Kerbal Space Program

Alewa wrote:

I think we misunderstood each other. This is a mission from a normal career. A similar situation I had with another job from my career. After the execution, I got my rewards, and in the menu to the right there were checkmarks everywhere (during the flight). After a while all the ticks disappeared, the mission was displayed as new. But on my return to the space center it disappeared.

Then seems you mean a contract, not a mission. Mission has a very different meaning in KSP.

The situation you provided shows what follows:
- persistent.sfs and 4.sfs are both the same, before the flight (note date/time); wtf1.sfs is after you completed all but one of the tasks with both contracts. There is no save showing when you got the bug.
- with wtf1.sfs, the Tracking Station has all filters off (but for unknown objects). There's just to toggle them on again (click on each one on their icon, top bar in TS) and they will stay on. Nothing shows them being off was the result of a bug. TS filters haven't been seen to toggle by themselves anytime while replaying your save(s).
- with wtf1.sfs, the vessel is initially without signal, uncontrollable. No course correction can be made, and on its actual course it will never meet Kerbin (thus failing the last task with both contracts). Kerbin's position is also out of phase for a transfer window. On its actual course, the vesel will get close enough to Kerbin for a faint control to be possible by Year 6, Day 2 (1 Year, 168 Days after the situation with wtf1.sfs). Only at that point will be possible to correct course and achieve the last objective with both contracts.

However, there's the bug occurring before then. Previously reached objectives will be reset. I started to notice the reset at Year 5, Day 151 while timewarping; reducing warp and saving the game could arrive at Year 5, Day 242, time 00:12 when the reset will occur at any warpspeed.
There is no evident reason for the reset to happen: contract hasn't expired, no event shows in game nor in the log. Conversion of the specific date in UT doesn't point to any errors in the date/time format.

It is currently unknown to me, using a stock KSP 1.4.4, what could be the reason for that reset bug. It can't be ruled out to be something tied to what one of the add-ons you had installed has registered in the savegame. Only with a clean savegame, that saw no add-on involvement, this can be considered a bug in stock KSP, at which point will be handled by Squad Developers. Should no reproduction come with a clean KSP copy, you may want to consider listing the whole set of add-ons that were in use at any time during your game, any help in dealing with add-ons is better handled on https://forum.kerbalspaceprogram.com/index.php?/forum/70-technical-support-pc-modded-installs/ instead of the official bugtracker.

diomedea wrote:

Alewa wrote:

I think we misunderstood each other. This is a mission from a normal career. A similar situation I had with another job from my career. After the execution, I got my rewards, and in the menu to the right there were checkmarks everywhere (during the flight). After a while all the ticks disappeared, the mission was displayed as new. But on my return to the space center it disappeared.

Then seems you mean a contract, not a mission. Mission has a very different meaning in KSP.

The situation you provided shows what follows:
- persistent.sfs and 4.sfs are both the same, before the flight (note date/time); wtf1.sfs is after you completed all but one of the tasks with both contracts. There is no save showing when you got the bug.
- with wtf1.sfs, the Tracking Station has all filters off (but for unknown objects). There's just to toggle them on again (click on each one on their icon, top bar in TS) and they will stay on. Nothing shows them being off was the result of a bug. TS filters haven't been seen to toggle by themselves anytime while replaying your save(s).
- with wtf1.sfs, the vessel is initially without signal, uncontrollable. No course correction can be made, and on its actual course it will never meet Kerbin (thus failing the last task with both contracts). Kerbin's position is also out of phase for a transfer window. On its actual course, the vesel will get close enough to Kerbin for a faint control to be possible by Year 6, Day 2 (1 Year, 168 Days after the situation with wtf1.sfs). Only at that point will be possible to correct course and achieve the last objective with both contracts.

However, there's the bug occurring before then. Previously reached objectives will be reset. I started to notice the reset at Year 5, Day 151 while timewarping; reducing warp and saving the game could arrive at Year 5, Day 242, time 00:12 when the reset will occur at any warpspeed.
There is no evident reason for the reset to happen: contract hasn't expired, no event shows in game nor in the log. Conversion of the specific date in UT doesn't point to any errors in the date/time format.

It is currently unknown to me, using a stock KSP 1.4.4, what could be the reason for that reset bug. It can't be ruled out to be something tied to what one of the add-ons you had installed has registered in the savegame. Only with a clean savegame, that saw no add-on involvement, this can be considered a bug in stock KSP, at which point will be handled by Squad Developers. Should no reproduction come with a clean KSP copy, you may want to consider listing the whole set of add-ons that were in use at any time during your game, any help in dealing with add-ons is better handled on https://forum.kerbalspaceprogram.com/index.php?/forum/70-technical-support-pc-modded-installs/ instead of the official bugtracker.

Yup, all that you say is accurately, but the "explore Duna" contract was almost completed by another craft (as I remember), so you can ignore it. Also I didn't change TS filters by myself, it seems that it happen before the 4.sfs. Here's is the list of mods that I used before. As you can see mod "scatterer" was expired after another update but it worked without errors.

Alewa wrote:
Yup, all that you say is accurately, but the "explore Duna" contract was almost completed by another craft (as I remember), so you can ignore it. Also I didn't change TS filters by myself, it seems that it happen before the 4.sfs. Here's is the list of mods that I used before. As you can see mod "scatterer" was expired after another update but it worked without errors.

None of the mods listed brings me any idea about them messing with contracts. However, that may be not relevant after some further investigation brought to find the cause of the contract objectives reset. Started by noticing that the following is logged whenever objectives get reset (see OrbitDriver message log.png):

[OrbitDriver]: On-Rails SOI Transition from Sun to Kerbin.
Transition UT Range: 42012756.491 - 42012756.511.
Transition UT: 42012756.501. Iterations: 24.
[ContractParameter]: on Contract Complete the Supreme Mun 4 Rally. set to Incomplete.
[ContractParameter]: on Contract Complete the Supreme Mun 4 Rally. set to Incomplete.
[ContractParameter]: on Contract Complete the Supreme Mun 4 Rally. set to Incomplete.

Now, I would expect that objectives get reset, if your spacecraft "большой круиз" actually entered Kerbin's SOI but then left without making that last objective (which requires to be on a crash-course with Kerbin, meaning to get on a suborbital trajectory with the planet). That's default behaviour with the contract system, so to allow other vessels to try a contract if one couldn't complete it.
However in the case with your wtf1 situation (updated in qk_y5d242h0012, where objectives will get reset within 60 seconds), is evident your spacecraft doesn't enter Kerbin's SOI. So the question, what is happening? I then removed in turn each and every vessel (including debris) that the OrbitDriver would be tracking from the game. Nothing changed, but then tried to track the asteroids near Kerbin and discovered one that was about to enter its SOI... exactly at the moment the objectives reset. You can see asteroid FJT-890 in screenshots81 (just before its SOI change) and in screenshot82 (just afterwards). A closer look at it in TS (screenshot83) reveals it is in a hyperbolic orbit that will make it leave Kerbin's SOI within days.
I haven't clear what is so special about that asteroid: most of those discovered around Kerbin will enter its SOI at some moment. Perhaps is because its orbit may bring it close to Mun, so fulfilling one other of the contract objectives. Certainly the contract system, that was at this time waiting for a vessel to enter Kerbin's SOI, is considering it to be "the vessel". One side note but of much importance: nothing in the save ties a specific vessel to a contract nor to any of the fulfilled objectives. It is certainly disturbing to find an uncontrollable "vessel" to spoil a contract, though asteroids are certainly parsed for their own contracts.
Now, what I found to be of help, and you could use as well: using the buttons in Tracking Station to first Track, then Terminate said asteroid (which actually stops tracking, doesn't remove it from the game immediately) has worked on said asteroid FJT-890 to remove it from vessels being considered by the contract system: screenshot84 shows 20 minutes after the moment objectives were reset, that still they are checked. Unfortunately this has to be done with all asteroids, until the mission ends.

In the end, to me is certainly a bug this fact that other "vessels" coming out of nowhere can activate just one condition (like the entry in Kerbin's SOI) that is associated with one contract, and consequently cause it to fail. Still need to find if this can happen in a clean game (where no add-ons were installed at any time).

diomedea wrote:

None of the mods listed brings me any idea about them messing with contracts. However, that may be not relevant after some further investigation brought to find the cause of the contract objectives reset. Started by noticing that the following is logged whenever objectives get reset (see OrbitDriver message log.png):

[OrbitDriver]: On-Rails SOI Transition from Sun to Kerbin.
Transition UT Range: 42012756.491 - 42012756.511.
Transition UT: 42012756.501. Iterations: 24.
[ContractParameter]: on Contract Complete the Supreme Mun 4 Rally. set to Incomplete.
[ContractParameter]: on Contract Complete the Supreme Mun 4 Rally. set to Incomplete.
[ContractParameter]: on Contract Complete the Supreme Mun 4 Rally. set to Incomplete.

Now, I would expect that objectives get reset, if your spacecraft "большой круиз" actually entered Kerbin's SOI but then left without making that last objective (which requires to be on a crash-course with Kerbin, meaning to get on a suborbital trajectory with the planet). That's default behaviour with the contract system, so to allow other vessels to try a contract if one couldn't complete it.
However in the case with your wtf1 situation (updated in qk_y5d242h0012, where objectives will get reset within 60 seconds), is evident your spacecraft doesn't enter Kerbin's SOI. So the question, what is happening? I then removed in turn each and every vessel (including debris) that the OrbitDriver would be tracking from the game. Nothing changed, but then tried to track the asteroids near Kerbin and discovered one that was about to enter its SOI... exactly at the moment the objectives reset. You can see asteroid FJT-890 in screenshots81 (just before its SOI change) and in screenshot82 (just afterwards). A closer look at it in TS (screenshot83) reveals it is in a hyperbolic orbit that will make it leave Kerbin's SOI within days.
I haven't clear what is so special about that asteroid: most of those discovered around Kerbin will enter its SOI at some moment. Perhaps is because its orbit may bring it close to Mun, so fulfilling one other of the contract objectives. Certainly the contract system, that was at this time waiting for a vessel to enter Kerbin's SOI, is considering it to be "the vessel". One side note but of much importance: nothing in the save ties a specific vessel to a contract nor to any of the fulfilled objectives. It is certainly disturbing to find an uncontrollable "vessel" to spoil a contract, though asteroids are certainly parsed for their own contracts.
Now, what I found to be of help, and you could use as well: using the buttons in Tracking Station to first Track, then Terminate said asteroid (which actually stops tracking, doesn't remove it from the game immediately) has worked on said asteroid FJT-890 to remove it from vessels being considered by the contract system: screenshot84 shows 20 minutes after the moment objectives were reset, that still they are checked. Unfortunately this has to be done with all asteroids, until the mission ends.

In the end, to me is certainly a bug this fact that other "vessels" coming out of nowhere can activate just one condition (like the entry in Kerbin's SOI) that is associated with one contract, and consequently cause it to fail. Still need to find if this can happen in a clean game (where no add-ons were installed at any time).

I don't know if it should be so that after completing the contract and waiting a bit time all checkmarkers disappear but it didn't anything bad. Just looks strange (4.jpg). Objectives are to take thermal data.

diomedea wrote:

After completing the contract (all objectives checkmarked) it should be archived, rewards awarded for completing it with the relevant message shown in the contract app; in the savegame it would be registered among "CONTRACT_FINISHED". Bugs about contracts not completing have been reported in the past, some fixed. Sorry can't find from your pic what case it is about; it may be a new one.
However I understand you're reporting the above as further proof about the objectives being reset (checkmarks disappearing). Is this with a new unmodded game? In that case, a save before this issue happens would be really helpful.
Would be helpful also if you could isolate the moment objectives are reset and find if any message shows in console (open with Alt-F12) or log just before the "[ContractParameter]: on Contract <contract_name>. Set to Incomplete" lines. Perhaps the cause this time is different, but still enough to spoil the Contract System.

Okay, I did it again and understood that it is not applicable to our theme. The reason is going to menu (press "Esc" button) when you have fulfilled contract. There was nothing in console about it. I attached save files in ms1.zip archive if you need (spacecraft "чит" is right on the cource to last obj). The game should be clean becase I deleted it with Steam, manually deleted all files from "\\commom\Kerbal Space program" folder and installed again. I keep searching.

diomedea wrote:

Indeed. Save file helped understand. The contract actually completes after the last objective is met: on the Contract App, above the checkmarked objectives, the contract title turns green; it can be seen as archived from Mission Control. Is true the checkmarks get reset after pausing the game, but at that point have no more meaning: checkmarks can't be verified for completion anymore as the contract is not active, and would disappear from the Contract App after a scene change. No bug here. Thanks for helping.

I did it again! The same contract type - the same problem. Attached logs and saves. Contract: Eve, Minmus, Ike and Duna flight. Craft: "Полетушка". Approx time of detection: +2y270d. I also managed to somehow ignore the first spoiling asteroid, but the second did it

diomedea wrote:

Indeed. Save file helped understand. The contract actually completes after the last objective is met: on the Contract App, above the checkmarked objectives, the contract title turns green; it can be seen as archived from Mission Control. Is true the checkmarks get reset after pausing the game, but at that point have no more meaning: checkmarks can't be verified for completion anymore as the contract is not active, and would disappear from the Contract App after a scene change. No bug here. Thanks for helping.

I could skip a few asteroids by constant reloads and was stopped by the another one. Then I tracked and deleted (terminated tracking) all asteroids but it nothing changed. Save is attached, approx meeting +3y275d. The same craft

diomedea wrote:

Indeed. Save file helped understand. The contract actually completes after the last objective is met: on the Contract App, above the checkmarked objectives, the contract title turns green; it can be seen as archived from Mission Control. Is true the checkmarks get reset after pausing the game, but at that point have no more meaning: checkmarks can't be verified for completion anymore as the contract is not active, and would disappear from the Contract App after a scene change. No bug here. Thanks for helping.

Flew a little again, tracked the culprit (EXE-937). And it seems that if you do not look at the contract menu (do not check the checkmarks) and just perform the tasks, the contract will be passable (maybe theese checkmarks reset just only visual? I'll check this later). Also I noticed strange flashing icon of unidentified object on my ship when was at the tracking station and turned on timewarp. Saves are attached (again3.zip).

Thanks for those saves. They provide some interesting evidence:
1. is the confirmation the issue wasn't tied to any of the add-ons.
2. the issue (checkmarks with contract objectives being reset) can happen with different objectives. In the "Again" series the last objective is passing by Minmus, so it is the one stray asteroids are temporarily fulfilling, bringing to a reset of the whole set for the intended vessel.
3. The situation is temporary. I've played the "Again1" save up to the conclusion of the contract, 3 times had the objectives resetted during it; each time I then used a trick to make those goals show again checkmarked.

The trick: after the stray asteroid has messed up with goals, go to the tracking station; select the correct vessel and fly it. All already achieved goals are showing again. This actually proves those goals have never vanished, the issue being about the checkmarks needing a refresh (they seem to be computed again for a vessel that is becoming active).

Note 1: tested the above trick also with the situation in wtf1, and works in there too. Now I'd like to hear is it works for you too.

So, while all the above is a confirmation this bug exists, the effects at this time can be considered mild: contracts are not definitely ruined because of another object temporarily fulfilling just one of the objectives, but are just the checkmarks not being automatically refreshed afterwards for the active vessel. So, indeed, that confirms your last observation about the reset being only visual.

Note 2: the flashing unidentified object icon above the active vessel: confirmed here, this would need to be reported as a different issue. But is unlikely it will receive much attention, as the issue has no real effect on the game.