Why Redis node(s) report as unreachable?

[devesh bajpai]

redis-cli --cluster add-node 10.80.252.36:6379 10.80.246.65:6379 --cluster-slave >>> Adding node 10.80.252.36:6379 to cluster 10.80.246.65:6379 >>> Performing Cluster Check (using node 10.80.246.65:6379) M: f02c966118e48771dff786a399f9ad939d5b5d4e 10.80.246.65:6379 slots:[11644-11926] (283 slots) master S: e3a27f390903713101a626a230a443b2702527a5 10.91.135.28:6379 slots: (0 slots) slave replicates 11a95b16f7e749883a3ab8fbb1d0d08b5a98cdd8 S: abfd544ec6180176cd05f09c5952764c3a3a9e46 10.90.71.204:6379 slots: (0 slots) slave replicates 20fa54049ed8b0280c79e63e596fa38d162a77a4 S: 7950a6441fb0d5377cb8fa17f0496af407fa7475 10.80.113.153:6379 slots: (0 slots) slave replicates 2375eb1263ec5be76ee4b0831f099a8f59431273 M: ba6596af5e6db6960da6a50c996692468cf53e9b 10.81.128.50:6379 slots:[2694],[2710],[4903-4991],[7113-7168],[7945-7990],[11046-11134],[13484] (283 slots) master M: 5a58a10ff23ef9af9c627d2dba9cde1a13c45284 10.81.55.128:6379 slots:[4874-4902],[6826-6949],[6952-7080] (282 slots) master 1 additional replica(s) S: e44da6ff12236b4c9d754412f6d97768218e0e80 10.88.20.100:6379 slots: (0 slots) slave .... M: c7a7f8297f07b4b6e19452170d0c18d44cf18d5e 10.90.47.148:6379 slots:[2700],[2707],[3495-3582],[12669-12730],[12856-12895],[14087-14176] (282 slots) master 1 additional replica(s) [OK] All nodes agree about slots configuration. >>> Check for open slots... >>> Check slots coverage... [OK] All 16384 slots covered. Automatically selected master 10.80.246.65:6379 >>> Send CLUSTER MEET to node 10.80.252.36:6379 to make it join the cluster. Waiting for the cluster to join ..................................... Warning: 6 node(s) may be unreachable - Node 10.90.34.147 may be unreachable from: 10.91.39.41:6379 10.91.239.197:6379 Cluster bus ports must be reachable by every node. Remember that cluster bus ports are different from standard instance ports. - Node 10.80.252.36 may be unreachable from: 10.91.39.41:6379 10.91.239.197:6379 Cluster bus ports must be reachable by every node. Remember that cluster bus ports are different from standard instance ports. - The port 16379 of node 10.80.246.65 may be unreachable from: 10.80.252.36:6379 Cluster bus ports must be reachable by every node. Remember that cluster bus ports are different from standard instance ports. - Node 10.80.110.72 may be unreachable from: 10.91.39.41:6379 10.91.239.197:6379 Cluster bus ports must be reachable by every node. Remember that cluster bus ports are different from standard instance ports. - Node 10.80.54.159 may be unreachable from: 10.91.39.41:6379 10.91.239.197:6379 Cluster bus ports must be reachable by every node. Remember that cluster bus ports are different from standard instance ports. - Node 10.90.100.26 may be unreachable from: 10.91.39.41:6379 10.91.239.197:6379 Cluster bus ports must be reachable by every node. Remember that cluster bus ports are different from standard instance ports. .....................................

We're trying to figure out the reason for this behavior. Is there a way to accomplish this without flushing/resetting nodes or restarting cluster?

[al]

I had same problem.

Check SELinux.

A command like the one below may work(All host)

> sudo setenforce 0

2022年3月23日水曜日 21:46:14 UTC+9 thedeve...@gmail.com: