View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000840 | FSSCP | --------- | public | 2006-02-27 15:39 | 2012-02-28 14:19 |
| Reporter | frankie | Assigned To | taylor | ||
| Priority | normal | Severity | crash | Reproducibility | sometimes |
| Status | resolved | Resolution | fixed | ||
| Platform | Mac PowerBook G4 | OS | OS X | OS Version | 10.4.latest |
| Product Version | 3.6.7 | ||||
| Summary | 0000840: random but repeated crashes from invalid wings | ||||
| Description | I played through FS2 official campaign without any problems. I then added the MediaVPs and started Derelict. About a dozen missions later it started crashing regularly. I have since removed the MediaVPs and deleted the cache, but the crashes continue. | ||||
| Steps To Reproduce | 1: start a mission 2: get into a battle with missiles flying, ships exploding, and usually a voiceover 3: everything locks up, then FS2 exits a minute later | ||||
| Tags | No tags attached. | ||||
|
2006-02-27 15:39
|
|
|
|
I happens with every mission or just one in particular? I do see the basic problem (there's a missing out-of-bounds check for an array) but I have no clue where it's coming from at this point. |
|
|
It happens in Derelict's "The Iron Curtain" and also in Deus Ex Machina's very first mission. I have tried many variations on level of detail, MediaVP parts, and config settings, but none of that seems to make a difference. |
|
|
it's the same bug I posted (0000812) I think... I am trying different builds and settings, and I still get that problem (the official campaign works only with nomusic flag, derelict keeps crashing even with nomusic flag) |
|
|
Sounds similar. Once the total amount of action reaches a high enough level, plus multiple audio signals and a big explosion, I get a random risk of FS2 crashing. Missions with smaller ship counts seem to work fine. edited on: 02-28-06 15:34 |
|
|
Yeah then it's deffinatly same bug, game hangs most of the time when new part of mission gets loaded ( shipi entering or command gives new orders) when near an explosion or beamfire, or music starting to play, I recently formatted my pc,but I think I encountered this the first time with a build of last year.... |
|
|
btw, what system do u have, maybe whe have some similar hardware, could be a hardware problem? my sys specs: Asus A7v8X-E deluxe mb, AMD 3200+, 2x 512 corsair valueselect dual channel RAM, ATI 9800PR (flashed to XT) (drivers tried : ATI 5.x till omega 6.2, SB live! Platinum 5.1, Samsung 160GB sATA HDD, Saitek X36 Digital Joystick, Logitech MX700 MOuse and Logitech MX keyboard. Win XP Pro SP2 All other games function normally... no overheating problems, no strange crashes... |
|
|
BB, click "View Advanced" bug details for my profile. Might give you a chuckle to see how different our rigs are. |
|
2006-03-01 17:52
|
|
|
|
This problem appears to be semi-resolved. I downloaded nightly 20060228 and added the -nomusic flag, Iron Curtain mission played through without a hitch. However, a new and different crash occurs on the next mission load. Starting new bug. |
|
|
nope, tried latest build...3-1-06, still hanging (but I got one mission further:S) this time I could go back to desktop (meaning I dont have to reset my pc, wich was the case previous times) |
|
|
Damn, spoke too soon. Same exact crash a couple missions later (Shadowplay) in both 20060206 and 20060228. 0 FS2_Open-debug 0x0004b9e0 ai_attack_wing(object*, int, int) + 556 (aicode.cpp:3264) |
|
|
Now getting a similar problem in Homesick mission 15 "The Draft". Most likely time for a crash is when I'm in the blast of an exploding enemy. Given that I have to kill at least 10 bogeys to complete the mission ... about a dozen crashes so far. However, no crash report is generated by 20060228, either normal or debug build. Here's the last few lines from the logs I do have: console.log WARNING: "Trying to add a bitmap that does not exist in stars.tbl!" at /Users/taylor/src/cvs/fs2_open.pre/projects/Xcode/../../code/starfield/starfield.cpp:2964 WARNING: "Failed to add starfield bitmap 'dneb08' to the mission!" at /Users/taylor/src/cvs/fs2_open.pre/projects/Xcode/../../code/mission/missionparse.cpp:4966 ASSERTION FAILED: "wingnum != -1" at /Users/taylor/src/cvs/fs2_open.pre/projects/Xcode/../../code/ai/aigoals.cpp:2435 fs2_open.log: Frame 0 too long!!: frametime = 0.262 (0.262) Got event GS_EVENT_ENTER_GAME (2) in state GS_STATE_WEAPON_SELECT (16) Entering game at time = 326.385 Frame 1 too long!!: frametime = 0.488 (0.488) 1225 frames executed in 30.002 seconds, 40.831 frames per second. ANI lockspin.ani with size 63x63 (1.6% wasted) vm_forward_interpolate: Bad rotationFrame 18937: Weapon 0000652 CHASING cmeasure 0001571 Frame 18992: Weapon 0000765 CHASING cmeasure 0001612 /* 32 REPEATS OF THIS MESSAGE WITH DIFFERENT NUMBERS */ Frame 21360: Weapon 0000488 CHASING cmeasure #4611 Frame 21415: Could not create debris, no more slots left Frame 21415: Could not create debris, no more slots left /* 31 REPEATS OF THIS MESSAGE WITH IDENTICAL NUMBERS */ Frame 21415: Could not create debris, no more slots left Frame 21442: Weapon 0000488 CHASING cmeasure #4903 Frame 21442: Weapon 0000634 CHASING cmeasure #4903 Frame 21442: Weapon 0000723 CHASING cmeasure #4903 Frame 21442: Weapon 0000500 CHASING cmeasure #4903 Frame 21442: Weapon 0000397 CHASING cmeasure #4903 Frame 21467: Weapon 0000397 CHASING cmeasure #4991 EOF |
|
|
The starfield warnings are safe to ignore. It's just saying that the mission is trying to use a bitmap that the game can't find. It's not fatal and will work either way. The wingnum issue is the problem though, and basically the same thing that you've had. I haven't been able to reproduce this yet to fix but it's still on the todo list. |
|
|
anything we can do to help? |
|
|
Not yet at least. I've been working on the new universal binary build for OS X and the new Linux builds so I haven't had a chance to try and track this down again. I should have time to debug again this coming weekend though. |
|
|
I've got a repeatable crash whenever the second of the two fighters escorting the pair of enemy freighters is destroyed. I'm attaching a debug log. |
|
2006-03-08 17:05
|
|
|
|
Sorry, I should've mentioned... Derelict: Shadowplay 20060308 UB build Mac OS X 10.4 |
|
|
Aha, here's a testcase that should make this much easier to debug. No gameplay required. Warzone campaign's 4th mission (Cleansing Fire) crashes with the exact same ai_attack_wing error ... during mission load! =================== STARTING LEVEL LOAD ================== Reassigning player to squadron 108th Someone passed an extension to bm_load for file 'wraiths108.pcx' ANI Loading.ani with size 26x26 (18.8% wasted) Starting model page in... Beginning level bitmap paging... loading warp model 5632 SHOCKWAVE => Loading default shockwave model... SHOCKWAVE => Default model load: FAILED!! Falling back to 2D effect... MISSION LOAD: 'WarzoneE1M4.fs2' Hmmm... Extension passed to mission_load... Reassigning player to squadron 108th Someone passed an extension to bm_load for file 'wraiths108.pcx' Using alternate ship type name : Jamming Device Starting mission message count : 203 Ending mission message count : 204 Current soundtrack set to -1 in event_music_reset_choices Current soundtrack set to -1 in event_music_set_soundtrack Loading model 'corvette2t-01.pof' IBX: Found a good IBX to read for 'corvette2t-01.pof'. IBX-DEBUG => POF checksum: 2323280796, IBX checksum: 2287035310 -- "corvette2t-01.pof" Loading model 'cargo06.pof' IBX: Found a good IBX to read for 'cargo06.pof'. IBX-DEBUG => POF checksum: 1932983486, IBX checksum: 714305600 -- "cargo06.pof" Loading model 'freighter06.pof' IBX: Found a good IBX to read for 'freighter06.pof'. IBX-DEBUG => POF checksum: 1148147902, IBX checksum: 2985564716 -- "freighter06.pof" Loading model 'freighter07.pof' IBX: Found a good IBX to read for 'freighter07.pof'. IBX-DEBUG => POF checksum: 896595278, IBX checksum: 2689747922 -- "freighter07.pof" Loading model 'Platform2S-01.pof' IBX: Found a good IBX to read for 'Platform2S-01.pof'. IBX-DEBUG => POF checksum: 2390238059, IBX checksum: 3996802094 -- "Platform2S-01.pof" Loading model 'bomber2t-01.pof' IBX: Found a good IBX to read for 'bomber2t-01.pof'. IBX-DEBUG => POF checksum: 1352596829, IBX checksum: 856117191 -- "bomber2t-01.pof" Loading model 'fighter2t-03.pof' IBX: Found a good IBX to read for 'fighter2t-03.pof'. IBX-DEBUG => POF checksum: 59332783, IBX checksum: 1004083819 -- "fighter2t-03.pof" Loading model 'fighter11.pof' IBX: Found a good IBX to read for 'fighter11.pof'. IBX-DEBUG => POF checksum: 1221070737, IBX checksum: 2073696110 -- "fighter11.pof" (the identical crash occurs here) |
|
|
I still haven't had a chance to find the underlying cause of this, but the ai_attack_wing() crash should be fixed in the next build. I'm going to have to trace back through all of the changes and figure out where this started going wrong. |
|
|
I was finally able to track this down to an earlier spot and fixed it there. I haven't had a single invalid wing crash since. It will need testing by other before I can consider it truely fixed though. A new OS X build will be available later today. |
|
|
New build is available here: http://icculus.org/~taylor/fso/nightly/20060326-OSX.dmg |
|
|
The new build won't launch for me. No errors, or anything, just nothing happens after I double-click it. |
|
|
The wrong set of Ogg and Vorbis frameworks got in at first but I updated the build when I realized that. Check your console log and see what it says. Also you can try viewing the APP contents and copying the Ogg and Vorbis frameworks to /Library/Frameworks/ and see if that fixes it. |
|
|
Re-linked, re-packaged, re-uploaded. See if it works better now. |
|
|
Works perfectly. Thanks very much. |
|
|
The build is now working, or the bug is fixed? Or both? :) |
|
|
Both! It's a lot bigger than previous builds. Does it have something they don't (aside from the bugfix)? |
|
|
Booyah! The new build played straight through my previous guaranteed crash points (Homesick 15 and Warzone 4) without a hitch. |
|
|
Bigger filesize you mean? That seems to just be because of the SDK to make a universal binary build. The main reason I don't like them in fact. The binary isn't actually "bigger" in the basic sense, it just won't let all of the debug symbols and things get stripped out the same, and it doesn't compress nearly as well (hence the bigger download). I've never been able to figure a way around either of those two things. The question has been asked several times on the Apple dev lists without any real answer, but perhaps I'll get a solution at some point. Glad to hear the crashing is resolved. I'll go ahead and resolve this bug but if you start getting something similar again just reopen and I'll take another look. |
|
|
Fixered. |
|
|
Fix committed to trunk@8535. |
|
fs2open: trunk r8535 2012-02-28 09:19 Ported: N/A Details Diff |
Diaspora code: Fix for Mantis 840 - Sounds for DRADIS, patch by m_m |
Affected Issues 0000840 |
|
| mod - /trunk/fs2_open/code/ship/ship.h | Diff File | ||
| mod - /trunk/fs2_open/code/ship/ship.cpp | Diff File | ||
| mod - /trunk/fs2_open/code/radar/radarsetup.h | Diff File | ||
| mod - /trunk/fs2_open/code/radar/radarsetup.cpp | Diff File | ||
| mod - /trunk/fs2_open/code/radar/radardradis.h | Diff File | ||
| mod - /trunk/fs2_open/code/radar/radardradis.cpp | Diff File | ||
| mod - /trunk/fs2_open/code/hud/hudparse.cpp | Diff File | ||
| mod - /trunk/fs2_open/code/hud/hud.h | Diff File | ||
| mod - /trunk/fs2_open/code/hud/hud.cpp | Diff File | ||
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2006-02-27 15:39 | frankie | New Issue | |
| 2006-02-27 15:39 | frankie | File Added: FS2_Open-debug.crash.log | |
| 2006-02-27 20:24 | taylor | Note Added: 0004966 | |
| 2006-02-28 02:34 | frankie | Note Added: 0004970 | |
| 2006-02-28 19:57 | bonusbartus | Note Added: 0004975 | |
| 2006-02-28 20:33 | frankie | Note Added: 0004976 | |
| 2006-02-28 20:34 | frankie | Note Edited: 0004976 | |
| 2006-02-28 20:54 | bonusbartus | Note Added: 0004977 | |
| 2006-02-28 21:00 | bonusbartus | Note Added: 0004978 | |
| 2006-03-01 17:42 | frankie | Note Added: 0005004 | |
| 2006-03-01 17:52 | frankie | File Added: fs2_open.log | |
| 2006-03-02 11:03 | frankie | Note Added: 0005019 | |
| 2006-03-02 21:52 | bonusbartus | Note Added: 0005028 | |
| 2006-03-03 01:36 | frankie | Note Added: 0005030 | |
| 2006-03-03 01:39 | taylor | Status | new => assigned |
| 2006-03-03 01:39 | taylor | Assigned To | => taylor |
| 2006-03-07 17:31 | frankie | Note Added: 0005069 | |
| 2006-03-07 17:44 | taylor | Note Added: 0005077 | |
| 2006-03-07 21:50 | bonusbartus | Note Added: 0005095 | |
| 2006-03-07 21:57 | taylor | Note Added: 0005098 | |
| 2006-03-08 17:04 | Muriac | Note Added: 0005107 | |
| 2006-03-08 17:05 | Muriac | File Added: fs2_open.Muriac.log | |
| 2006-03-08 17:15 | Muriac | Note Added: 0005108 | |
| 2006-03-12 09:52 | frankie | Note Added: 0005118 | |
| 2006-03-13 18:35 | taylor | Note Added: 0005119 | |
| 2006-03-13 18:37 | taylor | Summary | random but repeated crashes => random but repeated crashes from invalid wings |
| 2006-03-25 11:07 | taylor | Note Added: 0005235 | |
| 2006-03-26 09:36 | taylor | Note Added: 0005244 | |
| 2006-03-26 21:44 | Muriac | Note Added: 0005247 | |
| 2006-03-26 22:59 | taylor | Note Added: 0005249 | |
| 2006-03-26 23:46 | taylor | Note Added: 0005250 | |
| 2006-03-27 02:37 | Muriac | Note Added: 0005252 | |
| 2006-03-27 02:52 | taylor | Note Added: 0005253 | |
| 2006-03-27 02:53 | Muriac | Note Added: 0005254 | |
| 2006-03-27 03:00 | frankie | Note Added: 0005256 | |
| 2006-03-27 04:08 | taylor | Note Added: 0005258 | |
| 2006-03-27 04:08 | taylor | Status | assigned => resolved |
| 2006-03-27 04:08 | taylor | Resolution | open => fixed |
| 2006-03-27 04:08 | taylor | Note Added: 0005259 | |
| 2006-11-01 04:30 | taylor | Relationship added | has duplicate 0000845 |
| 2012-02-28 14:19 | Echelon9 | Changeset attached | => fs2open trunk r8535 |
| 2012-02-28 14:19 | Echelon9 | Note Added: 0013368 |
