Deimos invisible ground kills astronauts/ships

While going down for a Deimos landing, my ship poofed out of existence, like a crash, but while still in space not at the surface. I sent another ship, and i was able to land near the equator and plant a flag, but when I sent my Kerbal on a jetpack adventure to the North, where there's an odd looking spike in the surface, the Kerbal "poofed" in mid-space.
So I guess it might have something to do with height maps or something. It seems it's acting like there's ground when there's not ground yet. This is similar to bug #87, about invisible ground, but that issue is different. After one good landing attempt I experienced an odd negative gravity pulling the ship up as described in bug #87, but that issue isn't catastrophic.

Maybe this catastrophic issue exists because Deimos is not spherical. So this bug could potentially affect any non-spherical moon/asteroid. Plus Deimos is tiny - probably smaller than the fictional Duna moon, even though this is RSS and almost everything is 10x larger.

[siimav]

I can confirm both the spiky north pole and the collision with invisible terrain issues. I haven't been able to pinpoint what causes those terrain collisions because my second landing attempt with a different trajectory was successful. I didn't notice anything like that around Phobos so I think that issue is isolated to only Deimos.
@PhineasFreak any ideas what might be wrong here?

[PhineasFreak]

Unfortunately i do not think that i am the correct guy to ask about Kopernicus CBs. I do also believe that it might be caused by the fact that the body mesh is not round and the fact that it is one of the smallest CBs that you can come across. If so, one could also test and see if the same thing applies to the RSSExpanded bodies (some of them, like Bennu, are irregular and less than 200 m across).

Maybe some spots on the terrain are too close to the center of the body? KSP does not like vessels that come very close to the center of a CB.

[raidernick]

I've had this same issue happen on Io before, where there is some floating invisible "ground" a km or more above the actual "visible" ground that you end up hitting and exploding. As far as I know this was never fixed. I had tested several of the other moons around the solar system and had the issue sporadically happen on callisto and triton, which means it likely exists on others as well.

[PhineasFreak]

Some (many) years later I think I found what the actual problem was: on the first RSS run Kopernicus creates the 3D models of the CBs and stores them as binary (.bin) files. Any change in the height maps (be the actual height map or the height mods) will not update the model on subsequent KSP runs, you have to manually delete the .bin file and recreate it. If an RSS installation is updated with changes in the CB heights it can lead to all sorts of problems (collision mesh above or below ground, vessels phasing through, mismatches between SS and PQS terrains).

Since then RSS has remained more or less unchanged and without further reports of this bug I do believe it had something to do with the process outlined above. Kopernicus also includes a new file hash check where it will automatically recreate the .bin files when it detects any changes in the CB configuration file(s).