David's Astronomy Pages
Notes - Session 612.2  (2018-06-08, Observatory Commissioning)

   
Bullet Session Aims & Highlights
   
Bullet Operational Issues
  - Critical Issues (0),  Major Issues (2),  Minor Issues (0),  Small Defects (0),  Continuous Improvement (0)
   
Bullet Observing Plan
   
Bullet No images from this session  >>
   
2018-06-09
Bullet CCDApp2 Freezing - Investigation
 

Session Aims & Highlights (2018-06-08)

Main Aims

This was a daytime session to continue formal commissioning of the new Dome Observatory including :

  1. System Stability. Check stability of POTH.Hub (Telescope + Dome) & Observatory Control Software following major issues during the nighttime session on 2018-06-06
  2. Observing Plan. Test automated acquisition of images of a sequence of targets (Plan) that have been scheduled by a newly developed genetic algorithum optimisation program working with target data held in SQLite (test to the limits possible with a daytime test with no actual stars)

Equipment & Software

Highlights

Notes

Summary Plots & Logs

Observing Plan Executed as Day Time Test, 2018-06-08
(Note: Failed targets are due to them being below the horizon at the time of the test)
 
Image

Back to Top


Operational Issues (2018-06-08)

[ Prev | Next ]

Critical Issues

Major Issues

Minor Issues

Small Defects

Continuous Improvement

[ Prev | Next ]

 

Back to Top


Observing Plan  (2018-06-08)

POTH/system instability during the last session had prevented the intended full testing of automated acquisition of a sequence of targets (Job Queue) that have been scheduled as an 'Observing Plan' by a newly developed genetic algorithum optimisation program working with target data held in SQLite.

To try to keep commissioning on track a daytime test was considered to be the next best thing. This was undertaken under near full test conditions (i.e using live scope slewing, slaved dome rotation and CCD camera imaging,  but with the dome shutter closed and without high precision target centering using plate solution of locating images (obviously no stars in daytime !). 

The test executed the Observing Plan that had been previously created for the 2018-06-06 session.   Being run during the day time, several of the targets intended for viewing at night, lay close to the horizon/below the working airmass limit or totally lay below the horizon at the time of the test. These targets flagged up as 'failed', but as a test they successfully showed that the executor was doing its job by only allowing slews to 'valid' targets.

First attempt at the test failed when CCDApp2 froze/locked-up when returning focus to the application after going to another application.   A second attempt succeeded when CCDApp2 was kept as the top window throughout the test. This problem is under investigation.

Observing Plan about to be executed as Daytime Test (screenshot)
Image
 
Observing Plan Completed (screenshot)
(Note: Some targets have failed due to them being close to or below the horizon at the time of the test)
Image
 
Observing Plan Executed as Day Time Test
(Note: Failed targets are due to them being close to or below the horizon at the time of the test)
 
Image

Back to Top


CCDApp2 Freezing - Investigation  (2018-06-09)

The Observatory's control program (CCDApp2, later renamed as AstroMain  has developed a tendancy to freeze/lock up when executing a Job Queue, when it's not the top window and an attempt is made to take back focus and make it the top window again.  This is a problem that has been seen during indoors testing with Simulator Scope & Camera etc and has only appeared since making software changes in 2018 to enable Dome Operations and accomodate new Job Queue routines.   After freeze/lock-up occurs the only way out is to kill the CCDApp2 process, but this results in instability in POTH.Hub with resulting problems with scope and dome connection.

It was thought that the problem most likely relates to some illegal Threaded Operation with interference to GUI operation involving  window repainting. Precise cause of the problem and solution is difficult to track down.

A series of indoor tests were performed in order to try to understand the circumstances under which CCDApp2 program freezes whilst executing a Job Queue and the circumstances under which it doesn't.

Test    Description Result
1   CCDApp2 kept at Top Window throughout
 (2018-06-08 Test)
  Job Queue executes to completion
2   Focus repeatedly moved to other application windows and back again to  CCDApp2 whilst executing Job Queue
(with Queue Tab open)
  Job Queue proceeds ok through first 12 targets whilst switching between applications (after clicking back on CCDApp2 there is sometimes there is a pause until CCDapp2 window comes to the top with focus). Whilst on the 13th target CCDApp2 froze completely
(window shows 'Not Responding').
3   Focus repeatedly moved to other application windows and back again to  CCDApp2 whilst executing Job Queue
(with another Tab, other than the Queue2 tab, open)
  Job Queue executes to completion
4   Like 3 but CCDApp2 run in VS2017 Debug mode, with breakpoint in thread   Job Queue hangs after clicking continue in VS2017 after control had passes to the breakpoint in VS2017 Normally this would return control to the running CCDApp2 application, but in this case the UI just freezes.
5   Like 2 but CCDApp2 run in VS2017 Debug mode   Job Queue hangs almost immediately one the first attempt to return to CCDApp2.
6   Repeat of 2   Job Queue executes to completion ' Responding').
         

Back to Top