D12954: Fix side-effect causing distant jobs to abort.

Eric Dejouhanet noreply at phabricator.kde.org
Fri May 18 06:22:05 UTC 2018


TallFurryMan created this revision.
TallFurryMan added a reviewer: mutlaqja.
Restricted Application added a project: KDE Edu.
Restricted Application added a subscriber: kde-edu.
TallFurryMan requested review of this revision.

REVISION SUMMARY
  This change corrects the case where an ASAP job is scheduled to start at a later date/time because its score is not yet positive.
  With this change, only executeJob does switch the job state to JOB_BUSY, so that no other part of the code can unexpectedly start the job.
  
  Note that periodic score update is prepared in this commit to display negative scores properly, but score computation needs rework beforehand.
  
  But this does not solve the problem of duplicated repeated jobs where only the pipeline tracking activity is enabled, which has similar symptom.
  
  A new test vector, 'distant_jobs_no_twilight', has two north-hemisphere targets with opposite transit times.
  This allows testing a job immediately, and having the scheduler sleeping to wait for the next.

TEST PLAN
  Use 'distant_jobs_no_twilight' to test sleeping mode.
  Without the change, the first job completes, but the second job runs immediately, aborts, and does weird things.
  With the change, the first job completes, and the second job properly waits for its schedule.
  
  If using the telescope simulator, you may observe a problem with parking, which unexpectedly stops the scheduler, but lets the system in an incorrect state.
  
  Use all other test vectors to hunt for regressions.
  I have a weird issue with the capture tab on my setup and the RGBLRGB sequence: the CCD simulator has all captures stored as blue filter, will investigate.

REPOSITORY
  R321 KStars

BRANCH
  bugfix__distant_jobs (branched from master)

REVISION DETAIL
  https://phabricator.kde.org/D12954

AFFECTED FILES
  kstars/ekos/scheduler/scheduler.cpp

To: TallFurryMan, mutlaqja
Cc: kde-edu, narvaez, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20180518/9e542a41/attachment-0001.html>


More information about the kde-edu mailing list