Home > database >  Accidental force merge Elasticsearch
Accidental force merge Elasticsearch

Time:10-28

I executed force merge and forgot to add index name which I want to merge... Now the task is ongoing for some time and it has description:

Force-merge indices [], maxSegments[4], onlyExpungeDeletes[false], flush[true]

Does it mean it tries to force merge all indices present in the cluster or what? I wanted to stop it, but it is not possible... What can I do in this case? it eats all the ram and I am afraid of nodes failure when it will be ongoing forever.

This is output for GET _tasks?detailed&actions=*forcemerge

{
  "nodes" : {
    "vCmAwy1VRZm0yyw7dF_52g" : {
      "name" : "node01",
      "transport_address" : "10.10.10.11:9300",
      "host" : "10.10.10.11",
      "ip" : "10.10.10.11:9300",
      "roles" : [
        "data",
        "data_cold",
        "data_content",
        "data_frozen",
        "data_hot",
        "data_warm",
        "ingest",
        "master",
        "ml",
        "remote_cluster_client",
        "transform"
      ],
      "attributes" : {
        "ml.machine_memory" : "50344366080",
        "ml.max_open_jobs" : "512",
        "xpack.installed" : "true",
        "ml.max_jvm_size" : "21361459200",
        "hwc" : "hot",
        "transform.node" : "true"
      },
      "tasks" : {
        "vCmAwy1VRZm0yyw7dF_52g:4016904417" : {
          "node" : "vCmAwy1VRZm0yyw7dF_52g",
          "id" : 4016904417,
          "type" : "transport",
          "action" : "indices:admin/forcemerge",
          "description" : "Force-merge indices [], maxSegments[4], onlyExpungeDeletes[false], flush[true]",
          "start_time_in_millis" : 1666792931893,
          "running_time_in_nanos" : 5189657499371,
          "cancellable" : false,
          "headers" : { }
        },
        "vCmAwy1VRZm0yyw7dF_52g:4017256604" : {
          "node" : "vCmAwy1VRZm0yyw7dF_52g",
          "id" : 4017256604,
          "type" : "transport",
          "action" : "indices:admin/forcemerge",
          "description" : "Force-merge indices [], maxSegments[4], onlyExpungeDeletes[false], flush[true]",
          "start_time_in_millis" : 1666794829797,
          "running_time_in_nanos" : 3291753989638,
          "cancellable" : false,
          "headers" : { }
        }
      }
    }
  }
}

Command which I executed:

curl -s -u'login:pass'-H 'Content-Type: application/json' -XPOST -k "https://10.10.10.11:9200/_forcemerge?max_num_segments=4&pretty"

CodePudding user response:

You could restart your nodes one after the other which would cancel the tasks, but at the risk of creating instability

  • Related