Hi
Loads of these errors:

[2018-07-18T23:23:00.126 CEST INFO ][ org.kde.kstars.fits] - Loading FITS file "/tmp/kstars/phd2.fits"
[2018-07-18T23:23:02.638 CEST INFO ][ org.kde.kstars.fits] - Loading FITS file "/tmp/kstars/phd2.fits"
[2018-07-18T23:23:02.640 CEST CRIT ][ org.kde.kstars.fits] - "Could not open file /tmp/kstars/phd2.fits. Error could not open the named file"
[2018-07-18T23:23:05.265 CEST INFO ][ org.kde.kstars.fits] - Loading FITS file "/tmp/kstars/phd2.fits"
[2018-07-18T23:23:08.172 CEST INFO ][ org.kde.kstars.fits] - Loading FITS file "/tmp/kstars/phd2.fits"
[2018-07-18T23:23:08.173 CEST CRIT ][ org.kde.kstars.fits] - "Could not open file /tmp/kstars/phd2.fits. Error could not open the named file"
Th frequency is related to the -in this case- 2 second guide exposure.

Eventually capture is aborted -the log ends abruptly- and kstars disappears from the screen.
The PHD2 window is refreshing fine and guiding is good. AFAICT, PHD2 doesn't write anything to /tmp. Any idea of the cause of these error messages?
TIA
Logs:
drive.google.com/open?id=1EAKu5qeg-KkyI05wsyL94wjh_9Dir2G_

Hi again
OK, found it:
-rw-rw-r-- 1 steve steve 5760 Jul 20 01:12 phd2.fits
It's created after ekos guide connects and starts guiding. It's updated every 2 seconds -the guiding loop exposure- but it doesn't seem to do anything e.g.making /tmp/kstars read only has no effect on the guiding, the latter seems to be handled totally within phd2...

Can you post a link to this file (phd2.fits) ?

It's the frame that's displayed in the phd2 guide window from which phd2 takes the star and lock position. indi removes /tmp/kstars upon exit so I can't produce another until my next session. Meanwhile is that OK?
Note that 'Recieve external guide frames' us unchecked.
TIA

phd2.fits:
drive.google.com/open?id=1b27IzYp7tGYBCaIFUnqpXoLDA0FAB5o3
TIA

So last December when I worked on improving PHD2 support, I found that PHD2 could send a star image back along with its other communications. Thus we could have a star image in the KStars guide module from PHD2. The image is sent via a JSON document, then saved briefly to a temp file, and then loaded into the guide module so that you can see the star if all goes well. It is a very small file, and is always saved in the same file, so it shouldn't take up a lot of space. I purposely made it a temp file so that it would get deleted when you are all done. Sometimes though, the JSON document gets interrupted due to issues on the network and the star image doesn't get saved. Then a couple of error messages should print including some saying that the PHD2 message couldn't be read and one saying that it can't open the file. I believe that I have removed the other messages though because some people were getting annoyed at them. The end result is that that single frame from PHD2 won't get displayed. You are seeing the star image most of the time correct?

Ah, yes. I see. Yes, most of the time I do see the star.
Thanks for your reply and clarification.

'

Would it be possible to move the errors to a higher level of logging? There are so many of them, they make the default log difficult to read.
TIA,
Steve

I'm not sure, that error is not printed out by the part of the code that is involved with PHD2 guiding in KStars. If it was disabled somehow, it would affect other KStars features. I might be able to look into it if I have time. Maybe I could make PHD2 not try to save/load the image if there was a transmission error for that Star Image. I'm in the middle of some other things right now, but. I can look at it soon.

This should be already fixed by 2.9.7 release of KStars.

I've got a fix for this, I'll be submitting a PR soon.

I also have a fix to prevent ekos from trying to process partial response messages from phd2 before they are received in their entirety; I'll send a PR for that too.

Andy

Sorry if this ends up being a double post. I just posted a few minutes ago but it looks like it did not go through.

I've got a fix for the phd2.fit error -- I'll be sending a PR shortly. Here's the branch if anyone cares to look: github.com/KDE/kstars/compare/master...agalasso:phd2_fixes

It also contains a fix for the problem with ekos trying to process partial messages from phd2.