Re: Observability in Postgres - Mailing list pgsql-hackers

From Greg Stark
Subject Re: Observability in Postgres
Date
Msg-id CAM-w4HOd0+17zMFct0JVJCdnfJkkLx=DT3SMcNHs9q2fja6zag@mail.gmail.com
Whole thread Raw
In response to Re: Observability in Postgres  (Stephan Doliov <stephan.doliov@gmail.com>)
Responses Re: Observability in Postgres  (Michael Banck <michael.banck@credativ.de>)
List pgsql-hackers
On Tue, 15 Feb 2022 at 22:48, Stephan Doliov <stephan.doliov@gmail.com> wrote:
>
> I am curious what you mean by standard metrics format? I am all for standards-based but what are those in the case of
DBs.For environments where O11y matters a lot, I think the challenge lies in mapping specific query executions back to
systemcharacteristics. I am just thinking aloud as a newbie to this community. 


I was about to reply simply that the standard the open source world is
coalescing around is OpenMetrics. This is basically a codification of
the prometheus format with some extra features.

But on further thought I think what you're asking is whether there are
standard database metrics and standard names for them. A lot of this
work has already been done with pg_exporter but it is worth looking at
other database software and see if there are opportunities to
standardize metrics naming for across databases.

I don't really expect a lot of deep examples of that though. As soon
as you dig under the surface the differences quickly add up so you'll
have different labels with different semantics. But sure, it would be
nice if you could do simple high level queries like "number of active
connections" using the same metrics on mysql, postgres, and ...
whatever.


--
greg



pgsql-hackers by date:

Previous
From: Michael Paquier
Date:
Subject: Re: make tuplestore helper function
Next
From: Justin Pryzby
Date:
Subject: Re: make tuplestore helper function