Closed Bug 1486800 Opened 6 years ago Closed 6 years ago

maxRunTime is stolen by purgeCache service

Categories

(Taskcluster :: Workers, enhancement)

enhancement
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: pmoore, Assigned: pmoore)

Details

Attachments

(1 file)

Recently I discovered that when purgeCache service was down (due to other causes) the task initialisation process was delayed by around 15 minutes.

This was due to the worker using an exponential backoff algorithm to check for purge requests.

The maxRunTime was only 20 mins for the task, and the 4.5 minutes it had left to run the task were not enough, which caused the task to time out, and me to notice the problem!

The max run time should not include task environment setup time - it should be the time from when the first command gets executed.
Commits pushed to master at https://github.com/taskcluster/generic-worker

https://github.com/taskcluster/generic-worker/commit/43feda73a4f4c918bea7d307e33e418d1a43fe9c
Bug 1486800 - don't start the max run timer until the task initalisation (feature initialisation) has completed

https://github.com/taskcluster/generic-worker/commit/48279177841a520033acd24a7f7227e63adda0de
Merge pull request #125 from taskcluster/bug1486800

Bug 1486800 - don't start max run timer before task initialisation has completed.
r+ from bstack in github
Assignee: nobody → pmoore
Status: NEW → ASSIGNED
Released in generic-worker 11.0.1
Status: ASSIGNED → RESOLVED
Closed: 6 years ago
Resolution: --- → FIXED
Component: Generic-Worker → Workers
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: