Metrics: Wire up objstore_bucket_operation_duration_seconds in iter operations
#79
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
gotjosh
commented
Sep 26, 2023
| @@ -27,6 +27,7 @@ require ( | |||
| golang.org/x/oauth2 v0.4.0 | |||
| golang.org/x/sync v0.1.0 | |||
| google.golang.org/api v0.103.0 | |||
| google.golang.org/grpc v1.53.0 | |||
There was a problem hiding this comment.
Unsure if this was forgotten or specific to my setup - happy to remove it if undesired.
There was a problem hiding this comment.
Looks like you are correct. This might be a miss on previous prs
…r` operations Trying to get some statistics in Mimir about the time we're spending in our `iter` operations I realised that despite these operations showing up as part of the labels their value was always 0. Soon enough and with the help of Peter, I realised that these were not wired up in `objstore.go`. This PR does exactly that and measures the duration of the whole `iter` operation. There's one big caveat though, `iter` operations take a callback as an argument which is executed per each item returned. The computed duration will include whatever time we spend on each of these callbacks which might an undesirable effect for users of the library. Sadly, I can't get around this without changing the interface for `iter` to (e.g.) return JUST the time the network operation took. In the meantime, I've done the best I've could without incurring in a big change and documented the behaviour as part of the Histogram's HELP. Co-Authored-By: Peter Štibraný <895919+pstibrany@users.noreply.github.com> Signed-off-by: gotjosh <josue.abreu@gmail.com>
Signed-off-by: gotjosh <josue.abreu@gmail.com>
Signed-off-by: gotjosh <josue.abreu@gmail.com>
This was referenced Sep 26, 2023
yeya24
approved these changes
Oct 20, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Trying to get some statistics in Mimir about the time we're spending in our
iteroperations I realised that despite these operations showing up as part of the labels their value was always 0.Soon enough and with the help of Peter, I realised that these were not wired up in
objstore.go.This PR does exactly that and measures the duration of the whole
iteroperation. There's one big caveat though,iteroperations take a callback as an argument which is executed per each item returned. The computed duration will include whatever time we spend on each of these callbacks which might an undesirable effect for users of the library.Sadly, I can't get around this without changing the interface for
iterto (e.g.) return JUST the time the network operation took. In the meantime, I've done the best I've could without incurring in a big change and documented the behaviour as part of the Histogram's HELP.Verification
I've used a test which I then deleted because the histogram is based on time and we don't seem to mock time so it's not worth committing.