Jump to content

[1.0.4] ScienceAlert 1.8.9: Experiment availability feedback (July 13)

xEvilReeperx

By xEvilReeperx
in KSP1 Mod Releases

 Share

Recommended Posts

troyfawkes

troyfawkes

Posted
Posted

So the log told me to share this...? I was wandering around on Minmus with Jeb doing some pretty nefarious things, but I'm not sure what caused the surfaceSample error. I didn't get this error on Mun with my landers. This might have been due to connecting multiple ships with KAS?

[ERR 22:49:17.733] ScienceAlert, Failed to create surfaceSample ScienceSubject. If you can manage to reproduce this error, let me know.

[ERR 22:49:17.733] ScienceAlert, Exception was: System.NullReferenceException: Object reference not set to an instance of an object

at ScienceSubject..ctor (.ScienceExperiment exp, ExperimentSituations sit, .CelestialBody body, System.String biome) [0x00000] in <filename unknown>:0

at ResearchAndDevelopment.GetExperimentSubject (.ScienceExperiment experiment, ExperimentSituations situation, .CelestialBody body, System.String biome) [0x00000] in <filename unknown>:0

at ScienceAlert.Experiments.Observers.ExperimentObserver.UpdateStatus (ExperimentSituations experimentSituation, System.Boolean& newReport) [0x00000] in <filename unknown>:0

[LOG 22:49:17.747] [KAS] Grab - Grabbing part :KAS.CPort1

[LOG 22:49:17.758] [KAS] Detach(Base) Attach mode is Docked:False,Coupled:False,FixedJoint:False,StaticJoint:True

[LOG 22:49:17.758] [KAS] Detach(Base) Attach type is : StaticJoint

[LOG 22:49:17.759] [KAS] Detach(Base) Removing static rigidbody and fixed joint on Radial connector port

[LOG 22:49:17.775] FF: part couple event

[LOG 22:49:17.776] FF: from vessel KAS.CPort1 (Vessel)

[WRN 22:49:17.776] [PartJoint]: None of the provided nodes was valid!

[LOG 22:49:17.777] RemoteTech: SatelliteManager: OnVesselDestroy(58bac3d4-d5be-47d5-9ab1-61b21ce9fb96, Jebediah Kerman Debris)

[LOG 22:49:17.778] RemoteTech: SatelliteManager: UnregisterProto(58bac3d4-d5be-47d5-9ab1-61b21ce9fb96)

[LOG 22:49:17.786] ScienceAlert, ExperimentManager.OnVesselWasModified: rescanning vessel for experiment modules

[LOG 22:49:17.787] ScienceAlert, Done

[LOG 22:49:17.788] [iR GUI] vessel kerbalEVA (Jebediah Kerman)

[LOG 22:49:17.788] [iR GUI] 0 groups

[ERR 22:49:17.922] ScienceAlert, Failed to create surfaceSample ScienceSubject. If you can manage to reproduce this error, let me know.

[ERR 22:49:17.922] ScienceAlert, Exception was: System.NullReferenceException: Object reference not set to an instance of an object

at ScienceSubject..ctor (.ScienceExperiment exp, ExperimentSituations sit, .CelestialBody body, System.String biome) [0x00000] in <filename unknown>:0

at ResearchAndDevelopment.GetExperimentSubject (.ScienceExperiment experiment, ExperimentSituations situation, .CelestialBody body, System.String biome) [0x00000] in <filename unknown>:0

at ScienceAlert.Experiments.Observers.ExperimentObserver.UpdateStatus (ExperimentSituations experimentSituation, System.Boolean& newReport) [0x00000] in <filename unknown>:0

[ERR 22:49:18.038] ScienceAlert, Failed to create surfaceSample ScienceSubject. If you can manage to reproduce this error, let me know.

[ERR 22:49:18.039] ScienceAlert, Exception was: System.NullReferenceException: Object reference not set to an instance of an object

at ScienceSubject..ctor (.ScienceExperiment exp, ExperimentSituations sit, .CelestialBody body, System.String biome) [0x00000] in <filename unknown>:0

at ResearchAndDevelopment.GetExperimentSubject (.ScienceExperiment experiment, ExperimentSituations situation, .CelestialBody body, System.String biome) [0x00000] in <filename unknown>:0

at ScienceAlert.Experiments.Observers.ExperimentObserver.UpdateStatus (ExperimentSituations experimentSituation, System.Boolean& newReport) [0x00000] in <filename unknown>:0

Link to comment
Share on other sites
xEvilReeperx

xEvilReeperx

Posted
  • Author
Posted
So the log told me to share this...? I was wandering around on Minmus with Jeb doing some pretty nefarious things, but I'm not sure what caused the surfaceSample error. I didn't get this error on Mun with my landers. This might have been due to connecting multiple ships with KAS?

Great, I've been hunting this bug a long time. Any detail you can provide on what you were doing at the time would be very helpful, especially if you happen to have detailed steps that reproduce it every time. I have a suspicion on the cause but I've never been able to reliably reproduce it

Link to comment
Share on other sites
troyfawkes

troyfawkes

Posted
Posted

I'll keep an eye out, I have to do this same mission out to Ike so I might get similar results.

Short story: Rocket over Minmus was 3 stages [at this point], 1) crasher with no probe, 2) miner with probe, 3) joyride monoprop thing with capsule+Jeb.

Landed the crasher, decoupled (so now it has no probe, maybe classed as debris?), landed the miner next to it, landed the joyride next to that. KAS connected all three, went in and out of the capsule a couple times with Jeb, decoupled because things started vibrating a bit. I'm not sure at what point I took a surface sample but I'm sure I did.

Here's the full log of that session just in case the context around it helps (https://drive.google.com/file/d/0B9QXdjV-pBueWVVfMG5lM1NwUWM/view?usp=sharing).

If you can figure out the NaN error at the end I'd be super happy too :) totally unrelated and I'm assuming it's clipping related, but not sure...

Link to comment
Share on other sites
  • 2 weeks later...
taniwha

taniwha

Posted
Posted (edited)

xEvilReeperx This patch fixes the NRE thrown when EL loads a .craft file for determining the vessel cost. It should also make SA more robust on normal launches as the exact same condition (packed, protoVessel is null) happens then, too (I discovered this while writing KerbalStats).


diff --git a/ScienceAlert/ProfileData/ScienceAlertProfileManager.cs b/ScienceAlert/ProfileData/ScienceAlertProfileManager.cs
index de3a7f7..2303392 100644
--- a/ScienceAlert/ProfileData/ScienceAlertProfileManager.cs
+++ b/ScienceAlert/ProfileData/ScienceAlertProfileManager.cs
@@ -317,7 +317,11 @@ namespace ScienceAlert

                 // it's possible the new vessel is in fact packed (almost certain to be a DiscoverableObject)
                 // so we need to be careful not to access any parts if it is
-                uint mid = newVessel.packed ? newVessel.protoVessel.protoPartSnapshots[newVessel.protoVessel.rootIndex].missionID : newVessel.rootPart.missionID;
+                // However, packed and newVessel.protoVessel == null means this
+                // is a ship that has been newly launched or loaded by
+                // Extraplanetary Launchpads for processing, so newVessel.rootPart
+                // is guaranteed to be valid.
+                uint mid = (newVessel.packed && newVessel.protoVessel != null) ? newVessel.protoVessel.protoPartSnapshots[newVessel.protoVessel.rootIndex].missionID : newVessel.rootPart.missionID;

                 Log.Debug("ProfileManager.OnVesselCreate: new vessel mission id = " + mid);

[edit]After getting some sleep, I realized the real problem: vessel.packed does not mean that vessel.rootPart is invalid. It means only that the vessel is on rails (eg, timewarp). It is !vessel.loaded (ie, vessel.loaded is false) that indicates vessel.rootPart is invalid.

Edited by taniwha
Link to comment
Share on other sites
xEvilReeperx

xEvilReeperx

Posted
  • Author
Posted
xEvilReeperx This patch fixes the NRE thrown when EL loads a .craft file for determining the vessel cost. It should also make SA more robust on normal launches as the exact same condition (packed, protoVessel is null) happens then, too (I discovered this while writing KerbalStats).

Thanks. Newly-spawned asteroids were what lead to my code and it seems they lead me to a wrong assumption :)

Link to comment
Share on other sites
Probus

Probus

Posted
Posted

Love your Science Alert Mod. Been playing with it since you first created it.

Lately it has been doing something strange every so often. When I hit EVA Report in SA it "throws" my kerbal a good distance away from the vessel and puts him in a orbit with greatly decreased horizontal velocity. Its only happened twice out of about the last 10 times, but that is a significant amount. Do you already know about this problem/conflict or would you like me to research it a bit for you?

Link to comment
Share on other sites
Thunderous Echo

Thunderous Echo

Posted
Posted (edited)

  • Added feature: Blizzy's toolbar made optional

How do you turn it to Blizzy's Toolbar Mode?

EDIT: Figured it out for myself.

For anyone else with this problem:

ToolbarInterface

{

ToolbarType = BlizzyToolbar

}

Maybe there should be some kind of guide on the settings file?

Edited by Thunderous Echo
Link to comment
Share on other sites
Schmonzo

Schmonzo

Posted
Posted
How do you turn it to Blizzy's Toolbar Mode?

EDIT: Figured it out for myself.

For anyone else with this problem:

ToolbarInterface

{

ToolbarType = BlizzyToolbar

}

Maybe there should be some kind of guide on the settings file?

you can also change it ingame by right clicking the science alert button in the regular toolbar. After doing that you are shown the options menu.

Also science alert does not seem to work for EVAs in low orbit for me, anyone else got that problem or did I just mess up my settings?

Link to comment
Share on other sites
xEvilReeperx

xEvilReeperx

Posted
  • Author
Posted
Also science alert does not seem to work for EVAs in low orbit for me, anyone else got that problem or did I just mess up my settings?

Any chance you have enabled SCANsat integration? If so, experiments with results that vary by biome will be ignored unless you have the relevant biome data scanned already.

Link to comment
Share on other sites
Wisq

Wisq

Posted
Posted
Any chance you have enabled SCANsat integration? If so, experiments with results that vary by biome will be ignored unless you have the relevant biome data scanned already.

Not the original reporter, but FYI, I found I ended up disabling SCANsat integration because it was failing to recognise some rarer biomes on the Mun, even though I recall that I definitely had the whole thing 100% biome-scanned. Unfortunately I don't have more details; I'll try reenabling it and report any issues, once the ExtraplanetaryLaunchpads compatibility is fixed.

Link to comment
Share on other sites
Schmonzo

Schmonzo

Posted
Posted
Any chance you have enabled SCANsat integration? If so, experiments with results that vary by biome will be ignored unless you have the relevant biome data scanned already.

jupp, thanks! Makes sense too I guess, no science spam till I scanned the buggers :)

Link to comment
Share on other sites
Ph34rb0t

Ph34rb0t

Posted
Posted

H/U for those running ScienceAlert and the stuff from Tarsier: Hubertnnn did a quick workaround fix for the .dll.

http://forum.kerbalspaceprogram.com/threads/62941-v0-24-Tarsier-Space-Technology-%28v4-5c-Now-with-GALAXIES!!!%29?p=1515634&viewfull=1#post1515634

ScienceAlert loads now and even KerbalAlarmClock now correctly saves any alarms.

Link to comment
Share on other sites
WooDzor

WooDzor

Posted
Posted

So I've been having a bit of a problem with the EVA science function lately. Quite often it crashes my game (as in I can't save, can't go back to the menu). Yesterday I had the most bizarre thing happening to me. I was with Bill and Bob on minmus and they were driving around in their little rover. I decided to check on the MEM, and suddenly it told me I had some EVA todo. Great! press the EVA button; suddenly I have 2 Bobs :-S Might this be related to the same problem as Ph34rb0t is reffering to; where by Science Alert looks at an older KAC safe?

Let me know if I you want me to try to replicate the bug again.

W

Link to comment
Share on other sites
allista

allista

Posted
Posted

Good day, xEvilReeperx. Thanks for the great mod!

I want to redirect a bug-report my user made recently, because, as I found, it concerns ScienceAlert.

Like ExtraplanetaryLaunchpad my mod uses ShipConstruct.PutShipToGround -> ShipConstruct.AssembleForLaunch routine to spawn vessels. And when it happens, ScienceAlertProfileManager crashes with NullRef. in OnVesselCreate handler:


NullReferenceException: Object reference not set to an instance of an object
  at ScienceAlert.ScienceAlertProfileManager.OnVesselCreate (.Vessel newVessel) [0x00000] in <filename unknown>:0 
  at EventData`1[Vessel].Fire (.Vessel data) [0x00000] in <filename unknown>:0 
  at Vessel.Initialize (Boolean fromShipAssembly) [0x00000] in <filename unknown>:0 
  at ShipConstruction.AssembleForLaunch (.ShipConstruct ship, System.String landedAt, System.String flagURL, .Game sceneState, .VesselCrewManifest crewManifest) [0x00000] in <filename unknown>:0 
  at AtHangar.HangarStorage+<convert_constructs_to_vessels>c__Iterator0.MoveNext () [0x00000] in <filename unknown>:0

The problem is reproducible only in career mode, of course. In my mod this causes a ship stored in editor to disappear upon launch. And I would guess that the same thing would happen with ships built with ExLaunchpads.

Link to comment
Share on other sites
  • 2 weeks later...
Wisq

Wisq

Posted
Posted

Any chance of seeing a new release in time for the holidays that addresses the EPL/Hangar incompatibility bug? I would really love to do a holiday break playthrough using both EPL and ScienceAlert at the same time.

Link to comment
Share on other sites
xEvilReeperx

xEvilReeperx

Posted
  • Author
Posted
Any chance of seeing a new release in time for the holidays that addresses the EPL/Hangar incompatibility bug?

Yep, updated OP. This release only fixes that compatibility bug; some other features I've been working on aren't quite ready yet.

Also, now that it's been out a while what do you guys think of vessel profiles? I've received surprisingly little feedback on that so far. If you have a suggestion or criticism don't be shy :)

Link to comment
Share on other sites
Wisq

Wisq

Posted
Posted

Awesome, thanks!

Re: profiles, I haven't had a chance to investigate further, but the last time I used ScienceAlert (prior to reporting the EPL incompatibility issue), I had issues with profile saving/loading not seeming to work. Generally, each time I quit & restart KSP, I had to adjust everything to my liking by hand. But with the incompatibility fixed, I'll see if I'm still having the issue and try to give an actually-useful report if it's not. :)

Link to comment
Share on other sites
Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...