Plate solving issue that likely needs a message to assist users

Sequence Generator Plate Solving
1

I just wanted to file a bug report for the plate solving feature of SGP 2.6.0.13. I am not sure why, and it may have just been a fluke, but when I upgraded the other day, it apparently wiped out a few settings. This included resetting the PlateSolve filter to None, from Lum.

When I first tried to start some sequences last night, the center feature and plate solve feature kept failing. Some dialogs would flicker onto the screen for a moment then disappear. I wasn’t sure what was going on at first, exited and restarted a few times, and even downgraded once. I started digging through the logs, and eventually realized that the issue was the “none” filter selected in PlateSolve settings. Once I set that back to Lum, plate solving worked.

While it was helpful to have the logs, I think this is a situation that should be more clear to the user. Some kind of dialog should be displayed explaining that PlateSolve couldn’t continue because the filter chosen could not be set (I am not really sure why “None” was even an option, since my filter wheel does not have any filter labeled None, so certainly, such a filter could not have been set.)

2

Not sure I follow you. None is a valid option and should not issue a warning. If none is selected, it means that SGPro should attempt to solve using the current filter (and does not require a “none” position to exist).

Need clarification here. There was no filter chosen so there is no apparent warning. Is the plate solve not continuing? Seems like it is and it’s failing?

3

The plate solve never seems to begin. When it starts, there is a fraction of a second where I can see a dialog appear, but it’s gone before anything renders. Then the sequence reappears and no further actions occur.

In the log, I noticed this:

[02/20/17 19:09:38.721][DEBUG] [Telescope Thread] Center telescope message received…
[02/20/17 19:09:38.721][DEBUG] [Telescope Thread] Solving with Plate Solver PlateSolve2…
[02/20/17 19:09:38.752][DEBUG] [Telescope Thread] Setting filter for auto center…
[02/20/17 19:09:38.754][DEBUG] [Telescope Thread] Could not set the centering override filter! Aborting auto center…
[02/20/17 19:09:38.754][DEBUG] [Telescope Thread] Aborting auto center 1…
[02/20/17 19:09:38.754][DEBUG] [Telescope Thread] Auto center aborted, closing centering dialog…
[02/20/17 19:10:09.513][DEBUG] [Telescope Thread] Center telescope message received…
[02/20/17 19:10:09.513][DEBUG] [Telescope Thread] Solving with Plate Solver PlateSolve2…
[02/20/17 19:10:09.554][DEBUG] [Telescope Thread] Setting filter for auto center…
[02/20/17 19:10:09.554][DEBUG] [Telescope Thread] Could not set the centering override filter! Aborting auto center…
[02/20/17 19:10:09.554][DEBUG] [Telescope Thread] Aborting auto center 1…
[02/20/17 19:10:09.554][DEBUG] [Telescope Thread] Auto center aborted, closing centering dialog…

I checked out the PlateSolve settings after seeing the above, and noticed my filter setting, which had been set to Lum before, was now set to None. I set it back to Lum, and plate solving/centering started working again.

What I am requesting is, when an issue like that occurs, instead of just allowing the requested action to fail and end without any notification to the user, it would be helpful to at least get some kind of message dialog saying what happened (with an auto-close timeout even, to avoid potentially hanging up anything else). Whatever the issue, whatever the cause. I was just left wondering what the heck was going on, trying various things, until I finally delved into the logs, which is never all that fun.

In this case, the issue was apparently “Could not set centering override filter!” I am not sure why None was not working, even if it should have been a valid option. Regardless of that, though, an error was occurring, apparently a well defined one, and an unexpected result was experienced in the UI (plate solving never even started and the centering window only flickered onto the screen for a fraction of a second before returning to the sequence window.)

4

just FYI i have seen this exact issue many, many versions ago so this is not unique to 2.6.x.x. same behavior, quick flash of the centering box with a “silent” failure to start the platesolve process. re-setting the platesolve filter also resolved the problem for me.

rob

5

@jon.rista

Please attach logs so we can see what is happening.

6

I just encountered the issue again. Sadly, my full log file is over 12 megs, as I was running some extensive tests on 2.6.0.14, to see how long it could go without crashing or what have you. I can truncate the log, eliminate the middle part that contains repetitive F&F exposures, and include the header and all the lines related to the imaging sequence I was trying to start.

Would that suffice?

7

It would be best to zip it or use something like Dropbox to get it to us. Also a general time of when you noticed the issue would be helpful.

Thanks,
Jared

8

I just experienced this issue with 2.6.0.12 I only found this thread after upgrading to 2.6.0.16 after which I went and checked my logs and saw the same message Jon reported. I’ve since gone to the plate solve settings and confirmed that the filter is correctly set to L in 2.6.0.16 but haven’t tried it again to confirm that it works.