I have deployed a logging project on OpenShift and have created a PV on a GlusterFS, which consists of a 3 servers running in a cluster.
I have created a volume on a GlusterFS and enabled sharding feature on it. However, Elasticsearch doesn't run on it as expected, and I get this error:
FAIL: Configuration for 'config' failed because of UnavailableShardsException[[.searchguard.logging-es-6akrt1z2-11-66509][0] primary shard is not active Timeout: [1m], request: [index {[.searchguard.logging-es-6akrt1z2-11-66509][config][0], source[{"searchguard":{"dynamic":{"http":{"xff":{"enabled":true,"remoteIpHeader":"x-forwarded-for","trustedProxies":".*","internalProxies":".*"}},"authc":{"authentication_domain_proxy":{"enabled":true,"order":0,"http_authenticator":{"challenge":false,"type":"proxy","config":{"user_header":"x-proxy-remote-user"}},"authentication_backend":{"type":"noop"}},"authentication_domain_basic_internal":{"enabled":true,"order":1,"http_authenticator":{"type":"clientcert","challenge":false},"authentication_backend":{"type":"noop"}}}}}}]}]]
How could I fix this?
OK, I solved the issue. It looks like default shard size on a volume was to small, so I set it to a 50GB. Now, a new question arises, what happens if I set Elasticsearch to use 2 shards, and then a shard grows beyond 50 GB?