Bug 464775 - Remote slewing and plate-solving target update problem
Summary: Remote slewing and plate-solving target update problem
Status: REPORTED
Alias: None
Product: kstars
Classification: Applications
Component: general (show other bugs)
Version: 3.6.2
Platform: Other Linux
: NOR normal
Target Milestone: ---
Assignee: Wolfgang Reissenberger
URL: https://invent.kde.org/education/ksta...
Keywords:
Depends on:
Blocks:
 
Reported: 2023-01-25 00:37 UTC by Philipp Auersperg-Castell
Modified: 2023-01-29 15:48 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Philipp Auersperg-Castell 2023-01-25 00:37:49 UTC
SUMMARY
***
Remote slewing and plate-solving target update problem:

when slewing with remote access(tested with kstars and stellarium as remote clients),
there occurred 2 problems:

- Name of target is not displayed in the Capture view.
- target coords are not forwarded to Align, so plate solving gives
 not the delta from position to desired position but instead the delta
 between park position and slew position, so 'slew to target' does not work.
***

 This bug has been introduced in version 3.5.6.


STEPS TO REPRODUCE

1. run kstars/ekos (3.6.x) and simulator telescope and ccd, GSC installed for plate solving simulation.
  - run ekos and hit 'start'.
  - telescope mount in park position.
2. unpark the mount.
3. on another machine start kstars and connect via Tools/Devices/Device Manager...
4.  on client machine slew the telescope to M81
   - either by selecting it in the map and rightclick and select in context menu 
       'Telescope Simulator/Slew'.
   - or by clicking ctlr-F and finding M81 and clicking "Details..." and then 
       "Center in Telescope".
5. Telescope as expected points to M81.

See logs at the end of this report!

OBSERVED RESULT
- In Ekos/Capture view the text box labeled 'Target:' is _not_ updated.
- in the 'Align' tab when clicking 'Capture&Solve' with the 'Solver Action' set to 'sync':
    the solver finds the position but gives a wrong error like: 
    "673084 arcsec. RA:541515 DE:-399754".


EXPECTED RESULT
- in the Capture view the 'Target' text box shows "M_81".
- 'Capture&Solve' should gives an error in the range of <60 arcsec.

when doing the above-mentioned steps from the local kstars map, everything
works fine as described in EXPECTED RESULT.

PROPOSED FIX
I have provided a merge request that produces the expected result:
 https://invent.kde.org/education/kstars/-/merge_requests/834  .


SOFTWARE/OS VERSIONS
Tested on Ubuntu/amd64 and Raspberry arm64.
This bug does not seem to be a platform-specific problem.

ADDITIONAL INFORMATION

Log with broken align:
2023-01-25T01:08:55 Mount is synced to solution coordinates.
2023-01-25T01:08:55 Syncing to RA (09h 57m 41s) DEC ( 68° 57' 26")
2023-01-25T01:08:55 Solution coordinates: RA (09h 57m 41s) DEC ( 68° 57' 26") Telescope Coordinates: RA (09h 57m 43s) DEC ( 68° 57' 15") Target Coordinates: RA (23h 56m 00s) DEC ( 180° 00' 00")
2023-01-25T01:08:55 WCS information updated. Images captured from this point forward shall have valid WCS.
2023-01-25T01:08:55 Effective telescope focal length is updated to 3 131,54 mm.
2023-01-25T01:08:55 Solver RA (148.94092) DEC (69.06839) Orientation (90.00727) Pixel Scale (0.24766) Parity (neg)
2023-01-25T01:08:55 Solver completed after 0.20 seconds.
2023-01-25T01:08:54 Image received.
2023-01-25T01:08:48 Capturing image...
2023-01-25T01:07:52 World Coordinate System (WCS) is enabled.


Log with correct align:
2023-01-25T01:21:12 Mount is synced to solution coordinates.
2023-01-25T01:21:12 Syncing to RA (09h 57m 42s) DEC ( 68° 57' 26")
2023-01-25T01:21:12 Solution coordinates: RA (09h 57m 42s) DEC ( 68° 57' 26") Telescope Coordinates: RA (09h 57m 44s) DEC ( 68° 57' 15") Target Coordinates: RA (09h 57m 44s) DEC ( 68° 57' 15")
2023-01-25T01:21:12 WCS information updated. Images captured from this point forward shall have valid WCS.
2023-01-25T01:21:12 Effective telescope focal length is updated to 3 133,03 mm.
2023-01-25T01:21:12 Solver RA (148.94257) DEC (69.06839) Orientation (90.01308) Pixel Scale (0.24754) Parity (neg)
2023-01-25T01:21:12 Solver completed after 0.23 seconds.
2023-01-25T01:21:11 Image received.
2023-01-25T01:21:05 Capturing image...
2023-01-25T01:18:02 World Coordinate System (WCS) is enabled.
Comment 1 Wolfgang Reissenberger 2023-01-29 15:48:05 UTC
There is no stable way to reproduce this issue. I can confirm, I could observe it and that there were situations when the target was not set. But after recompiling everything it *seems* to work - no matter whether the INDI server is local or remote.

It looks like a concurrency issue...