Hi.. I have some troubles with a small Cluster (3 nodes + 3 replicas). The cluster goes down and unavailable (CLUSTERDOWN) for a short time (but several times a day). I have no clue how to fix this.
My redis config looks like this
maxmemory 25769803776
tcp-keepalive 0
tcp-backlog 65536
maxclients 10000
port 7000
dir /var/lib/redis/7000/
appendonly no
protected-mode no
cluster-enabled yes
cluster-node-timeout 5000
cluster-config-file /etc/redis/cluster/7000/nodes_7000.conf
pidfile /var/run/redis/redis_7000.pid
logfile /var/log/redis/redis_7000.log
loglevel notice
my logs
15948:C 02 Apr 2022 23:10:42.473 * DB saved on disk
15948:C 02 Apr 2022 23:10:42.474 * RDB: 0 MB of memory used by copy-on-write
758:M 02 Apr 2022 23:10:42.572 * Background saving terminated with success
758:M 02 Apr 2022 23:15:09.260 * Marking node 7317c5bf2435fedbe2dc0b54a6a07977c43e311d as failing (quorum reached).
758:M 02 Apr 2022 23:15:09.260 # Cluster state changed: fail
758:M 02 Apr 2022 23:15:10.239 * Marking node 008ec4d6a3c71a117bbc08008e6d37acd0c8e29d as failing (quorum reached).
758:M 02 Apr 2022 23:15:10.265 * Marking node cac1f0e1092af9c45a1910502f060853a8f51190 as failing (quorum reached).
758:M 02 Apr 2022 23:15:10.373 * Clear FAIL state for node 008ec4d6a3c71a117bbc08008e6d37acd0c8e29d: replica is reachable again.
758:M 02 Apr 2022 23:15:10.884 * Clear FAIL state for node cac1f0e1092af9c45a1910502f060853a8f51190: replica is reachable again.
758:M 02 Apr 2022 23:15:11.937 # Failover auth granted to cac1f0e1092af9c45a1910502f060853a8f51190 for epoch 43
758:M 02 Apr 2022 23:15:11.940 # Cluster state changed: ok
758:M 02 Apr 2022 23:15:12.698 # Failover auth denied to 7317c5bf2435fedbe2dc0b54a6a07977c43e311d: it is a master node
758:M 02 Apr 2022 23:15:13.484 * Clear FAIL state for node 7317c5bf2435fedbe2dc0b54a6a07977c43e311d: master without slots is reachable again.
758:M 02 Apr 2022 23:20:21.295 * 100 changes in 300 seconds. Saving...
758:M 02 Apr 2022 23:20:21.296 * Background saving started by pid 15983
Comment From: madolson
Hello, we use this github as a place to report questions about Redis implementation, not general trouble shooting. Please consider these resources instead https://redis.io/community/ for getting help about running Redis.