Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Telemetry Alert for GC_SLOW_PHASE on 2019-11-01
27 views
Skip to first unread message
telemetr...@mozilla.com
Nov 4, 2019, 12:33:06 AM11/4/19
to dev-telemet...@mozilla.org, dev-teleme...@lists.mozilla.org
We have detected a change in the Telemetry probe GC_SLOW_PHASE in Firefox Nightly builds from 2019-11-01.
Alert details: http://alerts.telemetry.mozilla.org/index.html#/detectors/1/metrics/1358/alerts/?from=2019-11-01&to=2019-11-01
Changes new to Nightly builds on 2019-11-01: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=4a19c3f74d0cf11a6a2dc143b3b43a88a96f0932&tochange=518df4329a20f76b3bffd5f6201b007fdebc274f
The value of GC_SLOW_PHASE over time: https://telemetry.mozilla.org/new-pipeline/evo.html#!measure=GC_SLOW_PHASE
What to do about this:
1. File a bug to track your investigation. You can just copy this email into the bug Description to get you started.
2. Reply-All to this email to let the list know that you are investigating. Include the bug number so we can help out.
3. Triage the alert. You can find instructions here: https://docs.telemetry.mozilla.org/tools/alerts.html#triaging-a-telemetry-alert-email
If you have any problems, please ask for help on the #telemetry IRC channel or on Slack in #fx-metrics. We'll give you a hand.
What this is:
We have a system called cerberus[1] that compares Telemetry collected on different Nightly builds and looks for sudden changed in value distributions using the Bhattacharyya Distance[2]. It found such a change in GC_SLOW_PHASE on 2019-11-01 so it asked its buddy medusa[3] to send this email to the dev-telemetry-alerts mailing list and to all email addresses listed in the alert_emails field of GC_SLOW_PHASE's definition.
You can do this!
Your Friendly, Neighbourhood Firefox Telemetry Team
[1]: https://github.com/mozilla/cerberus
[2]: https://en.wikipedia.org/wiki/Bhattacharyya_distance
[3]: https://github.com/mozilla/medusa
Alert details: http://alerts.telemetry.mozilla.org/index.html#/detectors/1/metrics/1358/alerts/?from=2019-11-01&to=2019-11-01
Changes new to Nightly builds on 2019-11-01: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=4a19c3f74d0cf11a6a2dc143b3b43a88a96f0932&tochange=518df4329a20f76b3bffd5f6201b007fdebc274f
The value of GC_SLOW_PHASE over time: https://telemetry.mozilla.org/new-pipeline/evo.html#!measure=GC_SLOW_PHASE
What to do about this:
1. File a bug to track your investigation. You can just copy this email into the bug Description to get you started.
2. Reply-All to this email to let the list know that you are investigating. Include the bug number so we can help out.
3. Triage the alert. You can find instructions here: https://docs.telemetry.mozilla.org/tools/alerts.html#triaging-a-telemetry-alert-email
If you have any problems, please ask for help on the #telemetry IRC channel or on Slack in #fx-metrics. We'll give you a hand.
What this is:
We have a system called cerberus[1] that compares Telemetry collected on different Nightly builds and looks for sudden changed in value distributions using the Bhattacharyya Distance[2]. It found such a change in GC_SLOW_PHASE on 2019-11-01 so it asked its buddy medusa[3] to send this email to the dev-telemetry-alerts mailing list and to all email addresses listed in the alert_emails field of GC_SLOW_PHASE's definition.
You can do this!
Your Friendly, Neighbourhood Firefox Telemetry Team
[1]: https://github.com/mozilla/cerberus
[2]: https://en.wikipedia.org/wiki/Bhattacharyya_distance
[3]: https://github.com/mozilla/medusa
Jon Coppeard
Nov 4, 2019, 6:01:05 AM11/4/19
to telemetr...@mozilla.com, dev-telemet...@mozilla.org, dev-teleme...@lists.mozilla.org
This is due to a bug fixes related which landed recently and is
expected. We now see fewer slow phases due to waiting for parallel
tasks to complete.
https://bugzilla.mozilla.org/show_bug.cgi?id=1592537
Jon
> _______________________________________________
> dev-telemetry-gc-alerts mailing list
> dev-telemet...@mozilla.org
> https://mail.mozilla.org/listinfo/dev-telemetry-gc-alerts
expected. We now see fewer slow phases due to waiting for parallel
tasks to complete.
https://bugzilla.mozilla.org/show_bug.cgi?id=1592537
Jon
> dev-telemetry-gc-alerts mailing list
> dev-telemet...@mozilla.org
> https://mail.mozilla.org/listinfo/dev-telemetry-gc-alerts
0 new messages