Hi,
I have restarted the Elsaticsearch worker nodes. After restarting, one of the indexes is stuck on INITIALIZING mode and it is been quite a long time.
I ran:
curl -k -XGET 'https://eu1.elasticsearch.svc.cluster.local:31000/_cluster/allocation/explain?pretty' -H 'Content-Type: application/json' -d'{"index": "index-1","shard":2,"primary": true}'
{
"index" : "index-1",
"shard" : 2,
"primary" : true,
"current_state" : "initializing",
"unassigned_info" : {
"reason" : "NODE_LEFT",
"at" : "2020-11-10T16:16:37.317Z",
"details" : "node_left [ktIXm-hVRliZA7FbWElCAw]",
"last_allocation_status" : "throttled"
},
"current_node" : {
"id" : "PrPXcMPpTuGvVxBkwXNt1w",
"name" : "elasticsearch-es-default-1",
"transport_address" : "10.1.218.232:9300",
"attributes" : {
"ml.machine_memory" : "12884901888",
"ml.max_open_jobs" : "20",
"xpack.installed" : "true"
}
},
"explanation": "the shard is in the process of initializing on node [elasticsearch-es-default-1], wait until initialization has completed"
}
The reason stated NODE_LEFT, I am not able to understand that because out of all shards only one shard is stuck and all the other shards on that elasticsearch-es-default-1
node are assigned.
index-1 2 p INITIALIZING 10.1.218.232 elasticsearch-es-default-1
index-1 2 r UNASSIGNED
What can be done to fix this?