[elasticsearch] recommended method to automate an api call (e.g. ILM) post deployment?

This is helm related so I am not asking in the ES forum.

I would like to automate the following api call:

PUT _template/my_template
{
  "index_patterns": [
    "logstash-*"
  ],
  "settings": {
    "number_of_shards": 8,
    "number_of_replicas": 1
  }
}

I am trying with these values in the helm chart, essentially the idea is to wait until the ES cluster has started (relies heavily on health check mechanism that creates the “start file”)… is this a recommended approach?

lifecycle:
  postStart:
    exec:
      command:
        - sh
        - -c
        - |
          #!/bin/sh
          # Add a template to adjust number of shards/replicas
          TEMPLATE_NAME=my_template
          START_FILE=/tmp/.es_start_file
          INDEX_PATTERN="logstash-*"
          SHARD_COUNT=8
          REPLICA_COUNT=1
          while [ ! -f ${START_FILE} ]; do
            sleep 1
          done
          curl -XPUT "http://localhost:9200/_template/$TEMPLATE_NAME" -H 'Content-Type: application/json' -d'{"index_patterns":['"$INDEX_PATTERN"'],"settings":{"number_of_shards":'$SHARD_COUNT',"number_of_replicas":'$REPLICA_COUNT'}}'

FWIW have not gotten it to work yet, pod remains in Initialization phase:

elasticsearch-master-2                   0/1     PodInitializing   0          12m8s

But let me know if I am on the right track please.

cc @jmlrt @Crazybus would love to get your input.

[pbecotte]

https://helm.sh/docs/topics/charts_hooks/

Helm hooks are the best way to do this IMO, but these helm charts dont have any extension point built in for it.

edit- sorry, didn’t realize you were running from the ES pod

In your wait loop, instead of waiting for that file, wait for the api to return a 200 before you issue the PUT command- I’d guess its a race condition.

https://kubernetes.io/docs/tasks/configure-pod-container/attach-handler-lifecycle-event/#discussion

[naseemkullah]

Thanks @pbecotte,

I like the idea of removing the dependency on that file, so I’ve tried:

lifecycle:
  postStart:
    exec:
      command:
        - bash
        - -c
        - |
          #!/bin/bash
          # Add a template to adjust number of shards/replicas
          TEMPLATE_NAME=my_template3
          INDEX_PATTERN="logstash3-*"
          SHARD_COUNT=8
          REPLICA_COUNT=1
          ES_URL=http://localhost:9200
          while [[ "$(curl -s -o /dev/null -w '%{http_code}\n' $ES_URL)" != "200" ]]; do sleep 1; done
          curl -XPUT "$ES_URL/_template/$TEMPLATE_NAME" -H 'Content-Type: application/json' -d'{"index_patterns":['\""$INDEX_PATTERN"\"'],"settings":{"number_of_shards":'$SHARD_COUNT',"number_of_replicas":'$REPLICA_COUNT'}}'

Which seems like a sufficient solution.