Kerbal Space Program

General consensus has been to not open a report unless the issue has been confirmed on a 100% stock install (probably excluding some part mods, I'm not sure) - apparently, even things which seem completly unrelated can screw things up cough*DebRefund*cough

• Revert to SPH seems to undo the bug for a time; a quick F5/F9 cycle might also work
• If the craft reaches 100,000 km, explosions occur (reproduced)
• If you try to EVA a Kerbal, their EVA packs don't have much of an effect. If you then switch back to the vessel, the Kerbal will appear to fall towards the parent body's CoM (I don't know if this is acceleration towards the CoM, or the vessel accelerating away from the CoM). This is very likely a related bug.

I have been able to reproduce the issue on the runway ( http://steamcommunity.com/sharedfiles/filedetails/?id=329263674 ), with a combo of messing with time warp and gravity hacking; gravity is ON in the picture.

The part which remains fixed appears to be the root part; the first time I encountered this (and everything blew up), only my cargo bay remained, which was the root part.

The "related bug" I mentioned: at one point, I deployed an LV-N, and the fairings (which are NOT physicsless, unlike all the other ones) appeared to fall back to Kerbin. I do not know if this bug is really related or not.

I am on the Win32 version, though launching under x86-64, so the core bug would not be x64 exclusive, though some of the effects the OP listed might be.

• Navyfish Docking Port Alignment Indicator (4.0, 6/29/14, KSP ???)
• "Spaceplanes Plus" (integrated into KSP)
• Kerbal Attachment System (0.4.9, 10/8/14, KSP 0.25)
• Procedural Fairings (3.10, 10/11/14, KSP 0.25)
• Karbonite (0.4.2, 10/7/14, KSP 0.25)
• SCANsat (8.0, 10/8/14, KSP 0.25)

Danny2462 might have encountered this in 0.24 (stock?), http://youtu.be/kTg3vB8x3TY?t=9m15s . I have only encountered this in the presence of time warp. Another video from Danny2462, which I don't have a link to, shows a Kerbal with an apparently low-functioning EVA pack; both of these glitches were caused by misuse of the Klaw, but there might be a common fundamental flaw. (DISCLAIMER: I am only reporting the gravity issue as possibly being related, not the destruction of Kerbin)

I'll see about getting this tested in 100% stock.

Might be related to http://forum.kerbalspaceprogram.com/threads/97844 - I was using the Klaw to..."dock"...with a ship at one point, but I don't know if it was in the same game session as either of my incidents (infinite-ascend-until-100K plane and the LV-N fairings). Even then, I can't think of a way for the issue to persist into another vessel, so the Klaw probably isn't the only factor.

http://www.reddit.com/r/KerbalSpaceProgram/comments/2kr6aw/strange_bug_ruining_my_first_eve_mission/
Almost certainly stock.

I seem to be having this problem as well. I run vanilla KSP on windows and I do not use any mods. It first occurred as I time warped a long nuclear engine burn with a ship that is built in 2 parts, connected with the advanced grabbing unit. Reloading the game did not solve it in this case, whenever I reloaded and attempted timewarp it would say my ship was under acceleration even though all engines were off. Staging view showed the ship to be bobbing out of control at the AGU connect point, I tried releasing with the intent of reconnecting but my upper stage just blew up. I restarted a new career and the bug came up again with just a plain rocket that I was using to clear the atmosphere - it is totally stuck and slowly accelerates away from kerbal.

Right now, I'm tempted to say there is a relation to all the Klaw issues; if I connect two Klaws together like so:

OO]<   >[OO
----> <----

Internally, how is the Klaw implemented, compared to docking ports? Does it just act as a port which doesn't require a matching part on the other craft? How are ambuigious or unexpected situations resolved? I once had an issue of my own where a function was inadvertantly returning a bad (and sometimes "good", as in the data existed) value instead of complaining like it should have been: https://github.com/TruePikachu/unSIR0/commit/0c279056fa6d743905cf0b6bd6bb112441991bec shows what I mean.

gforce: Try creating a new Career save without loading the old one

Trying to glean the salient points from this and the linked Reddit, it would appear that it is very similar to #2753 and possibly a duplicate.

We need to distill a simple set of reproduction steps for this issue to be presented to the developers for attention.