Proposal: Telemetry event instead of predefined log message on batch timeout #1347
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
katafrakt
commented
Nov 18, 2024
katafrakt
force-pushed
the
telemetry-for-batch-timeout
branch
from
f2ced2e to
b5de3b5
Compare
|
@katafrakt this is great! Can we add an item to the changelog with an example handler that would create the same sort of log message we currently do for anyone who wants to maintain backwards compatibility? |
|
@benwilson512 sure, I added something |
benwilson512
approved these changes
Nov 19, 2024
Instead of producing an eritrary log message with an arbitrary level, emit a telemetry event [:absinthe, :middleware, :batch, :timeout], so everyone can subscribe to it and do whatevet they want (including logging the same message as was here before).
katafrakt
force-pushed
the
telemetry-for-batch-timeout
branch
from
375b8ab to
8898d57
Compare
|
Reverted changes in typespec, as they were causing failures on the CI |
|
Thank you! |
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Outdated suggestions cannot be applied.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.
I really love the work done about batch timeouts in #1298, with one exception I noted in a comment there. On a timeout, Absinthe produces a predefined log message on error level. This goes against my thinking that libraries generally should not produce logs on runtime (only on startup or maintenance actions, such as running mix tasks), especially when there's a better option available.
This PR suggests adding a new telemetry event
[:absinthe, :middleware, :batch, :timeout]instead, so everyone can attach to it and do whatever they want (including logging the same message on the same level).More context
I was about to bump Absinthe at our company project, but the approach with the log and not telemetry caused at least three hold-backs:
inspect(batch_fun)is large.error-level log message, but I'm not sure we want to trigger it in case of the timeout. Sowarninglevel, at least for evaluation period, would work much better here.Other considerations
This is technically a breaking change, in case someone already relies on these log messages. But I think the capability was not advertised in changelog or anywhere, so perhaps it's safe to change. If not, perhaps we can at least give a configuration option to choose between the log message and telemetry.