After 3-4 hours of capture without problem, I have also problem with "Downloading..." stuck after a meridian flip, stopping it and I can continue to capture.

So problem comfirmed also with Touptek driver updated (IMX585 cooled) and Star Adventurer GTi.

Hi
Also with omegon veTEC. I think this is the same issue:
www.indilib.org/forum/ccds-dslrs/14136-o...ra-disconnected.html

Log attached to that post.

Thanks

We can use this workaround, example M78 capture a few minutes before FLIP, and restart a few minutes after, so align is not triggered, night goes ok! 300 capture without a problem.

So I can confirm that there is a problem, with ALIGN & CAPTURE stuck in downloading after a Meridian Flip.

This makes no sense to me, why would it just get stuck after meridian flip? In the INDI control panel, is the camera stuck? aborting and restart from INDI control panel works?

Jasem, thanks for reading, if you look at my logs at the beginning you can see that apparently Ekos does not acknowledge the driver successfully completing the exposure. The driver does not report anything abnormal and completes successfully, but Ekos is stuck on Downloading. An abort from Capture itself (stop button) will unfreeze it. But after a few exposure, the camera driver crashes and exit with -1, without anything in the logs justifying it.
The meridian flip part is not logical, you're right, but I did some dry runs for you on a bad weather night, and I could repro this 5 times in a row, always after a flip. Empirical, but consistent! Logs always look like the ones I posted above.

Exactly the same for me "Downloading..." stuck, I press "stop capture" and can continue (I didn't have the patience to wait for the camera driver to crash yet).

The first time, I only noticed it in the morning, everything else seemed to still work.

But I can add some hint, during that night I have 2 flip (on 2 different target), first one aborted a capture and the capture resume correctly, after the second flip, "Downloading..." stuck, seem like the capture BEFORE flip leave some problem, that can't resume after the flip.

Hi
Thanks @FRANSESCO. Your workaround looks promising.
In my case, after the flip, the camera disconnects. The driver needs to be restarted to carry on the schedule:

[2023-12-20T01:41:49.095 CET INFO ][ org.kde.kstars.ekos.mount] - Meridian flip: slewing to RA= "06h 20m 42s" DEC= " 22° 33' 19\"" Hour Angle "00h 00m 50s"
[2023-12-20T01:41:49.095 CET DEBG ][ org.kde.kstars.ekos.mount] - Setting meridian flip status to "MOUNT_FLIP_RUNNING"
[2023-12-20T01:41:49.095 CET INFO ][ org.kde.kstars.ekos.mount] - "Meridian flip running..."
[2023-12-20T01:41:49.095 CET DEBG ][ org.kde.kstars.ekos.capture] - updateMFMountState: "MOUNT_FLIP_RUNNING"
[2023-12-20T01:41:49.095 CET DEBG ][ org.kde.kstars.ekos.capture] - updateMeridianFlipStage: "MF_INITIATED"
[2023-12-20T01:41:49.095 CET DEBG ][ org.kde.kstars.ekos.capture] - Capture State changes from "Calibrating" to "Meridian Flip"
[2023-12-20T01:41:49.101 CET INFO ][ org.kde.kstars.ekos.align] - Target coordinates updated to JNow RA: "06h 20m 42s" DE: " 22° 33' 19\""
[2023-12-20T01:41:49.198 CET DEBG ][ org.kde.kstars.ekos.scheduler] - Capture State "Meridian Flip"
[2023-12-20T01:41:49.199 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Starting Goto RA=6.34487 DE=22.5552 (current RA=6.33959 DE=22.5663) "
[2023-12-20T01:41:49.200 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] GOTO ALign Nearest: delta RA = -0.005430, delta DEC = 0.008842 "
[2023-12-20T01:41:49.201 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Aligned Eqmod Goto RA=6.33944 DE=22.564 (target RA=6.34487 DE=22.5552) "
[2023-12-20T01:41:49.201 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Setting Eqmod Goto RA=6.33944 DE=22.564 (target RA=6.34487 DE=22.5552) "
[2023-12-20T01:41:49.202 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Slewing mount: RA increment = -5567912, DE increment = 4171970 "
[2023-12-20T01:41:49.203 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Slewing to RA: 6:20:42 - DEC: 22:33:19 "
[2023-12-20T01:41:49.203 CET DEBG ][ org.kde.kstars.ekos.capture] - updateMeridianFlipStage: "MF_FLIPPING"
[2023-12-20T01:41:49.216 CET INFO ][ org.kde.kstars.ekos.guide] - "PHD2: Guiding Stopped."
[2023-12-20T01:41:49.216 CET DEBG ][ org.kde.kstars.ekos.capture] - Guiding state changed from "Guiding" to "Aborted"
[2023-12-20T01:41:49.219 CET INFO ][ org.kde.kstars.ekos.guide] - "Autoguiding aborted."
[2023-12-20T01:41:49.219 CET DEBG ][ org.kde.kstars.ekos.scheduler] - Guide State "Aborted"
[2023-12-20T01:41:49.287 CET INFO ][ org.kde.kstars.ekos.guide] - "PHD2: Lock Position Lost."
[2023-12-20T01:41:49.548 CET DEBG ][ org.kde.kstars.ekos.mount] - Mount status changed from "Tracking" to "Slewing"
[2023-12-20T01:41:51.055 CET INFO ][ org.kde.kstars.ekos.guide] - "PHD2: Guiding Stopped."
[2023-12-20T01:41:51.056 CET DEBG ][ org.kde.kstars.ekos.scheduler] - Guide State "Aborted"
[2023-12-20T01:41:53.008 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Align: current point is 2 "
[2023-12-20T01:41:57.034 CET INFO ][ org.kde.kstars.indi] - EQMod Mount : "[INFO] Align: current point is 0 "
[2023-12-20T01:42:11.234 CET INFO ][ org.kde.kstars.indi] - Astroshop veTEC571C(USB2.0) : "[ERROR] Camera disconnected "
[2023-12-20T01:42:11.485 CET INFO ][ org.kde.kstars.indi] - Astroshop veTEC571C(USB2.0) : "[ERROR] get Temperature error. Generic failure "

After which there is no further capture.
Thanks for taking a look.
Steve

Hi everyone
There seems to be a similar issue here:
indilib.org/forum/ekos/14138-after-merid...ure-time-failed.html

Workaround. Question about avoiding the meridian flip align::
Given an exposure sequence, do we have a simple way of stopping 15 minutes before and resuming 15 minutes after the MF is due?
A scheduler setting perhaps?

TIA,
Steve

Use the scheduler untile time (look at my image before), but you need to know the hour in whitch MF is due.

I found a strange bug with meridian flip but not exactly like issue reported above. The issue reported above is most likely to some cable snap during the flip as it is most likely a hardware issue. I will also test with my QHY268 once weather clears up.

Thanks Jasem.
For what it's worth, I already completely ruled out cable snags. My sessions are never unattended and I actually walk up close during flip making sure all the cables are clear. I also tried several different cables. This is definitely software, also because when I restart Indi server, it resumes and never crashes again.

hello jasem, it can't be a cable problem, just because even though you don't touch anything you just press stop and play on the capture module to resume regularly (at least in my case, it never disconnected my Touptek) after the meridian flip.

Yesterday I did some testing taking advantage of the full moon, it is possible that "pausing" the capture module and then resuming, gets stuck in downloading, this might make sense with the Flip, it is not the Flip that is the problem, and that between pause and resume in some cases you get stuck in "Downloading..."

I'm curious what bug you found though (point out a piece of code as well ).