PHD2 app crashed, exits without warning

[Christopher Boyd]

Hi All,

I've recently upgraded from old Win10 laptop to MeLe Quieter3.

PHD2 starts off fine but I get up in the morning to find PHD2 application has exited with no errors/warning. The problem has happened a few times, but is intermittent.

I've checked PHD2 guide and debug logs, as well as NINA log (attached) and there's nothing to suggest any error. I have turned off power saving setting for USB, updated to latest version of ASCOM and camera drivers, reverted back to prior versions of NINA and PHD2 that I was running on the laptop.

Windows Event Viewer logs an APCRASH (see attached) related to 'ntdll.dll'

I'm running:

-Mele Quieter3 Win 11

-PHD2 v2.6.10

-NINA v2.0.0.3001

-PegasusAstro Unity Platform v1.10.1906.57

-ASCOM 6.6 SP2 6.6.2.4195

Drivers

-ZWO ASCOM v6.5.19

-ZWO ASI Cameras v3.23

Gear:

-Pegasus Powerbox Advanced v2 (feeding power and data for everything, itself running off a 13.8v 10A power supply)

-AZEQ6

-2600MM

-ZWO 7x2"EFW (using the camera USB hub)

-ZWO EAF (using the camera USB hub)

-ZWO 174MM Guide Cam (directly into PPBA)

[Bruce Waddington]

You're running on a PHD2 release that's 2+ years old.  So the first thing to do would be to upgrade to the latest dev release and see what happens.

https://openphdguiding.org/development-snapshots/

Bruce

[Christopher Boyd]

Yes I was originally on the latest release and had this issue so reverted back to an older release that I have had success with.

[Jens Scheidtmann]

Hi Christopher, 

Can you then please share logs and event infos for the new version?

Thanks, Jens

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/11f1d45b-4b45-43d2-8f51-22ed556d8856n%40googlegroups.com.

[Bruce Waddington]

For these kinds of problems, we can avoid wasting time by following these guidelines:

1.  Run the 2.6.11dev6 release of PHD2

2.  Be sure you are running the latest version of the drivers for your camera

3.  When you upload log data, upload *all* the data, not just subsets.  To do that, use the built-in Update Logs feature in PHD2   https://openphdguiding.org/getting-help/

Bruce

[Jens Scheidtmann]

Hi Christopher,

Looking at the ntdll.dll exception code and googling it, it seems to be an "access violation" error, that can have a lot of different causes, from viruses to misbehaving application software.

So when trying to reproduce this with the current version of PHD2: 

1) Sometimes, you get some more infos printed, when starting the program from CMD. So start PHD2 from CMD, get it running and then start N.I.N.A., SGP or whatever you use to control your rig. Hopefully the PHD instance gets picked up. 

2) Start Sysinternals "Process Monitor" attach it to the PHD2 process and see if this gives more information on where the error is coming from (this may be a big file).

3) Please also supply the files that are referenced in the "Information" entry from the Windows Event log

Keeping fingers crossed. 

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/369def0a-056d-46d5-a0a3-059ae6c853edn%40googlegroups.com.

[Christopher Boyd]

I updated to 2.6.11dev6 last night, rerun guiding all night and the same error.

Logs have been uploaded here:

https://openphdguiding.org/logs/dl/PHD2_logs_qyWa.zip

(and yes, I updated to the latest camera drivers as well)

[Christopher Boyd]

full PHD2 Log files are here:

https://openphdguiding.org/logs/dl/PHD2_logs_qyWa.zip

(from both older and latest 2.6.11dev6 versions)

HOWEVER, I can clearly see from the PHD2 logs that the application closed at ~11:30pm but there is NOTHING in the Windows Event Viewer at this time AND, looks like the temp files from the original "Information" tab have been cleared. I did manage to grab the WER file that was generated (attached here).

[Bruce Waddington]

Hi Christopher, I'm sorry you're having this trouble. I have a suspicion about what's happening but I need a little more info.  What is the reason for the 2h20m gap between 20:05 and 22:25?  Was this a planned delay of some kind?  Would this same thing have happened in the previous guiding sessions where you had the problem?  One quick thing you could do to help me isolate the problem would be to switch your RA guide algorithm from PPEC to Hysteresis.  I'm not suggesting this is a solution but it will help me know if my area of suspicion is correct.

Thanks,

Bruce

[Christopher Boyd]

Thanks 
Bruce, yes - planned delay - set everything up and calibrate etc ~8pm, target doesn't rise until~10:30pm so NINA advanced sequencer waits until then to start.

I'll try changing to Hysteresis.

[Christopher Boyd]

I tried running Process Monitor Last night. but this time Procmon.exe crashed! 

PHD2 application remained open all night, EXCEPT I'm an idiot and somehow accidentally turned off the 'start guiding' step in my sequence, so it sat there doing nothing all night anyway.

Is it a coincidence that this Promon crash happened a similar time to the prior PHD2 crashes?

I wonder if there's something happening at that time in my sequence (slewing, exposure, refocus etc) that might be causing something unexpected - power draw, scope position causing cable wiggle and connection disruption etc?

Or maybe some other Windows event?

I'm going to trawl through the last few NINA logs to see what's happening around these times.

Thanks for all the help and suggestions so far!

[Bruce Waddington]

Well, this is certainly interesting.  The fact that you encounter the problem so frequently suggests to me it may not be a PHD2 problem.  It's almost like a timer goes off somewhere and the first task that starts with 'P' gets nuked. <lol>  The PPEC algorithm is sensitive to wall-clock time and perhaps Procmon is as well.  That's why I suggested trying Hysteresis which doesn't really care about wall-clock time.  I would have recommended looking at the power supply but it sounds like you are running off 120V AC to feed the Pegasus box.

Bruce

[Christopher Boyd]

Essentially yes, I'm in Australia so 240V from the house. but I'm running that through a regulated 13.8V PSU into Pegasus box.

I think I may have fixed the issue. There's a few things I did/changed that may have helped (however the caveat is that it's has been an intermittent issue so I may have just got luck last night).

I reviewed all of the PHD2, NINA and PPBA logs to see hat else was happening at the time when PHD2 application exited, and found the only thing that was consistent was that it occurred just after NINAs autofocus routine (set to run every 60 mins). Looking in my NINA schedule, I moved some of the 'start guiding' and 'autofocus' routines around just in case I was causing some conflict.

I also grabbed 2x new NTDLL.DLL files from a trusted source (fellow in the local asto group) who runs essentially the same setup as mine), and copied into Windows/System32 and Windows/SysWOW64 DLL folders. I then ran sfc /scannow and DISM.exe to register and repair any broken DLL files (ref https://www.youtube.com/watch?v=FI1no8bTAMM).

As suggested I also changed to hysteresis for guiding.

I'll continue to monitor and repot back whether I think I've confirm the fix.

Thanks for everyone's help!

Guiding ran fDism.exe