×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Mac KStars 3.5.7 beta new build -- please test

  • Posts: 26
  • Thank you received: 1
Do we know when the code was pulled from indi-3rdparty to compile this version?
2 years 2 months ago #79535

Please Log in or Create an account to join the conversation.

  • Posts: 593
  • Thank you received: 274
Hi Rob,

I had a bit of clear sky this evening so I tried 3.5.7 (original beta). I run Kstars on my Mac and Indi on a Rpi via Astroberry.
1. All devices connect OK (2 ZWO cameras, ASI EFW, ASI EAF, EQMod Mount, Pegasus PPBA).
2. Polar Alignment worked fine (noted the MF disabled during the PA).
3. Auto-focusing worked as usual.
4. Alignment / Plate solving worked as usual.
5. Scheduler. Loaded a previous sequence but on running, it went to the idle state and got stuck there. All the Job Constraints were on their default values and none should have prevented the sequence from starting. I rebuilt the sequence and tried twice more - same issue. Looking in the log I can see some msgs that might help... "Scheduler failed to register with dbus", "scheduler is stopping..."
Hi Rob,

I had a bit of clear sky this evening so I tried 3.5.7 (original beta). I run Kstars on my Mac and Indi on a Rpi via Astroberry.
1. All devices connect OK (2 ZWO cameras, ASI EFW, ASI EAF, EQMod Mount, Pegasus PPBA).
2. Polar Alignment worked fine (noted the MF disabled during the PA).
3. Auto-focusing worked as usual.
4. Alignment / Plate solving worked as usual.
5. Scheduler. Loaded a previous sequence but on running, it went to the idle state and got stuck there. All the Job Constraints were on their default values and none should have prevented the sequence from starting. I rebuilt the sequence and tried twice more - same issue. Looking in the log I can see some msgs that might help... "Scheduler failed to register with dbus", "scheduler is stopping..."
  

File Attachment:

File Name: log_19-03-27.txt
File Size:406 KB
2 years 2 months ago #79536
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
David, the indi-3rdParty code should have been pulled today since my build script uses the most recent master of INDI and KStars as the default.
2 years 2 months ago #79537

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
John, the Scheduler is heavily dependent on dbus, so if dbus is not working, then that would make the scheduler not work properly.  So dbus was not working properly for you?  In this build, I am trying to use the default craft build of dbus instead of using the custom recipe that I made a few years ago.  So if dbus does not work, we may have to bring back my old recipe.
2 years 2 months ago #79539

Please Log in or Create an account to join the conversation.

  • Posts: 26
  • Thank you received: 1
Hi Peter,

Can you send/post a log file for the dome moving across 0 degrees when slaving and I'll have a look.
I do have the Autosync threshold set to 2 degrees in Ekos | Nexdome | Slaving (not 0.5).
Meridian side is Mount.

David
2 years 2 months ago #79540

Please Log in or Create an account to join the conversation.

  • Posts: 992
  • Thank you received: 155
Thanks Dave! I changed the slaving to 2 degrees vs 0.5 and that seemed to see the problem!
I will test over the coming nights and see how it goes. Thanks again for the tip!
2 years 2 months ago #79546

Please Log in or Create an account to join the conversation.

  • Posts: 26
  • Thank you received: 1
Great, I think the Ekos slaving value just needs to be greater than the Nexdome internal dead zone. So if you need more rotation accuracy you can try reducing it to 1.0, or even 0.6. Hope it fixes the issue for you!
2 years 2 months ago #79547

Please Log in or Create an account to join the conversation.

  • Posts: 2
  • Thank you received: 0
Hi Rob,

I tested with KStars-3.5.7-beta3 (and also beta2). I skipped KStars-3.5.7-beta because I read it had some problems with the Indi drivers. I have only focused on my own problem: starting a remote session from my Mac to the RPi with StellarMateOS on it (drivers used: Skywatcher AltAz {no problems}, SkySafari {no problems}, ZWO ASI CCD {crashes}).

I first started my devices and Ekos on the RPi. Everything works well.

As far as I can see my problem with 3.5.6 has not been resolved. The remote device/driver (indi_asi_ccd) still crashes and causes (or is the result of) the local macOS version of KStars/Ekos to crash. I really don't know what causes what, actually. But considering my last tests, I think it is the macOS version.
What is different though, is the fact that the remote device/driver (indi_asi_ccd) does NOT restart as in 3.5.5 and 3.5.6. The driver crashes and the tab in the remote Indi manager just disappears. When I start KStars/Ekos after some time on my Mac and try to connect remotely it only finds my mount (and the SkySafari) driver, since the ZWO ASI driver has died.

Additionally I tested 3.5.7.beta2 with a profile for a local setup, connecting just the ZWO ASI178MC directly to my Mac and using a telescope simulator.
In this test the indi_asi_ccd just kept on crashing, continuously trying a restart. Using ASI Studio on the Mac the camera works just fine.

I also tested with 3.5.7.beta3 locally. KStars/Ekos crashed again. I restarted and connected to the still existing instance of Indi server. The tab for the ZWO ASI CCD had disappeared. I'm on macOS Monterey and attached the macOS crash log. 

File Attachment:

File Name: log_crash_kstars.rtf
File Size:47 KB


If you need any additional info from me please let me know.

Cheers,

Marc
2 years 2 months ago #79613
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 6
  • Thank you received: 0
Hi,

I tested last night beta3. With previous version (3.5.6) I had serial problems and KStars was very unstable and indidriver can crash anytime. I have MacBook Pro early 2011, Onstep 4.12 and ASI224MC.

This beta-version seems to be stable and scheduler ran whole night without problems. Only problem was connecting to indiserver. KStars crashed almost every time, when I tried to connect equipment. If I let indiserver running, connecting seems to work but starting indiserver seems crash Kstars.

Then there is something strange with Fitsviewer. I have disabled fitsviewer from settings because I don't want to see every single image using scheduler. In the version 3.5.6 I can still see the image using capture but that not happen with beta-version. I have to enable Fitsviewer to see preview image and the disable it before using scheduler.

Then there may be some issue with focus/polar alignment and with auto stretch. It seems that auto stretch is not enabled with those functions because my stars were dimmer than normally with same settings. But I'm not really sure about that.
2 years 2 months ago #79614

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Hi JAMBBE,

Yes, with the first two betas that I posted there were packaging problems because of all the changes I had to make, my packaging script had some bugs.  The testers in this forum thread really helped me sort out the issues though with their testing, so I am most appreciative.  It should be good now, thanks for trying it out and reporting that it worked well.

There is a bug (I mentioned it earlier in the thread) that I see sometimes when starting an INDIServer where KStars sometimes but not always crashes and I don't know the cause.  This issue has existed for a few months and is not at all related to the INDI drivers issues that were reported with the betas above.  The INDIServer still starts successfully however, and if you restart kstars and reconnect to the INDIserver you already started everything works fine.  I wish I knew why this happens but I haven't figured it out yet.  It does seem that it might be dependent on Mac versions, since it happens on my Monterey Mac, but not on a Big Sur Mac that I tested it on the other day.  Now that a couple of more people can look into it with the work I did the last two weeks maybe we can investigate this further.

The other things you are mentioning with FITSViewer and and auto stretch are probably things that got changed in this version of KStars.  I am not familiar with any of the software changes this time around since I have been really focused on making all the changes so that we can get daily builds of KStars for Mac on the KDE Binary server.  They don't sound like Mac only issues, nor do they sound like anything I have been working on.  A good idea might be to post these questions/bugs in a new thread where more people might see it.  The only people who will look at this thread are those interested in the Mac beta build.

Please let me know if you notice any other issues and thank you for testing,

Rob
 
2 years 2 months ago #79629

Please Log in or Create an account to join the conversation.

  • Posts: 300
  • Thank you received: 57
Rob wrote:

"There is a bug (I mentioned it earlier in the thread) that I see sometimes when starting an INDIServer where KStars sometimes but not always crashes and I don't know the cause.  This issue has existed for a few months and is not at all related to the INDI drivers issues that were reported with the betas above.  The INDIServer still starts successfully however, and if you restart kstars and reconnect to the INDIserver you already started everything works fine.  I wish I knew why this happens but I haven't figured it out yet. "

Pretty sure that's not just a Mac bug because that exact same behavior happens frequently to me on ubuntu 20.04.3 LTS running on a generic AMD mini-PC.
Last edit: 2 years 2 months ago by Scott Denning.
2 years 2 months ago #79636

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Scott Denning, very interesting.  Is this on INDIServer startup or on shutdown?  I remember there was a cross platform KStars crash that happened sometimes on indiserver shutdown, but I thought this new crash on startup that I started seeing just a few months ago was Mac only.  The. shutdown bug was addressed I believe.

Thanks,

Rob
2 years 2 months ago #79637

Please Log in or Create an account to join the conversation.

Time to create page: 1.381 seconds