Bug 397332 - No-step scheduler job does a meridian flip and stucks
Summary: No-step scheduler job does a meridian flip and stucks
Status: RESOLVED FIXED
Alias: None
Product: kstars
Classification: Applications
Component: general (show other bugs)
Version: 2.9.7
Platform: Other Linux
: NOR normal
Target Milestone: ---
Assignee: Jasem Mutlaq
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2018-08-10 09:23 UTC by TallFurryMan
Modified: 2018-09-03 08:59 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:


Attachments
Session log - check the last activities to see the problem (1.40 MB, text/plain)
2018-08-10 09:23 UTC, TallFurryMan
Details

Note You need to log in before you can comment on or make changes to this bug.
Description TallFurryMan 2018-08-10 09:23:41 UTC
Created attachment 114389 [details]
Session log - check the last activities to see the problem

While looking for an issue with slewing proceeding while not required, I created a scheduler job with a sequence that by chance went near the meridian, and caused a meridian flip.

Because the scheduler job is not requiring tracking, the capture module should not request a meridian flip. We could add a warning for this when building the scheduler job.

However, when the Capture module executed the flip, the mount obliged but the flip timed out. The Capture module retried, but no timeout was configured and as the mount did not reply or something else went wrong that didn't gave the relevant notification, the Capture module hung there and the whole Scheduler+Capture got stuck.

Log attached.
Comment 1 TallFurryMan 2018-08-10 09:28:10 UTC
What I didn't say is that the mount was pointing at the north pole.
When requested to do a meridian flip, that is, just slew to the target again, it does nothing special as it is pointing at DEC=90.
That seems to be the root cause: the Capture module is unable to see that at DEC=90 whatever it requests the mount to do will end up with HA=0, thus will trigger a meridian flip.
That should be fairly easy to work around.