I have been trying to narrow down an issue that prevents me from imaging past a meridian flip and I think I am getting closer. I am not fully knowledgeable with the system, but I feel there is an issue either myself or KStar or Ekos or any and all three.

I have been working well but ever since 1.88 update things have gone weird. In summary My Polar alignment and get really good results. After completing the polar alignment I do a check to see if the alignment matches Kstars. The image below says not. First things first so there is not a lot of wasted talk, I imputed my coordinates very accurately, also my time is accurate to the second. I live in a time zone +9.5.

After a very accurate Polar Alignment ( I really like this feature):


I then carry out a simple capture and Solve do nothing:


I find myself way off target both in the solution results and Kstars solver box


I noticed that there was a couple of posts that say Kstars and Ekos use JNow, but I found when I clicked on the Mount Model it shows results in J2000.


I do not know if J2000 is the problem but in the end I can't display the right coordinates in KStars. I can initially plate solve and move to my first target but after a Meridian Flip I can no longer plate solve and move. I saw once I selected Carina Nebula on the second instance and the first solve failed at about 6000 arcsec off and could never move closer. Then it would change the name in the solution box but still couldn't move.

All the time Kstars showed I was on target.

Any help would be appreciated.

Just two quick comments:
If you have perfect PA that does not mean that automatically your main camera will be centered on the NCP. Errors in home position and things like cone error can easily put you off. Those errors don't matter for the PA process, PA itself is correct. So I would not call it a substantial error per se.
The other thing is that the poles are singular points in RA, and even tiny remaining errors in PA can lead to huge offsets. I strongly recommend to not use any plate solve close to the pole to sync your mount. Just go to some place further away from the pole to do a sync.

That is exactly what I did, Only for the first time since the issue started I checked the actual location of the pole, although i did not sync it, the picture shows that I did nothing. When i do my first slew t my target I arrive at 6000 arcsec out and it slew after plate solving within about 3 to 4 attempts. Afterwards after the first solve it works well, I take my images and the guide is very good and random guide are located around in a reasonable distribution circle.
When the system does a meridian flip then it fails all the time to align. The solver works though and produces the error needed for the mount to correct, but it never achieves the correction. It even keeps trying for 10 times then I have to stop it. I then have to park the scope then disconnect and shut down the indi server. When I start it up again everything go back to the correct settings and I can continue my imaging.
Even when I park the scope, after the first solve it parks to 177/92 deg. Yet according to Kstars I am on 180/90 deg.
On the other hand, if I was to slew directly to a target then park again, it would always head back to 180/90. Everything is repeatable. It appears as though even though it shows I am perfectly polar aligned which i think i am I am alway approx 6000 arsec off target and it appears that is consistent across the board.

I may have to create a video just to prove it.

I have been imaging mostly using EQMOD on ASCOM with APT for approx 7 years. So I do understand a lot, but as this has only started since 1.88 indi was updated. So there is either a setting that is corrupted, or a module that is not working. My last course of action will be to re-image, but this would be disappointing if it started working. It would clearly suggest that this software product is not meant for the general astrophotographer.

I have no choice as I am using the only product that has a driver for my focus controller. I hope i don't have the need to develop ASCOM drivers for it to go back to Windows.

Yes, I had seen that 'Nothing' was checked as action for the pole solve, just wanted to be sure.

But the 'update' rang a bell. Are you sure that all components are 'in sync', especially the libindi and indi-3rdparty versions? The eqmod driver isn't part of the main indi, but it has to be compiled against the version used.
I personally also recompile kstars when I change the INDI version. Not sure if that is really needed though.

Thanks for the update, I will check, although I am a hardware electronics person and not really good at software. I hope i can synchronize the updates, I am running Astroberry, so maybe I can put this to the Astroberry Forum instead. I am also going to look deeper into my hardware as well, I have a thought but not sure of the outcome.