Feature Idea: Reporting Slow ElasticSearch Queries

NewRelic does a fantastic job making the performance of my Postgres queries clear, and capturing slow instances of each query to aid in debugging. I would love to have similar support for my elasticsearch queries, but this does not seem to be available.

In particular, the two most useful features would be:

1. The ability to assign a name to each elasticsearch query, so that the transaction time graph is split by query.
2. The ability to submit slow queries to New Relic, and have them be displayed in the same way that slow SQL queries are displayed.

It appears that the notice_statement method from the ruby documentation would be appropriate for my second use case. Does a similar option exist in the Python code?

And what is the best way to go about implementing the first. Am I missing something simple?

Thank You,
Jim

New Relic edit

We take feature ideas seriously and our product managers review every one when plotting their roadmaps. However, there is no guarantee this feature will be implemented. This post ensures the idea is put on the table and discussed though. So please vote and share your extra details with our team.

@jimtla For your first point, you should check out this great doc on custom metrics, which I believe should accomplish what you are asking for. For your second point, that is a feature request and I’ll loop in @Linds to make this an official feature request in the Community.

Thank you Eileen -

It looks like custom metrics would let me record data about my distinct queries, and then separately plot that on an insights graph – this is a good idea, and definitely helpful. But using custom metrics will not have any impact on the APM “Databases” tab. Is that correct?

A feature request for the second would be amazing.

Thank you!
Jim

Thanks for the details, @jimtla! Your Feature Request has been added!

@jimtla That’s right. It shouldn’t have an impact on your Databases tab. I’m glad to hear that will work for you!

What we’re really looking for is a way to have this in the databases tab – being able to view all of our queries against our Postgres database is amazing, as is having examples of slow queries. The custom metrics strategy feels ok, but I think we’ll probably just use datadog custom metrics instead of doing custom insights queries to view this data.

Thank you for jumping in to help us understand your use case, @jimtla! I have passed all of this along to our product team!