P3D v4.5 (or 4.4) Crash with AIseparation or RC4

[Quoth]

Hi

I've been getting regular crashes with the faulting module showing as FSUIPC. No warning, just the spinning blue wheel and then back to desktop.

I'm using P3D v4.5, with Aerosoft Airbus 318/319/320/321, Radar Contact, GSX Level2, UK 2000 Gatwick and AIG traffic manager. No other add-ons currently installed or active.

I've attached the FSUIPC log file and the AppCrashView report.

Hope you can help

Kind Regards

Alan

 

FSUIPC5.log

App Crash View.txt

[John Dowson]

Hi Alan,

the usual cause of such crashes is a corrupt weather file, so you first need to eliminate that as a possibility. If you have a wxstationlist.bin file in your Appdata\roaming P3D4 folder (where your PREPAR3D.cfg file is), then please remove it. It will be regenerated when you start P3D. If the file isn't there, or if the problem persists, can you try adding

NoWeatherAtAll=Yes

to the [General] section of the FSUIPC5.INI file.

If that solves your issue, then it could be another corrupt weather file somewhere..

If it doesn't solve your issue, then as you are using an unregistered version it may be something to do with Radar Contact as it looks like that is all you have that is using FSUIPC. We can look into this once the weather has been checked.

Btw, do the crashes occur in a certain area or under certain conditions, or do they seem random? what do you mean by 'regular' - on every flight?

Cheers,

John

 

[Quoth]

Hi John 

Many thanks for your reply. Been following other threads with similar problems, don’t have the .bin file you mention. I’ll try the no weather addition to the .ini file and see if that helps.

As you said, I “think” I only use fsuipc for Radar Contact and I have an unregistered version. Would it be better to upgrade to the paid version?

The crashes appear random, not at any particular point and I’ve been running the same flight plan. When I say regular it happens about 4 out 5 flights, which as you can appreciate is very frustrating  🙂

Cheers

Alan

[John Dowson]

Quote

 Would it be better to upgrade to the paid version?

The versions are the same, its just that some features are disabled in the unregistered version. If its crashing with the unregistered version, it will also crash in the registered version. Also no point upgrading unless you are going to use some of the features it brings - check the user manual for this.

Let me know how it goes with the NoWeatherAtAll flag set.

Cheers,

John

[Quoth]

Thanks John, will do. One add on I forgot I was using is chase plane...

Cheers

Alan

[Quoth]

Hi John

Sorry but same problem unfortunately, I've attached the log file and crash report.

Hope you can help...

 

Regards

Alan

Crash View.txt

FSUIPC5.log

[John Dowson]

Hi Alan,

does it also crash if you disable Radar Contact?

John

 

[Quoth]

12 minutes ago, John Dowson said:

Hi Alan,

does it also crash if you disable Radar Contact?

John

Hi John

Not tried that yet, but it didn’t crash when I disabled AI (AIM OCI) I’ll try the reverse and disable RC, still testing various combinations at the moment. Currently just have Aerosoft Airbus 318/19 package, Radar Contact and AIM OCI installed (along with fsuipc and mkrws)

To be honest, I don’t believe it’s a fsuipc problem per se, but possibly more to do with bad data being received by fsuipc if that makes sense (as you can probably tell I’m not that computer literate!)

Regards

Alan

[Pete Dowson]

10 hours ago, Quoth said:

it didn’t crash when I disabled AI (AIM OCI)

I wonder, then if perhaps there's a corrupt BGL or aircraft in that AI installation. I use the AIM OCI for traffic, with all airlines enabled and kept up to date, but I only fly in Europe so a lot of them won't be seen.

The traffic data FSUIPC receives comes from SimConnect, but that in turn will come from the BGL route information and the Aircraft.cfg for the aircraft. I suppose it is possible for something there to be corrupted.

Pete

 

[Pete Dowson]

Further to my thoughts above, I've heard from another user with pretty much exactly the same error, but in his case there were actually THREE crashes reported in the Event Viewer, all within a second or two. the first of these was Prepar3D crashing with a .NET framework error. Then there were two FSUIPC crashes.

I think the FSUIPC crashes are because separate threads are in use and the crash of P3D is cascaded into those threads before the whole process is closed as a result of the main crash. So, could you check the Event Viewer again -- see if there were actually two or three in a row. If so, it is the first (earlier) one which is truly relevant.

Now FSUIPC does not use .NET framework at all. So I don't think it is really to do with FSUIPC. Note that this other user is using AIM OCI Traffic (as am I).

He is getting this with P3D 4.5.12 (the "hotfixed" one). Also his Win10 was recently updated. These things might be relevant. I am also using 4.5.12 but I have my Win10 fixed at release 1803 (as far as I'm able, anyway).

BTW John will be back to take over this support at the end of the week.

Pete

 

[Quoth]

Hi Pete 

Many thanks for your reply. Yes I am getting 3 in a row with the first being a .net error then the 2 fsuipc errors. Also using the latest v4.5 and my win 10 was recently updated.

Sorry for my ignorance but does AIM OCI use .net? Because as I said earlier I don’t get the crash with it disabled. Guess I should take this over to the AIG board but I’ve read most of that forum and haven’t seen anything about it crashing P3d 

Anyway, many thanks for your thoughts on this and please post if you have any other ideas although I appreciate that this doesn’t appear to be a fsuipc problem.

Regards

Alan

[John Dowson]

Hi Alan,

could you try the 5.151b version posted in this thread (in second to last comment):

Its a dll file but named as a .txt - just download it, rename to FSUIPC5.dll and use it to replace the existing one in your modules folder.

Thee are only some minor changes in this version, but I would be interested to know if you still get the .NET framework error with this version (was it a COM error?)

Thanks,

John

[Ray Proudfoot]

Hi chaps. I’m the other user Pete mentioned. I also had the .NET crash with 5.151b installed.

I reverted to 4.4 (still using 5.151b) and was able to fly out of EGLL up to Manchester and later returned to Heathrow and no crashes. But that’s with P3D v4.4 and Windows 10 build 1903. There’s another user on the CTD section on AvSim who had the same three errors as me with v4.5.

https://www.avsim.com/forums/topic/554065-45-ctds/?do=findComment&comment=4012882

[Quoth]

1 hour ago, Ray Proudfoot said:

.......There’s another user on the CTD section on AvSim who had the same three errors as me with v4.5.

https://www.avsim.com/forums/topic/554065-45-ctds/?do=findComment&comment=4012882

That’d be me (Gandalf) 😁 Looking for help wherever I can. I guessed from your post on Avsim you were the other user Pete mentioned here 😉

Edited May 29, 2019 by Quoth

[Ray Proudfoot]

14 minutes ago, Quoth said:

That’d be me (Gandalf) 😁 Looking for help wherever I can. I guessed from your post on Avsim you were the other user Pete mentioned here 😉

A different alias for every forum! 😄

[Pete Dowson]

6 hours ago, Quoth said:

Sorry for my ignorance but does AIM OCI use .net?

The Manager, AIM, might do, but it isn't running. The traffic just consists of regular Traffic BGLs plus SimObjects (aircraft models and textures) -- same, in fact as any other traffic package except for the likes of UTLive which injects its traffic instead of using BGLs.

I assume that P3D uses .NET framework, but I'm not even sure about that. One could check on its list of dependencies I suppose (there are programs that can do that), but I'm not sure I'd recognize which of the Windows services it calls upon are supplied by a .NET install and which aren't -- except a few of the latter which I'm certain definitely aren't.

Pete

 

[Quoth]

On 5/28/2019 at 10:48 PM, John Dowson said:

Hi Alan,

does it also crash if you disable Radar Contact?

John

 

H John

Just completed 2 flights without Radar contact and no crashes. Want to try some more before drawing any conclusions, also want to try with RC but with AI chatter and interaction turned off.

Haven't installed the ‘b’ dll as yet...

Regards

Alan

[Ray Proudfoot]

It may be a huge coincidence but when I had that crash at EGLL I was running RC4 via WideFS. I had saved the flight at the holding point so later that day loaded it and tested it again.

On 5 separate tests I could take off without a crash. RC4 was not running. All the tests were at the same time as the original crash. Why didn’t it crash? So I decided to start a new flight and this time I did use RC4.

As the aircraft rotated P3D crashed again. It seems almost impossible that RC4 could be responsible when it only reads data from P3D. I still feel AIG Ai has a bad texture or model but thought it worth reporting. I am using the ‘b’ variant of FSUIPC5.

[John Dowson]

Now I'm back, I'm going to install both AIM (I've been meaning to do this for a while now anyway) and RC4, one at a time and together, and see if I can reproduce. Once I can do that, I can see if the exception can be handled in FSUIPC.

[Ray Proudfoot]

The crashes are definitely linked with the amount of Ai in the sim and what is being fed to RC4. I never had the problem at EGCC simply because there is only a fraction of the Ai Heathrow has.

50% at EGLL and no problems with RC4. Hope you can get to the bottom of it John. Having all these wonderful airlines is great but not if it causes crashes.

[Pete Dowson]

1 hour ago, Ray Proudfoot said:

The crashes are definitely linked with the amount of Ai in the sim

It may just be that the reduced traffic level eliminates the one with the bad texture or whatever. So a test with traffic limited instead by moving half of the AIM-OCI scenery BGLs to a "saved" folder, one half then the other, might help distinguish between the rather too simplistic "too many" (what would that mean), and there being a "bad" one.

FSUIPC in any case limits the amount that RC can see, asit only has room for 96 ground and 96 airborne in total in its tables.

Are you setting the time the same for each test, or allowing that to vary now?

Pete

 

[Ray Proudfoot]

Hi Pete, if there was a bad texture then surely it would have caused a crash when Ai was at 100% with RC4 not running. It didn’t either today or earlier in the week with all testing at EGLL.

I flew at various times today from around 11:30 to 13:00 and no crashes. Tried another two around 14:00 - 14:30 and still no crash with Ai at 100%.

Five minutes after the last crash I tried again this time with RC4 running and it crashed with 100% Ai. As soon as I reduced it to 50% and loaded the same flight it was fine. Don’t see how it could be a bad texture. Crashes have only occurred with RC4 running.

[Pete Dowson]

44 minutes ago, Ray Proudfoot said:

Don’t see how it could be a bad texture. Crashes have only occurred with RC4 running.

Perhaps, then, more related to the Aircraft definition -- RC will be reading aircraft details for the ATC exchanges it does with them. That's why I said "or whatever", not just textures.  And also why I suggested doing the IPC logging, to see what RC was doing with FSUIPC just before the crash.

With such logging the log'll get large, but there's no need for whole logs, the last entries would do, 40-100 lines or so. In fact if you know when the crash will (probably) occur, dstart a new log (click new log button) just before.

Best though to make sure no other FSUIPC-using apps are running, to cut down on logging confusion. eg. EFB2.

Pete
 

[Ray Proudfoot]

Hi Pete, I'll do as you suggest tomorrow and generate a log.  The reason I'm so sure it's related to RC is because if it was a bad texture there would be many more posts on the AIG forum and there doesn't appear to be. The only suggestion was to reduce the Ai%.

In a topic on AvSim @Quoth had crashes with Ai set to 75% so the trigger point is somewhere between 50 and 75.

[Quoth]

Hi John

Just tried a couple of flights using the same scenario as before. With AI set at 50% I had a crash with the same 3 errors reported. After much swearing I retried with 40% AI and had a crash free flight.

I’ll keep testing. I did have logging on for the successful flight, so I’ll turn it on at 50% as well. I see what you mean about large files! 😁

Regards

Alan