Bug #6086

[API] Creating new job sometimes takes too long

Added by Peter Amstutz almost 2 years ago. Updated about 1 month ago.

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

0%

Category:API
Target version:-
Story points1.0
Velocity based estimate-

Description

We have been seeing a lot of timeouts in the java sdk when it is attempting to create a job. The Java SDK uses a default read timeout of 20 seconds, so this suggests that sometimes it is taking longer than 20 seconds to complete the processing associated with creating a job prior to sending a response. In addition to the short term fix of increasing the Java sdk timeout in #6071, we should investigate why job creation is taking so long.

History

#1 Updated by Peter Amstutz almost 2 years ago

  • Description updated (diff)

#2 Updated by Brett Smith almost 2 years ago

  • Project changed from Arvados Private to Arvados
  • Category set to API
  • Target version changed from Bug Triage to 2015-06-10 sprint
  • Story points set to 1.0

#3 Updated by Tom Clegg almost 2 years ago

  • Assignee set to Tom Clegg

#4 Updated by Tom Clegg almost 2 years ago

  • Assignee deleted (Tom Clegg)

#5 Updated by Brett Smith almost 2 years ago

I am currently leaving this on the 2015-06-10 sprint. Currently we believe that #6071 will be enough of a workaround for customer pipelines, but if it's not, I want this handy to steal and do as part of #6068.

If it doesn't come up again, it's not expected to be dealt with this sprint.

#6 Updated by Brett Smith almost 2 years ago

  • Target version changed from 2015-06-10 sprint to Arvados Future Sprints

#7 Updated by Tom Clegg about 1 month ago

  • Status changed from New to Closed

#8 Updated by Tom Clegg about 1 month ago

  • Target version deleted (Arvados Future Sprints)

Also available in: Atom PDF