Export metrics every 10 minutes instead of every minute #2260
Conversation
This file contains hidden or 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
zackattack01
approved these changes
May 15, 2025
James-Pickett
approved these changes
May 15, 2025
| @@ -345,7 +345,7 @@ func (t *TelemetryExporter) setNewGlobalMeterProvider(launcherResource *resource | |||
| // Create new meter provider and let otel set it globally | |||
| newMeterProvider := sdkmetric.NewMeterProvider( | |||
| sdkmetric.WithResource(launcherResource), | |||
| sdkmetric.WithReader(sdkmetric.NewPeriodicReader(metricsExporter)), | |||
| sdkmetric.WithReader(sdkmetric.NewPeriodicReader(metricsExporter, sdkmetric.WithInterval(10*time.Minute))), | |||
There was a problem hiding this comment.
Is this the kind of thing we could adjust at runtime, or is it baked too deeply?
There was a problem hiding this comment.
Sorry, just seeing this question now! We can't adjust this after provider creation. However, we have the same issue with the tracer provider (e.g. when we want to adjust the sampling rate), and we address it there by creating a new tracer provider anytime the relevant tracing flags change. We could do the same thing here -- make this into a flag, then on flag change, create a new global meter provider with the updated interval -- if desired.
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Outdated suggestions cannot be applied.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.
Our metrics exporter by default exports data points every minute. I had assumed that if no measurement had occurred during that minute, there would be nothing to export, so the export wouldn't happen. However, that appears to be incorrect -- the exporter will re-ship all meters with at least one data point every minute. I believe this behavior indicates that the latest data measurement is "unbroken", which seems reasonable and useful, but I also believe this is contributing to our higher metrics storage cost.
This PR changes the export interval to every 10 minutes instead of every minute. (I think we could comfortably bump this up to every 30 minutes or possibly every hour, but I'd like to confirm that this actually affects our billing data before making that change.) All metrics not yet exported will always be flushed and exported on launcher shutdown, so we don't risk losing data with this greater interval -- metrics will just take longer to show up. This seems like an acceptable tradeoff to me.