Closed Bug 1348071 Opened 7 years ago Closed 6 years ago

57 second response times from queue.taskcluster.net (or public-artifacts.taskcluster.net) last night

Categories

(Taskcluster :: Services, enhancement)

Product:
Taskcluster
Component:
Services
Type:
enhancement
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED INCOMPLETE

People

(Reporter: emorley, Unassigned)

References

Details

Last night at approx 2130-2230 UTC and then today at 0020-0130 UTC Treeherder saw spikes in response time fetching logs via queue.taskcluster.net (bug 1347945).

Treeherder uses the 'x-taskcluster-skip-cache: True' header, so this is either from slow responses from queue.taskcluster.net (which issues an HTTP 303), or else public-artifacts.taskcluster.net which is backed by Cloudfront/S3.

Unfortunately I can't distinguish between the redirect time and the main request time from our side on New Relic.

To find out which it is, would someone mind looking at the queue.taskcluster.net logs? (I'm presuming it's on the Taskcluster AWS account and not Heroku?)

Many thanks :-)
See Also: → 1348072
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → INCOMPLETE
Component: Queue → Services
You need to log in before you can comment on or make changes to this bug.