Parking from recovery mode

I got the same type of UI hang again tonight.

I was running a sequence and a sub was in progress. I wanted to abort the frame so I could run AF. I clicked Pause in the sequencer window, and got prompted to wait for current frame or abort immediately. I selected abort immediately (leaving Run end of sequence actions un-checked.) At that point the button on the sequence window changed to Aborting… but the status bar stayed at “Integrating ; Event 1; Frame 6 600s (594s)”.

At that point the UI was completely unresponsive (could not drag any window or click anything). I waited a while then killed the Application using Task Manager.

When Task Manager tried to close SGP, SGP put up the dialog “SGP still has gear connected. Are you sure you want to exit?”. I clicked No, and the dialog went away but the SGP UI went back to being stuck. I used Task Manager again to kill SGP, got the same dialog, clicked Yes, and SGP closed.

Log file

Andy

FYI I happened again. This time I just did Pause, keep the current frame, end of sequence options un-checked.

So apparently it just needs the pause to trigger the hang, neither the abort or the end of sequence options are needed.

I have the log if you need another one…

Andy

I hope this gets sorted - Looks like it’s happening to a few people …

1 Like

Same for me as well, here is a link to the log files.

I have had issues with the last to updates regarding stability. Thanks for looking into it.

Rodney

1 Like

I also have had many occurrences of the following issue with at least releases 24 and 25: UI hangs when I pause the sequence then select Abort. Run End of Sequence is unchecked.

I’ve just experienced the same issue - pausing to abort and SGP completely froze. Release 25 and PHD2dev09.

I aborted due to condensation within reducer (problem plaguing me at the moment) and will collect log once I’ve demisted!

Logs:

https://dl.dropboxusercontent.com/u/32050952/PHD2_DebugLog_2016-01-07_191745.txt
https://dl.dropboxusercontent.com/u/32050952/sg_logfile_20160107191543.txt

Admittedly, I am at a loss. These are not all the same issues. There are at least 2 issues here:

  1. Abort seems fine, but Park never gives control back to the calling function.
  2. Some other thing that, by all accounts in the logs, looks like everything worked just fine

I have been looking at this for hours… going on vacation and will try to pick it up again in a week or so. In the meantime, if anyone is easily able to reproduce this behavior (I have tried everything I can think of and the sequence aborts exactly like it should…) using simulator gear (that way we are guaranteed to have the same set of gear), I would be immensely grateful.

1 Like

Ken,

I tried to repro the problem in with simulators but could not. It sounds like Paul was able to repro it with simulators though as he reported in another thread.

Andy

Andy,
Strictly speaking, reproducing a situation like this is a deliberate exercise not an accidental occurrence like I had, hopefully there is something in my uploaded files which will give Ken a eureka moment. If not then I will be putting in some time to deliberately produce this problem in the hope that I will be able to reproduce it time after time and offer something solid for Ken to explore.

Paul

1 Like

Thanks, Paul! That would be great if you could repro it. It repro’d for me every time I hit pause with my imaging system regardless of whether I selected abort immediately/wait for current frame/run end of sequence options, but it did not repro for me when I tried with with simulators.
Andy

1 Like

ALERT ! ! !

Andy,
Pop over to my thread…I can confirm reproducibility of the failed abort issue and full instructions are there to do it yourself along with a sample SGP file to use and the logs from SGP and PHD2 for the reproduced failed abort

Fingers crossed eh ?

Paul

1 Like

Paul, Thanks for coming up with that repro! Hopefully it will allow Ken and Jared to fix the problem.
Andy

Hi all,

Was trying out things tonight and the exact same issue just happened. The sequence all went very well, at the end it runs the “end of sequence” actions. The mount is parked “physically parked, also shown “parked” in The Sky X”) but SGP stuck at “parking scope” status. Had to kill SGP in task manager.

Here is the SGP log

In the very bottom of the log you can find the error about parking and getting position of the scope etc. One thing weird I noticed is that once the SGP stuck at “parking scope”, I went to The Sky X and noticed the mount is actually “disconnected”. I then manually connected the mount back and found it’s parked, but apparently SGP somehow disconnected the mount and didn’t figure out it’s already parked.

Using Paramount MyT in my case so it has to be controlled via The Sky X (or by two software in a sense?), not sure if this is a problem.

Any fix from SGP yet? Or any user who encounter the same issue has found out the problem?

Thanks,
Yizhou

All these problems that seem to involve interaction between SGP and some device will almost certainly be helped if the logs from the device driver are provided. They will help to show what is happening from the driver’s perspective.

Another log source is the ASCOM DriverAccess logs. These are generated by the shim that provides the COM drivers with a .NET front end and are independent of both the driver and SGP. They will be present even if there is no logging in the driver.

Hummm… Tried again two hours later and it works fine this time… It proceeded fine, passed “parking scope” and show “sequence completed” or something like that at the end. Didn’t change anything…

Here is the log:

Parking happens around 5:28.

After that there are still a bunch of error code in debug log/ASCOM log saying ASCOM Telescope: Error in GetCurrentPos : GetAndFixRaDec. Can anyone explains what that means and is there something wrong?

Thanks,
Yizhou

TheSky disconnects when its park command completes. After that all attempts to read the scope properties will fail with a not connected error. This is what you are seeing.

There’s nothing wrong. The mount is disconnected so trying to read things from the mount will fail.

This zip contains my final logs from last night (2.5.0.6).

I expected the usual hang while parking scope, but the final details were unusual.

Events:

  1. sequence running normally until clouds came in about 5:45am, SGP now in recovery mode. Sequence set to terminate at 6:20.
  2. I abort recovery mode at 6:17:14am.
  3. Normal end of sequence commands are sent.
  4. Gemini2 stops tracking but mount and dome stop moving and stay stopped.
  5. Weird interactions going on with PHD2.
  6. Centering process starts up. (very strange)
  7. SGP eventually stops with Run Completed. UI remains ok entire time.
  8. I manually disconnect all equipment (which all still shows as connected) and close SGP.

@Chris, where do I find the ASCOM DriverAccess logs? There are none in my ASCOM logs directory.

Hey Guy’s,
Just passed a LOG to Ken for his long flight - Hopefully this is the big one for the end of sequence freeze in recovery mode !

:open_mouth:

Keep your fingers crossed

Regards
Paul

SGPro 2.5.0.7 beta is released. I am still unable to produce any hang situation so I am still reliant on folks running SGPro to this failure and sharing the logs. Thx for all the help.