Elasticsearch's default index recovery prioritisation for historical data

(Aaron Daisley) #1

Hi,

We're reviewing what indices should be prioritised when our Elasticsearch cluster restarts, and before making modifications using settings.priority, we were wondering if Elasticsearch, by default uses some sort of logic to decide to bring up newer indexes over historical indexes?

For example, say I have two indexes:
example-2019.06
example-2019.07

If ES did have the logic I described, example-2019.07 would be brought up first, followed by 2019.06.

I can't find anywhere that says it does, however.

Thanks

(Aaron Daisley) #2

I'm an idiot and didn't notice the following text on https://www.elastic.co/guide/en/elasticsearch/reference/current/recovery-prioritization.html

Unallocated shards are recovered in order of priority, whenever possible. Indices are sorted into priority order as follows:

  • the optional index.priority setting (higher before lower)
  • the index creation date (higher before lower)
  • the index name (higher before lower)
(Aaron Daisley) #3

My followup question would be:

Can you set index.priority through settings in index templates?

(system) closed #4

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.