Today while in chronocam, I had a crash with some error messages and a backtrace was output to the console (see attached).
Build version: 67523 Build ID: 4a235b9b44228fa788949998eaeddd10d8aa982f Graphics drivers: Nvidia (proprietary) v337.25
Same build as above, then trying to load the following replay results in a crash. LobbyID: 17004037682445174220 PS. If you can view the replay (I believe it's the right one), watch after planet smash as it does not kill all units/factories.
Thanks for ID! I'm going to check it with open source drivers and say if it's crash for me or not! Would be cool if any other Nvidia user able to check that goo.
Hm, it's work weird as always and can't find it. Not sure how, but I'll try to can try to request exactly this replay.
Also had the same thing happening now - some T2 building stayed there with very little health, after a minute or so my mouse wasn't working anymore so I restarted the game and tried to reconnect, but it would segfault every time.
I was actually able to play that game to the end. It was only when I went back to load the replay (was going to report the smash bug), when I found it crashes. But as of today I've also started experiencing crashes upon reconnects. I've attached one such crash-dump (same build as above).
Since this is an easy to reproduce, I thought I'd include a backtrace in case it helps: Code: [13:03:18.600] INFO onStateChanged PlanetState::UpdateFlags::TargetPlanet [13:03:18.600] INFO onStateChanged PlanetState::UpdateFlags::CollisionImminent [13:03:18.617] INFO onStateChanged PlanetState::UpdateFlags::Dead [13:03:18.617] INFO onStateChanged PlanetState::UpdateFlags::CollisionImminent [13:03:18.618] INFO onStateChanged PlanetState::UpdateFlags::*GroundHeightChanged Program received signal SIGSEGV, Segmentation fault. client::ClientPlanet::onStateChanged (this=0x7f15c20, flags=<optimized out>, eval_params=...) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/client/client_planet.cpp:220 220 /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/client/client_planet.cpp: No such file or directory. (gdb) bt #0 client::ClientPlanet::onStateChanged (this=0x7f15c20, flags=<optimized out>, eval_params=...) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/client/client_planet.cpp:220 #1 0x00000000005b7a0a in updateNote (dead_objects=..., future_work=..., past_work=..., paused=false, eval_params=..., note=0x8851aa0, this=0x7de6860) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/libs/paent/client_viewpoint.cpp:429 #2 ClientViewpointImpl::update (this=0x7de6860, wallclock_dt=0.114082336, eval_params=...) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/libs/paent/client_viewpoint.cpp:318 #3 0x0000000000707d32 in ViewpointController::update (this=this@entry=0x7de69e0, dt=dt@entry=0.114082336) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/libs/history/history_viewer.cpp:284 #4 0x00000000007083a6 in HistoryViewerImpl::update (this=0x739c4f0, dt=0.114082336) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/libs/history/history_viewer.cpp:124 #5 0x00000000004f2ef1 in client::ConnectionToServer::updateHistoryViewer (this=this@entry=0x821eff0, dt=dt@entry=0.114082336) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/client/client_connection.cpp:328 #6 0x00000000004f4528 in client::ConnectionToServer::update (this=0x821eff0, dt=dt@entry=0.114082336) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/client/client_connection.cpp:100 #7 0x000000000043bb56 in client::ClientGame::update (this=0xde31f0, dt=0.114082336) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/client/client_game.cpp:1356 #8 0x0000000000418f43 in main (argc=<optimized out>, argv=<optimized out>) at /var/lib/jenkins/jobs/PA_Live_Linux/workspace/pa/live/engine/game/darwin_linux_game_main.cpp:85 (gdb)
That replay now loads fine with the latest build (68331). Great work guys, I'm guessing it was the "persistent buffer mapping" issue? I look forward to playing some matches and further testing the stability.
While watching a (another) replay I experienced a crash. The following message was repeated a number of times before the crash (don't know if it's the cause): Code: ERROR WorldAudioDriver::createPlanetSound - failed to create sound [/SE/Weapons/air/fighter_fire] Edit2: I've now seen the above message with games that didn't crash so it's unlikely linked to the crash. Unfortunately, I was unable to reproduce this issue upon subsistent viewings of the replay (edit1: I did experience this again on another replay so have attached that as well now). Before it crashes the game freeze for a couple of seconds. I used to think this was just it attempting to upload the crashdump but sometimes I get theses freezes and the game recovers without a crash. Build version: 68331 Build ID: 5539803697b65f9e053af4b5a4be6b64398ac8de Graphics drivers: Nvidia (proprietary) v337.25
I just tried to play my first game with the current/above build and had a crash joining the lobby and again in the early stages of the game (a few of minutes in).
I was watching a replay of the above game to check if it'd crash at the same point without luck. It did crash at a later point (also not repeatable) and strangely after that it would crash every time upon loading the replay. After waiting several minutes it would load just fine again. In case you want to check server logs for the issue: LobbyID: 3587406168769051662
Nice to know I'm not alone. Please do include some system info, it might help in tracking down the cause.
I feel myself a bit dumb, but how did you manage to view it? I always can't get it in list, but I managed to load replay by patching "self.viewSelectedReplay" with ID specified. Though I still unable to get client crash. At what point it's should crash?
The game's completely stable for you? What display device are you using? I'll be upgrading my system shortly and so will also be able to test on Intel graphics. Apart from #33 which was temporary, all other crashes under 68331 seem to be random. :/
I didn't have time to watch it for too long. E.g I just checked different part of game in different location for about 5 minutes. It's was using Intel HD4600 on Kubuntu 14.04, 3.16 kernel and Oibaf PPA. Though I have some particle system glitch that only looks to appear with recent drivers, but client didn't crashed.
at the end of the match, the review game option pops up. Shortly after I select that and try to move the world the UI crashes. Also, I can see the replay viewer in the main menu. The latest games are displayed I can click a game and replay. That does not always crash.
