- edited description
Upgrade to latest nightly causes crash on focusing.
Left the scope imaging 2I/Borisov this morning and went to bed. Returned to find it had apparently hung during focusing, with no obvious reason why, 50 mins into a 3 hour run, which was disappointing!
It seems to have completed a successful hyperbolic fit, set focuser to best focus, and hung at that point.
This was my first run with nightly. Had been using #26 which works perfectly for me (aside from occasionally crashing shortly after starting, which never recurs).#008
Attached was view at 7am this morning when I got up. Will revert to nightly !#8
Comments (10)
-
reporter -
repo owner are we talking about a crash, meaning application is unresponsive, or it was just stuck after autofocus and you had to cancel and resume the sequence?
additionally logs for that session might be helpful.
Thanks.
-
reporter Sorry, not a crash by that definition. Unresponsive, for many hours! I suppose it could have been an issue like the autoguider not finding a star perhaps causing a hang? Where do I find the logs?
-
reporter - attached 20191111-170905-1.10.0.1026.7400.log
hopefully this is it!
-
reporter the hang was at about 0430UT 12/11/2019, give or take, so i dont think the log contains anything.
-
repo owner It might be possible that the hanging was caused by trying to pause guiding. I have adjusted the logic a bit in the latest nightly and hopefully this will be more robust.
-
reporter Hi Stefan! Thanks. In fact I experienced a similar thing last night using nightly
, where (I think) cloud made the star autoselect fail, judging by the ‘autoselectfail.fit’ files in the PHD2/ directory.#18So I think you may well be right, and its not NINA’s fault.
Is there any way to make the code repeatedly try to restart guiding at (say) 5 minute intervals if this happens? As it stands, a passing cloud on guiding start hangs the sequence.
-
reporter and yes, i also have ‘autoselect fail’ filed from the intitial ‘hang’ the previous night! Perhaps I need to loosen PHD2s criteria for a guidestar, if thats possible…
-
reporter Stephan I think this can be closed now! However another issue on restarting after guidestar autoselect fail might worthwhile.
-
repo owner - changed status to resolved
closed as per request.
A different issue for autoselection after a failure might be opened instead.
- Log in to comment