Add search_query_current and indexing_index_current

[locmai]

Signed-off-by: Loc Mai [email protected]

We are missing these 2 important metrics for our ES clusters:

indexing_index_current
search_query_current

[SuperQ]

These metric names are very confusing.

[locmai]

@SuperQ Big thanks for helping to review this! The _current naming that I follow similar to search_scroll_current one and also the pattern that already be in the code: QueryTotal => query_total, QueryCurrent => query_current.

Also updated the descriptions ^^

[SuperQ]

In Prometheus, we typically don't use direct 1:1 mappings for these kinds of things. The problem is a lot of systems (like Elastic) use statsd-style naming conventions, which are very hard to understand. We fix the naming in order to make metrics easier to use by the end user.

[locmai]

In Prometheus, we typically don't use direct 1:1 mappings for these kinds of things. The problem is a lot of systems (like Elastic) use statsd-style naming conventions, which are very hard to understand. We fix the naming in order to make metrics easier to use by the end user.

Great point. But the indices.go looks like everything is 1:1 mapping. I think I will fix the new ones first then.

[locmai]

Hi @sysadmind could you help me take a look at this PR?