Well, I just found out that it does occur with the EQMod driver set on Simulation mode as well.
So the issues I'm facing with my mount are not caused by this. Although this behavior during star cross tests is a bit odd and looks like a bug somehow.

I am getting this tonight With my EQ-8 and 3.5.0. The mount stops guiding, and even worse, seems to stop tracking, then guiding resets and the guide star is re-acquired and the cycle starts again. This is making long exposures nearly impossible. What I cannot tell is if the error is cause or effect.

Log attached

I'm getting something exactly like this where the guiding spikes massively at random points and I'm certain it has to be down to software and not the mount itself.

I've attached my log file as well.

I think you may have attached the wrong one too?

@mattbaker, could you add your equipment to your signature? I am just curious what equipment you are seeing the error on. Here is a hopefully helpful link: indilib.org/forum/new-forum-users/8195-w...-the-indi-forum.html

I double-checked my log attachment, it is the correct file. It was gzipped to save a lot of space.

Jim

Updated my signature and you are correct, it is the right file, my bad

Last night, I set up again. When I enabled guiding, I was seeing the same problem, so it was time to look around some more.
The behaviour was this for me:

1. Start guiding, go through the calibration routine, and everything looks good (except my guiding last night was > 1", but that was sky conditions.
2. After some time, while watching, the image from the guide camera would start to drift for a few frames, exactly the same as if tracking was turned off on the mount
3. The drifting would stop, the guiding would restart, and guiding/tracking would look normal again for a while. "Delta RMS threshold value exceeded. Searching for guide stars..."

I was looking at things in the logs that would indicate the motor needing to reverse direction rather than just slow down. If sidereal is "1" and guiding is ".5" in both RA and DEC, then the actual motor speed should qbe from .5 to 1.5, never negative.
The adjustment throughout the log are +/- 7.5 arc seconds per second, which is 1/2 sidereal in either direction, slowing down or speeding up the motor. Seems fine.

Next up, what other strange things were happening?

When I looked at one of the WARNINGs, I noticed that the mount logs just stopped for a period of 5 seconds

This corresponds to the time the mount drifted. The :f commands are just looking for a status from the mount. During this time frame, My guide camera uploaded 2 images. Each upload took only .002 seconds, so should not have been blocking anything in USB land.

The last successful command given to the mount other than status looked like this:

where commands J and K are start and stop respectively. All commands after that, until the read timeout are mount inquiry, f, j, and d.

There are no errors in the log, other than the timeout that I can see.

Next, I looked at the settings to see if anything did not make sense. I had added integral gain to attempt to fix some issues with guiding, so I set those back to 0. That was really the only change in Ekos. I will paste my settings below.
I looked at the INDI settings, and noted that the max guide pulse was set to 100, but the max guide pulse in Ekos was set higher than this. This did not seem to be proper, so I raised the INDI EQmod driver max pulse so that it was above the setting in Ekos. I also adjusted the minimum pulse settings to better align with each other. The EQmod documentation states that anything under 50ms really doesn't work quite right due to timings, but both values, INDI and Ekos were well below that... They just didn't align with each other very well, with the INDI driver being much more limiting. After making these changes and firing things back up again, the rest of my night went as shown in the screen capture below.



which produced the SHO Hubble-2 image:



I will post my settings when I fire the system up again. I am not sure if I have solved the problem or not, time will tell. I still don't know why the mount was becoming un-responsive and then decided not to after those changes. It could be that the INDI driver does not like being asked to do a pulse longer than it's settings allow?

Jim

So what I'm gathering from this is to up the minimum guide pulse in INDI and EKOS?

I don't know yet whether I solved anything, to be honest, but what I did was to make the Guider settings fit inside the INDI settings. That is, INDI would not try to limit the guider on the low end or the top end. Max guide pulse in INDI, I think I set to 500, and set max in Ekos to 300. For the minimum, I went with 20 in INDI and 30 in Ekos. I also set the integral to 0 in the Ekos/Kstars guider config. There are too many variables to point to any one thing and say it fixed it, but if I get another session like the tail of this last one, I will be happy.

Jim

I found this rather odd. I'm currently out at the moment and it still keeps happening. My guide rate is at 0.8 hence the 12.0329 but why is the RA then deciding to go at 1.8x sidereal

A few seconds later, it receives a command after the spike occurs and is told to do a 1776ms @ 0.8x when it then only does 0.2x

Your first set of logs is "Timed guide West", and the second is "Timed guide East". To guide west, the mount has to go faster than 1.0 (sidereal), so with your setting of .8, the combined speed is 1.8x, and the application computes how long it needs to go at this rate and gives it a time in milliseconds (79ms). The opposite is for guiding east, it needs to slow the mount down by .8, giving you a .2 rate.

Ahh, I see

I do keep getting this error when they occur

EQMod Mount : "[COMM] read error, will retry again... "