Unassigned shards: NODE_LEFT (Cloud)

[bogdan_copocean]

Hello guys,

I encountered a few problems with my Elasticsearch cluster (on cloud) and exactly, i have around 25 unassigned shards with the reason NODE_LEFT and one with REPLICA_ADDED (after I tried to add a replica).

I want to solve this problem without having to delete the indices and I looked around for a solution, but I didn't see one for the cloud.

Any suggestions?

Thanks!

[DavidTurner]

NODE_LEFT just says that the shard became unassigned because a node left the cluster. The more interesting question is why the shard is not being assigned again, and for that you should use the cluster allocation explain API.

[bogdan_copocean]

Hi David and thank you for replying. This is the output from the cluster allocation explain API:
{
"decider" : "same_shard",
"decision" : "NO",
"explanation" : "a copy of this shard is already allocated to this node [[test_index-000002][0], node[6NlWny21g5SdCMUzGtlcgWaszSQ], [P], s[STARTED], a[id=_AZRb0yaTJWOncpxzpNB0dsy8Q]]"
},
{
"decider" : "awareness",
"decision" : "NO",
"explanation" : "there are too many copies of the shard allocated to nodes with attribute [logical_availability_zone], there are [2] total configured shard copies for this shard id and [2] total attribute values, expected the allocated shard count per attribute [2] to be less than or equal to the upper bound of the required number of shards per attribute [1]"
}

It is pretty intuitive why this happened, but I don't know how to solve it.

[DavidTurner]

That's not the full output. That's just saying we can't allocate a copy of [test_index-000002][0] to this node because there's already a copy there.