Bug #3999

[Workbench] Pipeline timing number oddities

Added by Ward Vandewege almost 3 years ago. Updated 7 months ago.

Status:ClosedStart date:
Priority:NormalDue date:
Assignee:-% Done:

0%

Category:-
Target version:-
Story points0.5
Velocity based estimate-

Description

On qr1hi-d1hrv-21wrj86xjglhryh:

The first job is missing the second timing number "sum of all job wall clocks". But if you click through to the job it is listed just fine.

In the second job, the finished_at time is 12:26. But the last log entry is 12:15. What's up with that?

In the first and second job, the factor seems off (too low); the job wall clock number seems more or less accurate but the overall wall clock number seems too high (I guess that's the same bug as the previous point).

On 9tee4-d1hrv-2w460wv7g529l3l:

===========================
This pipeline started at 10:18 AM 9/25/2014. It completed in 19 hours 25 minutes 25 seconds at 10:19 AM 9/25/2014.

This pipeline ran for 19 hours 25 minutes 25 seconds and used 7 days 12 hours 51 minutes 18 seconds of CPU time (9.3⨯ scaling). ===========================
On the dashboard, the pipeline is shown (correctly) with a 1 minute runtime.

History

#1 Updated by Ward Vandewege almost 3 years ago

  • Story points set to 0.5

#2 Updated by Ward Vandewege almost 3 years ago

  • Description updated (diff)

#3 Updated by Ward Vandewege almost 3 years ago

  • Subject changed from [Workbench] Pipeline timing number oddities on qr1hi-d1hrv-21wrj86xjglhryh to [Workbench] Pipeline timing number oddities
  • Description updated (diff)

#4 Updated by Tom Clegg 7 months ago

  • Status changed from New to Closed

#5 Updated by Tom Clegg 7 months ago

  • Target version deleted (Arvados Future Sprints)

Also available in: Atom PDF