 20:11:19.175 # Connection with replica 10.0.1.7:6379 lost. 3442363:S 07 Jul 2022 20:11:19.175 * Before turning into a replica, using my own master parameters to synthesize a cached master: I may be able to synchronize with the new master with just a partial transfer. 3442363:S 07 Jul 2022 20:11:19.175 * Connecting to MASTER no-such-master:6379 3442363:S 07 Jul 2022 20:11:19.177 # Unable to connect to MASTER: (null) 3442363:S 07 Jul 2022 20:11:19.177 * REPLICAOF no-such-master:6379 enabled (user request from 'id=90 addr=/run/redis/redis.sock:0 laddr=/run/redis/redis.sock:0 fd=10 name= age=0 idle=0 flags=U db=0 sub=0 psub=0 multi=-1 qbuf=48 qbuf-free=40906 argv-mem=25 obl=0 oll=0 omem=0 tot-mem=61489 events=r cmd=slaveof user=default redir=-1')

and then infamous:
...
3442363:S 07 Jul 2022 20:11:24.184 # Unable to connect to MASTER: (null) 3442363:S 07 Jul 2022 20:11:25.187 * Connecting to MASTER no-such-master:6379
..

What is PCS/agent getting wrong there?
btw. OCF Redis does not need 'redis-sentinel.conf' , right?

many thanks, L.
_______________________________________________
Manage your subscription:
https://lists.clusterlabs.org/mailman/listinfo/users

ClusterLabs home: https://www.clusterlabs.org/

Reply via email to