|
|
|
I've had a similar problem with 1.19.1 (4) and a couple RC versions prior and Vista. Unfortunately I dont think my crash logs ever made it to Linden Lab as the crash logger would error out when restarting the viewer. My problem is this: when starting the viewer, on occasion Vista would freeze up and only a power cycle would fix it. Often I would need to remove the battery and leave it unpowered for 30 seconds or more before the computer would even attempt to boot. During the freezup, the screen appears normal but the keyboard and mouse are not responsive at all, but the drive activity led flickers occasionally. This usually happens when starting the viewer just before it displays the login screen, but has happened at least twice when restoring an already logged-in viewer that has been minimized. I do have the latest video drivers installed that are available for my system. Prior to the last couple RC viewers, my system was quite stable and never had any problem such as this.
Details from the help screen: Second Life 1.19.1 (4) Apr 2 2008 11:59:37 (Second Life Release) You are at 286506.8, 271777.7, 381.0 in Shantos located at sim4483.agni.lindenlab.com (63.210.158.133:12035) CPU: Intel Core 2 Series Processor (1994 MHz) Pasting this from the SLDev mailing list, from davep@lindenlab.com:
I think this is being caused by the extra impostor updates that are happening in 1.20. Unstable render pipe + more passes through the pipe = more crashes. Try disabling avatar impostors and see if that helps. Moving forward, we'll be looking at removing all GL dependencies from the newview project and creating an API agnostic interface to the graphics card. This is an early step in that direction, hot on the heels of the use of gGL.begin/gGL.end vs. glBegin/glEnd. I feel this flicker I reported may be related to the crashing. Others have reported similar flickering and momentary freezes or random crashes like I have experienced.
Could someone affected by this bug please try disabling avatar imposters and report what affect this has? It's our current leading hypothesis (see davep email, quoted above) though we haven't been able to repro the bug.
I've imported the ticket internally and attached it to the branch 1.20 QA flow. Are you sure this is a Vista only issue? I crash quite a lot with 1.20 on XP, and that just by rotating the cam around my avatar a bit. I have the feeling this is even worse when SL is still loading textures right after the login:
Login, textures still loading (lots still grey), rotate by alt-click and mousemove -> crash. Tillie – we're seeing this for XP as well in the crash reports. The common element is that the crashes all occur in the nVidia driver dll itself – frustrating for crash research because we don't get a call stack.
Please try turning off avatar imposters and note any change in crash frequency. Thanks, Periapse I turned off avatar impostures as suggested and have not crashed in over an hour since logging on. Seems your suspicions might be correct.
I'm not sure if this is similar to my issue, but the moment I tried running 1.20, it immediately crashes when I login. I see my avatar for about 2 seconds and then it just crashes and sends an error report.
I may reinstall later and try again, but it seemed pretty consistant to me. This is XP Pro SP2 with a Nvidia Geforce 8800 GTS running 169.21 (latest at time of posting). Ok this is more than just a software or driver issue: this burned out my LCD screens power supply entirely the other night. LL now owes me 8 bucks plus shipping and handling.
turning off avatar imposters did not prevent the camming crash from occuring.
I too am having this issue, I tried turning off the Avatar Impostures, and had good results for over an hour, then it locked up and crashed again. When I hard booted my machine, I brought SL back up and it locked again inside of 5 minutes. I am also having the same issue in the normal viewer since the update. If anyone knows of a good 3rd party client to use, please let me know, as I am now flushing my concierge money down the perverbial toilet.
Second Life 1.20.0 (84432) Apr 9 2008 03:59:09 (Second Life Release Candidate) CPU: Intel Core 2 Series Processor (2399 MHz) Also having this issue here. Turning Avatar Impostors off has not helped. The issue appears to be much worse in areas with a large number of avis, as I can go an hour or two without crashing if I avoid busy places. Worth noting though, that aside from the crashes - framerate and overall performance seems to be improved immensely from the last version - but the crashes are so frequent I've been forced into using the release version again. The crashes seem to happen only when I'm rotating/zooming the camera - and as Tillie mentioned, especially if the area is still rezzing and filled with grey objects.
Second Life 1.20.0 (84432) Apr 9 2008 03:59:09 (Second Life Release Candidate) CPU: Intel Core 2 Series Processor (2200 MHz) I experienced the same sort of crash on my first and only time using this Dazzle/RC version: a black screen for several seconds, followed bya notification from OS (Windows Vista) that the graphics device driver had crashed. My video system is also nVidia.
Second Life 1.20.0 (84432) Apr 9 2008 03:59:09 (Second Life Release Candidate) The crash occurred about 5 minutes after logging in, when I was doing nothing in particular – standing still and engaging in an IM conversation in an otherwise empty sim, or nearly so. Certainly no one else was even remotely near or in view, so the rendering of avatar impostors should not have been a factor. I wasn't even moving, or cam'ing around. The secondlife.log file doesn't show anything extraordinary at its end – it just cuts off abruptly in the midst of logging a series of INFO: process_improved_im events. Nope, was too premature in saying that turning off Avatar Imposters solves problem. This definitely does NOT solve it.
Now also getting C++ runtime error message reported when the client crashes after the video driver restart is reported by Vista. Also, there seems to be no crash logger activity when I restart the client after that.
I decided to trip down all my settings and see how long I would make it inworld, after about 40 minutes of stepping from Low to Ultra and back again, I decided to do a little building, as this is when the lock happens for me most of the time (That may be because I spend a considerable amount of time doing that, I don't know.) I upped my object mesh from the Low settings and these are the graphic settings I had right before I locked up today. NOTE: Atmospheric Shaders and Avatar Impostors were unchecked.
Just to be clear, this issue is regarding driver instability.. client crashes are unrelated. Did your display blank out Phree? (I am also getting client crashes once i've disabled imposters and at. shaders, but the driver is stable)
Foock, I posted here because a number of people were reporting the total locking of the system. I need to hard boot my machine when this issue comes up. I do not know what the root cause is. My machine just stops working and does not return. One moment, I am able to interact with the client, and the next I cannot. No movement, no mouse, nothing...just a static picture of where it stopped working with no access to windows or the task manager. If this is an unrelated issue, I apologize. I just do not with to record a duplicate problem, as they sounded related. I also am having the warning flickering before this event occurs.
Phree2Be, i apologise, I wasn't being accusatory, I just wanted to establish the nature of your crashing. When your machine freezes, do you notice any graphics clitching? weird lines at the top or bottom of the screen which shouldn't be there?
No apologies necessary, Foock. I'm pretty new to the jira, so I'm just getting my feet wet. I don't know all the protocol yet. Graphically, the screen appears to flicker when moving before the event occurs. I have also noticed, while building, it seems when I pan the camera around what I am working on, the focused prim/object "lags" in it's rotation slightly as compared to the rest of the world. I don't know if this is a related issue or not. I am not getting any lines or bars on the screen though. I am working with an SLI rig as well. I don't think my specs listed before reflect that.
Foock, I see some errors relating to Second Life, but logged as many times as this has happened. (ex. Faulting application SecondLifeReleaseCandidate.exe, version 1.20.0.18896, time stamp 0x47fca255, faulting module SecondLifeReleaseCandidate.exe, version 1.20.0.18896, time stamp 0x47fca255, exception code 0xc0000005, fault offset 0x0064a4e5, process id 0x658, application start time 0x01c89c2d0016c921.). Is this what you are talking about? I don't usually play around in the event viewer, not sure what I'm looking for.
I am so happy this issue has been reported. Prior to being forced to go to RC1.20, I have had a very stable viewer with almost no hard crashes. But as soon as I upgraded to 1.20, the problems stated.
I have done some investigation on the root cause as one of the sims that constantly crashes / freezes me has a mega prim sphere on it. When I removed the megaprim, the freezing stopped. This was not a problem with any of my previous versions of the Viewer. But I noticed that even with the megaprim removed, I still am crashing on this building skybox sim (yesterday I got the graphics card not responding crash about 15 times). this skybox has a lot of sculpties on it but it doesnt seem to be their simple presence - possible the increased # of them. I have an HP Pavilion Laptop (brand new) with Vista and running the latest driver on a NVidia GeForce 8400 graphics adapter. This "graphics driver not responding" error only happens when in secondlife and although since I started SL in January I have had a couple of these errors, in RC1.20 it has become a horrid part of life. tonight I will try turning off the imposter and the shading to see if the problem goes away. PLEASE FIX THIS!!! Thank you all for the detailed and useful input. As this crash spans several nVidia cards, and internally we still don't have a repro, your observations here are our best clues for diagnosing and fixing the problem.
I'll keep this group up to date on the progress of the investigation. Not had any crashes for a few months (lucky me), until I started to use 1.20. Now I get a crashes a few times a day.
Can't reproduce reliably. It seems to happen most in busy and/or heavily prim laden areas. Client closes and then the driver restarts. Sometimes I get this error (or similar) Client settings: System specs: Hi all...
I have disabled the "avatar imposter" and the "atmosphere shading" options and the crashing has stopped at the SIM that caused me nothing but grief! my avatar is not freezing at all and it moves smoothly and i have even tried pushing the graphics my agressively moving my avatar and camera controls. I placed the mega prim back in place (its a 40x40 sphere which is 30% translucent and has a script that spins the sphere) which always caused me freezing cycles and that is now causing me NO PROBS! I will continue using these settings to see how it goes - but I think you guys have isolated the root cause. So far things look good! Ok... to provide additional feedback....
I used my SL viewer all night. Initially as mentioned in my previous post I had both unchecked ( "avi imposter" and "atmospheric shading") and at the sim that always caused the crashing - all the issues went away. Then I left the sim and and went to an island where after about 15 minutes - I re-engaged the atmospheric shading ONLY. things continued to work well as I was jumping around multiple sims. Later last night things were starting to lag bad and I initially assumed it might be the bug cropping up so I unchecked the "atmospheric shading" which didnt change the lag. I chaulked up the lagging to general grid issues as others were complaining too. BUT I had still not crashed once. 8 hours after first logging in and with no crash... I returned to the SIM that always caused a crash. All was still good. So I experimented and Re-engaged the "atmospheric shading" I crashed within a minute of doing that. First the screen started getting unstable then blacking out then the dreaded "graphics driver not responding hit me". It clearly appears to be related to the "atmospheric shading" and I dont know if the "avatar imposter" is even a factor. But a trigger behind why its so consisent at my building sim and the "atmospheric shading" parameter might be that I am at 750Meters overlooking a vast expanse of ocean. Maybe - just a theory - this is rending overload for this parameter as there is soooo much atmoshere to deal with and the bug triggers. You SL developers know best... but I am thinking the bug has something to do with the parameter and the elevation of the location where the crash happens. I hope you all fix it because the atmospheric shading really adds to the visual effect of the SL environment and I would really not want to live life on SL without this parameter set for too long. The atmosphere looks horrid without this parameter! HOPE THAT HELPS ! A new release candidate, 1.20.1.84760 is available for download from the Test Software page:
http://secondlife.com/community/downloads-optional.php This one has avatar imposters turned off, because when this version went to QA, a couple days ago, the imposter code was suspected to be the culprit here. The imposter code was something that changed from 1.19 to 1.20, hence a likely possibility. Unfortunately the tests done by those in this JIRA indicate that the problem is more likely in the atmospheric shaders code. If you are able, please download the 1.20.1 rc and see if you are still crashing, but don't get your hopes up I am watching this JIRA, and will relay all the useful information to the devs researching this bug. I'll also report progress to you through these comments – we are going to resolve this! Thank you for your patience, and your continued efforts to help us. This issue seems to be affecting the mac as well, as 1.20 locks me up in no time.
Question though, you say avatar imposters are now disabled in 1.20.1 RC and yet when I look in the graphics preferences, it shows them enabled. Are they in fact disabled? Thanks. 1.20.1 rc makes absolutely no difference to me whatsoever. Crashes the moment I login.
I managed to notice for a split second all my IMs and offline things. Now I wish I logged on first in the normal client as I've lost those IMs now. SL never bothers to email me them anymore despite the preference being checked for it to do so. It emails me group notices, but not IMs. I wish this was handled better. As Fooke reported, with Avatar Imposters and Atmospheric Shaders disabled, 1.20 runs fine. But with one of these options on, it crashes and burns right after login. Here's the stack trace from a Solaris test. I was using svnversion 478 and the latest nvidia driver, 169.12,REV=2008.02.14.19.02.
Last night I used this latest release of the 1.20 viewer. Alhtough throughout the night I had not had 1 crash with the "video driver not responding", when my viewer arrived at two high elevation skyplatforms (over 700M) my viewer session and basically PC goes into quick frequent FREEZE UNFREEZE cycles (about 1 to 2 second internals of each state).
As soon as I leave the platform and return to earth it seemed to stop. While it was freezing on one of these platforms, I tried setting off all shading (both basic and atmospheric) and the avatar imposter was off by default. None of that helped. I was expecting that soon after the freezing started that I would get the graphics crash, but it didnt crash. just the cycle of freezings. It didnt matter if I moved the camera or the avatar... I am still trying to diagnose what is constant when I get this viewer instability. If I learn more - i will report back. I have an nVidia 8800GTS, and ran both the RC0 and the RC1. The RC0 would crash if you looked at it funny – run for ten minutes, then you'd edit something, or be typing in chat, and BOOM. Gone. Went back to the Standard viewer and hoped for a fix.
Read this article, tried turning off impostors, no help. D/l the RC1...it holds for a while, seems to work a bit better, but now, it seems to crash on a right-click select sometimes, or immediately upon login. I'll barely get in, and BOOM. Down again. No reason given. I did get a memory error once from a stack handler. I also get a lot of "the video card was reset and recovered", citing some nvddklmdrv.sys or something like that...it goes by too fast to write down. Black screen for about 20 seconds, then BOOM...crash with a "there seems to be something wrong with this progam" error. It REALLY doesn't like the nVidia driver. (which is Forceware revision 169.28, btw) Second Life 1.19.1 (4) Apr 2 2008 11:59:37 (Second Life Release) CPU: Intel Core 2 Series Processor (2399 MHz) If there's something specific you want to try, let me know. Crashed again today with the new RC of 1.20 that has the avatar impostors turned off. The crash occurred after about half an hour after login and it was immediately preceded by a series of freeze and un-freeze instances.
I am having this problem too and there doesn't seem to be a common factor; I can be walking or standing still, be in a busy place or just with one other person on an island.
I did find some updated drivers last night for my Nvidia card, so I am going to try them today. This is only for the RC viewer, the Official viewer doesn't seem to have this problem. Second Life 1.20.1 (84760) Apr 15 2008 12:13:49 (Second Life Release Candidate) CPU: Intel Core 2 Series Processor (1995 MHz) Just to give everyone on this JIRA thread an update:
Once it was recognized that the imposter code alone was not causing the problem, the developers working on this issue looked more carefully at the other things that changed between 1.19 and 1.20. We have a new branch that has several render fixes, and it is in QA now. Confidence is good that the driver instabilities will go away when this code goes live. The target is the next rc for 1.20, which should be out next week. Once again, thank you all for trying out the RC and helping us diagnose this issue. I'll post again when I have more info after QA looks at the fixes. This is still present in the latest RC2 from today. Just had a hard crash after two minutes of jerky movement.
I still have it under RC2 as well. Have noticed that it happens more frequently using long range camera operations.
RC2 is a whole lot WORSE than RC1 - I am now getting consistent crashes with same symptoms as above within 5min of logging on.
1.19.1.4 is ways better, even though it doesn't support 3D mice. The Dazzle UI (all 1.20 RC versions: RC0, RC1, RC2) has only introduced with it trouble, persistent instability and frequent crashes. 1.20 RC2 rendering is now even ugly due to lack of AA.
Confirmed here too, I had to go back to 1.19.1.4 it got so bad.
Second Life 1.19.1 (4) Apr 2 2008 11:59:37 (Second Life Release) You are at 222119.8, 256097.6, 22.2 in The Lost Furest located at sim3228.agni.lindenlab.com (216.82.20.217:13006) CPU: Intel Core 2 Series Processor (2133 MHz) Adding my Jira to this as it is a Nvidia graphics bug as well and I believe that it is connected in some way
if u want to use RC1 just go into shortcut and edit the --channel"slrc" part of the target so the "" part is different It's happening on my Mac Pro running Leopard 10.5.2 as well. I'm using the new Nvidia 8800GT that was just released for the Gen 1 Mac Pro machines, installed it today, and seeing these occasional freezes now. They last from about five to 30 seconds at a time, but nearly always recovers. I had been using the x1900xt card by ATI, and never had this problem. So, it would seem that this is a problem specific to the Nvidia cards in general, and the 8800GT in particular. Hopefully this will help you track and fix this.
I have been getting screen glitches (bars of chaos on the screen) on my computer. Atmospheric shaders and avatar importers are both off an it is set to mid detail. It usually take 10 -30 minutes for the glitch to happen but even if windowed it will cover the entire screen not just sl. But it will not immediately freeze giving me a few minutes to close sl before I have to hard restart the computer. This has happened with both RC1 and RC2 of the new 1.20 viewer. I have noticed I have a far older card then all the others here seem to have a 7300 GS nividia (damn low profile case).
specs I'm getting horrific flicker with this client, frequent crashes in the driver. Not quite as bad as some report, but the same for me too. Actually was more stable when using my 7300/7200GS than with the new 8600GTS I have installed this weekend. But they were using different NVIDIA drivers - I was using the latest quadro drivers for the 7300, which were pretty stable (though they flickered a tiny bit, not badly as long as I turned off bump/shiny). (cant use them for the 8600, bogs it down, 7300 was great though)
But now, ugh. 163.75 Forceware drivers now. Yes I know not the latest, I'm avoiding the latest ones like the plague due to the OpenGL bugs they have. additional - when I am camming the flicker is way worse than when not camming. The flicker is THE WORST when in flycam mode... like a strobe light going off. When the camera is reset to its usual position (reset view/esc) the flicker is almost nonexistent. almost. I still sometimes get flicker when I have chat windows up, inventory windows, etc. Which makes me suspect some kind of alpha blending problem.
I will agree with the comments here since the RC release of the 1.20 code on Friday the 18th. That release made the video driver crashing MUCH WORSE!!! As soon as I upgraded to it and I walked into the She Devils club, I crashed. I tried to changes settings in the advanced graphics driver - shutting off most of the parameters. No Luck - instant crash and freezing. I crashed about 6 times in attempts to get into that club. I even tried to remove any attached objects or scripts on me in case that was the problem. No luck. I could stand outside the club but as soon as I walked into the door - I crashed.
I had enough... I uninstalled the 1.20 client and I went back to the 1.19 client until hopefully this next release with the reported RENDERING fixes included is released. Then I will give it another chance. This bug is so frustrating. I so hope the developers have found the bugs that was reported they supposedly fixed in this upcoming release. Ugh. I'm sorry everyone. Apparently the attempts at fixing this, done in a special render branch, did not make it into this release. I will raise this particular issue at the Crash Hunters meeting today, and forward everything I learn on to this JIRA.
PS...
I know this is the wrong thread... but this is the first Issue I have ever "WATCHED" and I thought by "WATCHING" it, I would get emails whenever a new comment is added to this issue. In fact, after reading what it means to "WATCH" an issue, that is what is suppose to happen. Well I have not received 1 email on the email account I have registered in my SL account. Is the WATCH feature for JIRA broken? Does anyone else here get email alerts when this issue is updated? Who / how do I report an issue about JIRA itself? lol thanks for the advice in advance. Toysoldier, I've been getting the JIRA emails just fine as I've watched this issue. I suggest you log in to the SL website and check the email address on your account (Do you get offline IMs delivered to that address?). If it all looks good, then submit a support ticket (not a JIRA).
If you're using Gmail I've noticed that it started sending JIRA watching messages to my spam folder even though I never marked one as spam, so you might want to check there.
yes... I am getting offline IMs from SL - ton of them!
So I know my email account in SL is set up properly. I also looked into my email system and I do not have a spam filtering folder. I checked. I simply do not get JIRA alerts. I was going to open a SL support ticket and then it hit me - i am likely not get WATCHED alerts because I have a BASIC SL account and I bet that basic account holders do not get email alert service from JIRA. If this is the case then it is a bit ironic since my involvement in JIRA is assisting LL as much as its assisting me. this just further discourages a Basic Account holder from participating in providing feedback to LL to diagnose problems. Anyway... sorry again for the off-topic comments... Back to the problem at hand... This issue is still proving extremely difficult to resolve. I spoke with several in engineering about it today. There are changes in our render branch that are thought will fix this, but getting that branch merged into RC proved difficult.
The branch is in QA, and the target now is to get this merged into the next RC, which will be out Wednesday. We also decided not to do two rcs a week. I know that was frustrating for you all to download rc 1, and then rc 2 only a few days later – with no change in this issue (or actually making it worse). I'll let you all know how promising rc3 looks as it gets tested. Toysoldier – JIRA email is not strictly for premium members. I've pinged someone who understands the JIRA setup and we'll try to diagnose your JIRA email problem. I've found that by rolling back to the 1.19.0.5 viewer I'm able to avoid the system freezes. So, that's the viewer I'm going to have to use until this issue is resolved. The constant freezes with the newer "Windlight" viewers are simply too much to bear. Hopefully this issue will be fixed soon.
OK Periapse,
I did the tests. The club that repeatedly causes me to crash is called the SHE DEVILS club. I went in to the club with the "bump mapping and shiny" disabled. as soon as i entered the club, my session started its frequent FREEZE UNFREEZE cycle and then shortly after it crashed. I set the parameter back on and went back to the club and it was still freezing and unfreezing real bad. and after a bit of walking around. it crashed with the video driver error. I set it off again and went back in.... it was freezing... i was just standing there and it crashed again. So, that parameter did not seem to matter. SORRY. If you have some test machines with the nVidia card and driver... I can tell you that the SHE DEVILS club constantly makes me crash. if i leave the club, the problem goes away. So... I would ask your engineers to look in detail to what is inside that club. I can stand outside the club and my viewer is stable and hasnt crashed - that tells me it is not likely something to do with the region. I know the club owners and they replace the club with different models of clubs designs - and two different home/club designs has made me crash. I think you should really analyze that club and maybe u can find out the source of whats crashing the viewer. Tell me if u want to try something else.... until then, i am going back to the 1.19 Ohh one more note... On each crash I answered yes to the SL viewer asking if it can send a crash report to SL. Look for Toysoldier Thor Crash reports between 9:30pm and 9:50pm Central time (+2 hrs from SLT) I posted on the 18th that I was having this problem too. I upgraded my Nvidia drivers; I have to use the ones supported by HP. The new ones were version Release Date: 2008-03-25 Version: 7.15.11.7432 A. This made the problem worse, even 1.19.1 4 was going black screen with the message "Display driver nvlddmkm stopped responding and has successfully recovered." The display driver might have recovered but SL couldn't !
So I found some earlier drivers Release Date: 2007-11-10 Version: 7.15.11.5665 A. I have tested them on 1.20.2, with my graphics customised to Mid and I am having no freezes or crashes !! YAY ! I don't know if it makes a difference, the newer drivers were developed ... " Enhancements: Adds support for Microsoft Windows Vista Operating Systems with Service Pack 1 (SP1)." I haven't got SP1 for Vista. Would that have an effect? Hope some of this information helps Second Life 1.20.2 (85278) Apr 17 2008 19:35:32 (Second Life Release Candidate) You are at 235558.7, 270390.7, 44.0 in Kauai Beach located at sim7007.agni.lindenlab.com (8.10.144.9:13006) CPU: Intel Core 2 Series Processor (1995 MHz) no, I'm on SP1 Vista and this problem happened to me. LL's side this problem exists on.
I have similar issues as reported by Ella McMahon:
Second Life 1.19.1 (4) Apr 2 2008 11:59:37 (Second Life Release) You are at 262578.1, 257904.6, 27.2 in Grasmere located at sim3841.agni.lindenlab.com (216.82.23.68:13005) CPU: Intel Core 2 Series Processor (2194 MHz) *) I use HP-provided nvidia drivers for my series 8 card, not the ones off nivida.com/drivers. The latest version for my system is Version: 7.15.11.7432 A Release Date: 2008-03-25 . The graphics driver control panel reports this version as Forceware 174.32. nvidia.com/drivers lists this version as still in beta; I assume that it has been made available early to HP. In any case, this is the latest driver and I am not planning to go back to the previous version (Nov 2007). It's stable for every other use as far as I can see. *) Previous to the driver update, I tested 1.20 rc0-2 and found it unstable. Stuttering, freezing and crashes after a few minutes. *) After the driver update, the main viewer 1.19.4 has become unstable for me as well. At some point, there's flickering - think of a movie played on a projector slowed down to 10fps or a sort of strobe effect - then the driver crashes and recovers. The SL client is still running but has to be shut down. *) I have not had many problems with gfx driver instability over the last few months using the HP supplied driver from Nov 2007 (I believe they were Forceware 169). *) If there are known issue with this particular driver - 174.32 - please leave a comment here. *) I will try disabling atmospheric shaders with 1.19.1.4 and see if stability returns; I will try the same with rc3. Using the Release Canidate 1.20.2 viewer
I constantly get short complete system freezes (The entire PC freezes and is unresponsive) usually the freeze lasts for about 10 seconds or less then I am back working as before. Sometimes however the last key I was pressing repeats endlessly till I press another key (so your post looks like /me starts typinggggggggggggggggggggggggggggggggggggggggggggggggggggg ) Happens at ground level while walking, flying at low level, sitting on the ground, sitting in a 700m high skypad Have tried both maxing out graphics settings & turning them down to bare minuimum. Usually happens every 3-5 minutes Log off and use the main viewer in the exact same place / few moments later, no problems all night (Main viewer however does not have the space explorer support) Windows XP2 SP2 with all patches Second Life 1.20.2 (85278) Apr 17 2008 19:35:32 (Second Life Release Candidate)
You are at 286824.5, 266194.4, 84.1 in Triglav located at sim2316.agni.lindenlab.com (216.82.17.67:13005) Second Life Server 1.20.1.85162 CPU: Intel Core 2 Series Processor (2000 MHz) I crashed about 10 times this morning, and I was trying to work in my house. Same for me. Was hanging several times, then froze, then reboot. What might interest you: with 19.1.4 I have no crashes at all. For days. (nobody believes me this)
Second Life 1.20.2 (85278) Apr 17 2008 19:35:32 (Second Life Release Candidate) You are at 286590.7, 267185.8, 61.8 in Akhsharumova located at sim4382.agni.lindenlab.com (63.210.158.32:13004) CPU: Intel Core 2 Series Processor (2666 MHz) Before commenting on this Jira make sure you have downloaded the new RC3
http://blog.secondlife.com/2008/04/23/new-release-candidate-viewer-120-rc3-available/  1.20 RC3 seems to fix all related issues for me.
Just downloaded 1.20.3 and ran it on my Intel based Mac Pro with the Nvidia 8800GT video card. Still having the periodic lockups, just as bad as before. Also, it starts out not so bad, can run for a few minutes before the first lockup, then a couple of minutes before the next... after another few minutes the lockups happen over and over and over, lockup, unlocks for a few seconds, locks up again, unlocks for a couple of seconds, locks up again. At that point you're lucky to get two seconds of actual operation between freezes. sigh. Back to 1.19.0.5 for me.
I'm REALLY looking forward to when this bug is fixed. luckily i've been a bit more fortunate than most of you here it seems, i don't have most of the problems detailed here despite having an nvidia ge force 8800GTS (which seems to be affected for others), i have been noticing the flickering for a while and also noticed that the newly implemented anit aliasing settings seem to be completely non functional, and do not even any longer take effect when the settings in game are being over ridden by graphics drivers, still completely functional in the standard client though.
rc3 is doing much better for me with nvidia 8 series, no crashing or flickering so far. Unfortunately, I had to revert the driver from Forceware 174 to 156 because of apparently unrelated instability issues that made 1.19.1.4 unusable. However, I now see a new Smart heap library error message (MEM_BAD_POINTER) after exiting with Ctrl-Q. See VWR-6389
The last messages in the log were: INFO: LLAppViewer::cleanup: Goodbye Just FYI, I downloaded and installed the "Nicholaz Edition [Eye Candy EC-e]" patch for 1.19.1.4, and have been running it for about half an hour now. I have full WindLight functionality now, and NO system freezes of any kind. So, you may want to look into what makes the Nicholaz Edition work so well, and apply what you learn from that to the next RC.
RC3 seems to be working fine for me now - is a BIG improvement.
CPU: Intel Core 2 Series Processor (2393 MHz) I have an Nividia Quadro FX 4500 and had total stability with ultra settings and far distance viewing right through the Windlight viewer period... and that included using 2 copies of the Viewer on a single computer with the -multiple flag. As soon as this was rolled into the 1.19 RC viewer and Windlight thread removed.. I immediately noticed instabilities.. and especially I ALWAYS crash the FIRST loaded viewer when the second one has logged in with a different avatar - when using the -multiple flag.
Maybe this informatio CPU: AMD 64 X2 Dual Core Processor 6000+
Memory: 4096 MB OS Version: Microsoft Windows XP x64 Edition Service Pack 2 (Build 3790) Graphics Card Vendor: NVIDIA Corporation Graphics Card: GeForce 8800 GTX/PCI/SSE2 I've been hard crashing frequently for more than a week, many times within 10 minutes of being in SL. If I work out of SL in photoshop, no problems. CPU: Intel Core 2 Series Processor (1595 MHz)
Memory: 2046 MB OS Version: Microsoft Windows Vista (Build 6000) Graphics Card Vendor: NVIDIA Corporation Graphics Card: GeForce 8600M GS/PCI/SSE2 1.20 RC3 makes me crash every time i go into appearance. I had flashing of the grafix, crashing..etc.. Stopped using the RC and went to the official release and all fine. Perhaps give RC a rest. No point in banging your head against a brick wall
I found that RC2 stopped locking up my system regularly. It would lock me up once in a great while, and I could live with it.
RC3 started out great, then started locking me up time after time. I had total system lock ups 3x in about 30 or 40 minutes last night. I went back to RC 2. See Well, given the recent comments above I'm more puzzled than ever.
RC2 was much worse for most nvidia users. But Radar found that it actually increased stability I think there are multiple crash modalities at work here, which is why the developers are having such a hard time with this one. The only good news I can report is that as I study the Crash Reports from 1.20.3 I no longer see the nvidia crash signatures bubbling up to the top of the crash list. Instead, as you can imagine, almost all crashes are coming from avatarAppearance. Whether the main crash modality has been fixed, or whether it's just hiding behind the skirt of the huge Edit Appearance crash will only be discovered with 1.20.4, which is in QA right now and should be out today. Thanks again for your patience. Hello Periapse,
I have just tried the RC3 and so far the nVdia driver probems of freezing and crashing have completely disappeared within the club that I know caused crashing and freezing immediately. There was no signs of the problem. I walked around it for about 10 minutes. I will use this client all night and see how it goes. One new bug I have noticed (actually I was told about it from a friend that detected it) on RC3 is that when you EDIT APPEARANCE of your AVI, you are kicked out of the Viewer. but I trust this will be fixed. So far So good. I will report back later tonight. CROSSING MY FINGERS! I'm having this problem with the following configuration:
CPU: Intel Pentium 4 (Unknown model) (3199 MHz) This is more stable...but I'm still experiencing freezes in 1.20 that I don't experience in 1.19 or when SL is not running. When the freezes happen, I see one, or both, of my dual core CPU monitors spike. So essentially it's freezing the whole PC.
The freezes seem to happen mostly when I'm on sky platforms and when another avatar is near. I was having some freezes when my partner was near...now that he's gone...not having any freezes like that. But was funny, cause if I looked over the edge of the platform towards the sculptie owl...was having more freezes than if I looked towards center of the platform and couldn't see the horizon. Which may have nothing to do with anything...waiting for him to log back in and see if I start freezing again. OK, he's back. And yes getting freezes again. So odd.... Second Life 1.20.3 (85643) Apr 22 2008 18:39:39 (Second Life Release Candidate) You are at 223040.1, 277283.0, 626.2 in Melioria located at sim2943.agni.lindenlab.com (216.82.19.186:13004) CPU: AMD (Unknown model) (2310 MHz) 1.20 RC4 still locks up my Macbook Pro w/ Nvidia 8600GT M solid 5 - 10 minutes after starting the program. RC2 is the only one that did not.
@Periapse Linden: My main issue was never the appearance mode thing. I found 1.20 up UNTIL 1.20 RC2 to routinely lock up my mac within minutes. Then RC 2 did not. Then RC 3 and now RC 4 would do this. This happens while walking around or using camera to zoom around. It has nothing to do with the appearance mode crash, which only affected me (and everyone else) in RC3. I'm currently using RC2 as I can't keep hard booting my Mac everytime it locks up. Hardware Overview: Model Name: MacBook Pro GeForce 8600M GT: Chipset Model: GeForce 8600M GT System Software Overview: System Version: Mac OS X 10.5.2 (9C7010) OK Periapse...
Bad news... after about 1/2 hour and going to another island... I started experiencing the common freezing unfreezing cycle... and then i got the dreaded video driver crash. I happend 2 times and I gave up. Ithen noticed a RC4 and upgraded to that. I twas very easy to go to that same place and within minutes the freezing cycle started and then without me even moiving.... video driver crash. Sorry.... tell your developers that they havent figured it out. STILL CRASHING! I am gone from SL until Sunday. I have gone back to 1.19. Toy 1.20.4 RC4: Viewer suddenly freezes then disappears/vanishes completely from Windows (no trace, even in Task Manager) and without any error message, as in VWR-6389.
1.19.1.4 is the best viewer and most stable/smooth and fastest so far! I'm afraid 1.20 RC4 is not yet ready. This is an early beta not an RC! Second Life 1.20.4 (85828) Apr 24 2008 14:59:50 (Second Life Release Candidate) You are at 286971.1, 287965.3, 24.1 in International Island located at sim3676.agni.lindenlab.com (216.82.22.157:13004) CPU: Intel Core 2 Series Processor (1795 MHz) Dear Lindens,
there are sad news, you shall know. I attach a screenshot taken from my hardware-manager. You think, this is my problem and I shall update my drivers? 1) This problem is not just mine, but very, very common. 2) Driver update does not help. 3) The only idea which shall help reads "deactivate prefetch" - but it helped me neither. Never the less there are many, many video programs on my computer which run without problems. So there shall be a way to handle this. I cross fingers, you solve this hard puzzle!! Sincerely, pls read the comment of Rita Munro
Crashing back with RC4 - RC3 was best of the RCs so far but RC4 is back to the bad old days.
CPU: Intel Core 2 Series Processor (2393 MHz) Hi Rita, I don't have it (see my device manager screenshot above-
Cheers,-- Hi MB,
thank you for the link, but it does not help - as mentioned before. If I say: "driver update does not help" - I meant that. We are not a group of fools not able to check for the lastest driver. Rita Well, the Nicholaz edition of 1.19.1.4 was quite good for a while, but then it too started locking up after a couple of hours. I turned on "Activity monitor" in OS X and noticed that when the freeze would happen then the CPU usage display would go to solid red for whatever core SL was using at the time (geez, I have four cores and Leopard 10.5.2, you'd think by now SL would be able to take advantage of the other three cores instead of maxing one and leaving the other three to simply idle...) as the %system usage spiked and the system froze.
So, anyways, back to 1.19.0.4, that's all I can really do at this point. This is beyond frustrating. I'm thinking I may end up having to pull out my 8800GT and re-install my ATI x1900XT video card, something I really really don't want to do, but it's starting to look like I may not any option. sigh Here's hoping LL can find and fix this bug soon. This is beyond annoying. SL become UNUSABLE when it TOTALLY LOCKS UP every two minutes, for anywhere from 5 to 30 seconds per lockup. Just an update for all those listening in.
The code changes in 1.20.4 did manage to fix a large percentage of the nvidia crashes, however as evidenced here, there is still a long way to go. The good news is that 1.20.4 also has a lot of diagnostic info built in. This was added because we were not able to repro these crashes reliably. Now, when you crash, we should get more data. Those of you with some developer skilz may want to check your secondlife log file after a crash. Likely you will now see an error message at the very bottom. Our developers are scanning the Crash Reporter statistics for these errors (please make sure to submit crash reports). Whereas before all we saw was a huge spike for a generic (no stack) nvidia driver crash, we now see about a dozen different crash fingerprints that each have an llerror message telling us what to look for. Current theory is that certain inworld content (legitimate, but with pathologies) is triggering the crashes. This would account for all the reports of crashes at certain places, or when in the presence of certain avatars. I feel much more hopeful that the developers are getting a handle on this truly onerous crashing – and I'll keep this group posted on further results. Tonight I will log in with the 1.20.4 viewer and generate a few crashes. I can reliably create the crash at a different location then where I use to be able to generate the crashes prior to version 1.20.3. So obviously the developers made some progress.
I am not sure where to find these crash log files in my system but I will - as usual - submit the CRASH LOG reports that I get prompted for after each crash. PS Periapse..... I am still not getting any email notifications from this watched issue. No progress on getting this glitch in Jira fixed? Thanks Toy Happened to me again - don't seem to get any crash report proccessing going on when it happens though or next start up the SL Client to log back on.
This problem is driving me nuts. You guys haven't been able to find it because it was not introduced to 1.20. It happened between 1.18 and 1.19.4.
Since I upgraded from 1.18, SL is unusable. I foolishly upgraded to 1.19 and it crashed continuously. I then tried the release candidates, 1,2,3 and 4, with no improvement whatsoever. This is a texture cache problem - and was introduced in 1.19 I've just had a breakthrough. Something told me to disable my second display output (which normally just shows the Windows Desktop) and since I've done that, I haven't crashed.
How many of you guys have a second display? Try disabling it. i partcularly like the note that "Current theory is that certain inworld content (legitimate, but with pathologies) is triggering the crashes. This would account for all the reports of crashes at certain places, or when in the presence of certain avatars. "
Now this issue started with 1.19.1.4 Until then all was fine, even over a slow network, so what was changed to provoke this? And, more important, can we expect improvements soon ? The only other thing to do is quit SL completely !! And ask for refunds everytime this occurs!! Here 1.19.1.4 is and has always been perfect; no crashes or glitches. The crashes (and sudden viewer disappearances without error msgs under XP SP3) only started in the 1.20 RC series. I hope RC5 is better.
Unfortunately Second Life 1.20 RC5 has not fixed this issue for me. I hard crashed within 15 minutes of logging in on Second Life 1.20 RC5 .
The crashing seems to be totally random, I am on my workshop and not in a laggy area or near other avatars. I'm not wearing prims or large attachments on my avatar. I have crashed from appearance, while just standing talking in local chat with my business partner or simply standing still typing in IM. I can work on my computer with no problems without second life. RC5 died on me same way after an hour of working-OK, and then again withing a minute after reconnecting.
You know, seeing fewer crash reports may not indicate that anything has been fixed, it's just that after five release candidates, nVidia owners are getting weary of crashing. I have to agree with Maggie about nvidia owners growing weary. I've been logging in to SL a lot less this week because of these problems because it is so frustrating. So I don't think less crash reports means that it's happening less, it's simply that people are so tired of constant crashing they are not logging in as often.
well, when I grow weary of crashing, I switch back to 1.19.1 ...of course, I still do crash, but the crash reports count against 1.19.1, which is presumably getting little attention because "1.20 is much better".
I just hope whoever is summarizing the stats adjusts them for battle fatigue I've tried all versions up to RC5 and crash a minute or two into running. I run everything on max settings for graphics. 1.19.4 works perfectly. Here is my specs:
______________________ I'm on Vista 64 Ultimate. I do find it discouraging to have developers deliver unstable code and then say "we still don't have a repro".
If the users knew how to crash it reliably they'd tell you. What they want to run it reliably, and that seems to be elusive. I picked this up on the blog post of the new RC5:
" Ricky Shaftoe Says: May 1st, 2008 at 7:35 PM I trust LL is working on making the RC compatible with Vista 64 and nVidia's 64-bit video drivers?" This has also happened to me. The crashing seems to be corrupting my graphics drivers and causing me to also have to reinstall my wacom tablet several times. I don't know if its useful or not but I felt I should mention it. I just downloaded the RC5 and I thought it worked very nice, but now after approximately one hour use, my whole computer freezes again.
it freezes up every 10 to 5 seconds in a lag free sim, with nothing going on around me other than sitting and chatting with my partner. I am using Nvidia Geforce 8600 M GT with driver 6.14.11.6759 I dont see that the issue is resolved with this version either. Going back to the regular client. I re-opened the internal issue because the instability is still not fixed with rc5. So that we can keep focus, please remember that this thread is specifically for the nvidia driver instability issues which arose with the 1.20 rc. Check the description to see if it fits what you've observed. For other crashes please use one of the existing JIRAs, if it describes your symptoms (
Maggie – when I say "we still don't have a repro" I mean only that LL QA is unable to reproduce this in house. Several residents on this thread have in fact posted very consistent repros, but unfortunately we can't get the same thing to happen on our machines with debuggers attached. Periapse:
LL QA can't reproduce this? We bought two machines here with nVidia 8800s specificallly because LL recommended them as well-supported....does QA not have similar machines? All their test machines aren't on the same videe card, are they? Of course...if the problem is that the debuggers are interfering, then that can't be helped. But I can't believe that if somebody from QA, experienced with SL (meaning able to drive the UI as hard as a typical experienced user does, camera swings and all) got out into the main grid with any of the recent RCs on an nVidia card and spent a day in-world, that they wouldn't see this crash. And if they did do that without crashing, then it may mean that it's sensitive to network lag. My lifepartner and I bought nearly identical machines: HP Pavilion Ultimate d4996t PC
Second Life 1.19.1 (4) Apr 2 2008 11:59:37 (Second Life Release) You are at 173188.6, 285679.0, 99.3 in Harrington located at sim4420.agni.lindenlab.com (63.210.158.70:12035) CPU: Intel Core 2 Series Processor (2399 MHz) OK, that's 1.19 because I'm avoiding RC5 after the first three crashes. We have these two machines on the end of 15/15mbps Verizon Fios service, hardwired to the router, and we're frequently in-world at the same time a lot, and often in the same sim. When we're on at the same time, we crash a lot off the RC. When Siobhan is on by herself, she can run all day with very few problems. When I come home, I often crash within the first hour off the RC...and when I do, very frequently she will then crash within 10-15 minutes. Now, some of that may have to do with grid load...it's not server load because our region is always very lightly loaded. But when one of us gets crashy, we usually both come down with it so often that it makes me wonder if a packet coming down off the server isn't overwriting storage on both machines at the same time, but the bad data doesn't trip the machines at the same instant...it's a time bomb with a very short but variable fuse. If QA can't reproduce a bug like this, QA needs beefing up. Of course, everybody's QA needs beefing up; I'm a software engineer in RL and I feel your pain. Just to let you know i have the same problem.
Secondlife.exe application error . the istruction 0x69699edd referenced to memory 0x00000004. The memory cant be Read Second Life 1.19.1 (4) Apr 2 2008 11:59:37 (Second Life Release) You are at 254486.8, 261062.0, 351.4 in Eson located at sim3912.agni.lindenlab.com (63.210.156.70:12035) CPU: Intel Pentium 4 (Unknown model) (2793 MHz) Gaia Rossini So, isolating bugs is a delicate art. We have fixed a whole bunch of GL state bugs that were causing the nVidia drivers to crash quickly and frequently. In other words, I believe that all of the issues described here initially have been addressed.
That being said, we still have plenty of bugs in SL and we are working on them in priority order. The most difficult bugs to track down are memory leak / corruption bugs. We are working on a couple of projects to help with those. In the meanwhile, if SL runs fine for around an hour and then crashes, that is almost certainly a memory bug and not a driver bug. So, if you are seeing frequent nVidia driver bugs in 1.20.5, and have updated to the latest drivers, please open a fresh bug describing those symptoms. In the meanwhile I am going to resolve this issue because the vast majority of the nVidia driver bugs that have plagued 1.20 are no longer showing up in crash reports. Thanks, Here's my PC specs:
CPU: Intel Core 2 Series Processor (2405 MHz) I had luck with reducing crashes in 1.19.1.4 by deactivating a technology called "Splendid" while using forceware 169.21 after updating to the newest forceware 174.74 beta drivers I can't remember seeing any crashes in 1.19.1.4 nor 1.20.5 (except for the MEM_BAD_POINTER) This issue is not fixed. 1.20.RC5 still locks up my macbook pro and requires a hard boot about 4 - 5 times per night. I send crash reports when I can, but 90% of the time it doesn't ask. I'm tired of hardbooting my mac. No other app locks up my UI and requires me to power off without shutting down cleanly.
It's NOT a memory leak because I'm using menumeters and watching, and the memory use is stable. Not resolved. I've posted my system specs already in this thread. The issue is not fixed.
I am still crashing every 10 minutes with the error described in my previous post. It is not a memory error ( My pc was Scanned by a computerman for 2 days) Please Fix it!!!! Gaia Rossini Just to let you kno i have the same problem. I crash when I use ALT CTRL camera.
With respect Steve this is issue is NOT fixed. I have a new computer that is only 2 months old, my system specs below and I had no issues at all before a couple of weeks ago when this nvidia issue started. I crash within 10 - 15 minutes of being in SL, sometimes more frequently whether it is RC5 or 1.19.1.
Yesterday I had my longest period in SL for weeks, ironically during a meeting with a member of LL staff, when I was on for about 2 hours before crashing. Today I am crashing about every 5 minutes or so and its not just crashes to desktop its total machine shutdown which is not doing wonders for a nice new system that works perfectly smoothly outside of SL. I was on RC5 for about 5 minutes, shut down, I restarted my computer, logged in to SL, got past the log in screen and again had a total shut down. It is really unhelpful for you to say this is fixed and frankly quite demoralising for someone like me who is trying to run a successful business at 10 minute intervals with a graphics card that LL recommended but now seem unable to support. CPU: AMD 64 X2 Dual Core Processor 6000+ On the first evening that RC5 was out, I was amazed that I went the entire night without a crash and thouht the issue might have been resolved. Unfortunately, that crash free evening was a one time good dea,l and I'm back to crashing on a routine basis, often after teleporting and using the cam to pan around. I've also noticed the situation where I get sudden prim or surface patch graphic distortions (i.e. gross enlongations) and within 30 seconds, a crash occurs.
CPU: Intel Core Series Processor (1729 MHz) I totally agree wit Sachi.
I am Italian and I know a lot of people from my country that are having the same issue ( They can not post it here because they dont speak any English. PC specs already provided in my previous posts. Excuse me...you're saying "We fixed this bug, if you're still crashing open another bug"?
Why do I get the feeling somebody wants to make the bug stats look better? The symptoms I'm seeing are indistinguishable (at the user level, at least) from the ones I had before, and they certainly match the description of this Jira incident. They frequently include "Display driver nvlddmkm stopped responding and has successfully recovered"...although I've never seen a recovery after this happens. The 1.20 viewer is still unstable with the available nVidia drivers, and the crash is fatal. In my opinion, "1.20 causes fatal driver instability with some nVidia drivers" is still not resolved until there is not instability with nVidia drivers that leads to fatal crashes that clearly call out the nVidia support. An additional point of information: with RC4 and RC5, I can have several I have 2-3 second freeezes while the audio loops tighly (mabe 1/10 second of audio over and over again) just before nvlddmkm packs it in permanently after a black screen.
Well, I went back to RC2 today. Locking up my Mac and constantly powering off/on is just insane. Something about RC2 makes this happen far, far less frequently than RC1, 3, 4, and now 5.
Also, I can't submit a crash report most of the time because it never asks.. and sometimes when it does, I don't know if it's working because I choose "yes" to submit one but the comments dialog never comes, and SL opens right up. I seriously doubt you are just seeing less of this type of crash. I think you're just not getting the crash reports. It may helps LL:
yesterday I uninstalled all SL form my Pc and I downloaded 1.18.5.3 ( following http://wiki.secondlife.com/wiki/Get_source_and_compile#Ch Then I log in with the oldier version. NO CRASH , NO fatal Error , no SL.exe error for about 14 hours. Few minutes ago I downloaded again RC5........I logged in with it.... first crash in 10 minutes So I logged in with the OLD version... and again FATAL Error!!!! And I can repeat what Radar said : Also, I can't submit a crash report most of the time because it never asks.. and sometimes when it does, I don't know if it's working because I choose "yes" to submit one but the comments dialog never comes, and SL opens right up. I seriously doubt you are just seeing less of this type of crash. I think you're just not getting the crash reports. Dear Lindens:
1) We are harsh to you, because we love this game so much and are dissappointed not to be able to play it, like we were used to. Let me say first to the new version: the new version is better than one before! 2) 3) 4) To my mind you may have programmed the error-report-sending-switch with a reset, that does not "imagine" series of crashes. This is the log of my crash report It may helps. (you will find it in Files at the top of the page)
For other Resident . It may interest you to send to the Jiira the log of your crashes (for windows XP and be aware I am using the Italian version of XP and I am translating.......my translation could be not perfect)) START It will open a window...go in Seconlife/logs -->Crash report Added my crash report log too, I am sending the crash reports every time I log into Sl but I don't know if you are getting them. They are beahving a little differently to how they used to and don't seem to be taking any time to send the logs so maybe this also is not working properly. I had thought maybe it was something you had altered but reading other comments about crash logs not working I am not so sure now.
I have noted that UNCHECHING VOICE helps.
And from my crash log : LLIMSpeakerMgr::updateSpeakers ...it works few times then i crash...is it related to IMs? I have been flipping between the latest RC 1.20.5 and 1.19.4 and I have stopped actually getting the VIDEO DRIVER NOT RESPONDING crashes in the latest RC. BUT, I have noticed over the past few days in both clients that the viewer (both version) seems to now be going into states of 1-2minute freezes. Not completely frozen as I can see the avatars still animated but no more public or IM chats and i cant move. It seems to happen about 4 - 5 times a night. This may be totally unrelated.
But as for the original problem of the display driver not responding and crashing my viewer in 1.20.x... so far for the past few days... I have not encountered it. PS - thank for the few on here that provided advise on why I am not getting email notifications on this watched issued. I have tried UNwatching and REwatching this issue and it has not helped. I simply do not get any email notifications. Toy Let me try to explain why Steve resolved this JIRA. It's not because we believe everything is fixed. It's because this JIRA has grown so huge as to be useless to a developer trying to actually solve a specific problem. What he and the developers need are focussed JIRA that cover only one crash type.
This JIRA was started for a particular crash behavior, the key symptom of which is the "driver not responding" system alert. Many who have commented understand that (see Toysoldier comment previous). But many others do not, and have commented on all sorts of other crash behaviors, muddying the waters for LL developers. My point is that all your issues are important, and that we need separate JIRAs for them so we can address them. So while I am not going to attempt to close this issue again, I strongly urge you to open specific tickets for the other crash behaviors. In JIRA, choose VWR, bug, and for Affects Version choose the appropriate viewer (RC viewers are in "unreleased"). Feel free to announce the new JIRA number here, so that the many readers of this issue can determine if they experience the same crash, and if so can add their info to the new, specific, JIRA. If you are convinced that the crash you observe must be related to this issue, go ahead and Link this issue to the new one. Additional background for those interested: The two most difficult types of crashes for us to research are crashes in the driver dlls, and crashes that force a reboot of the machine. In the first case we sometimes get a crash report, but it has no stack info, so we don't know where to look for the problem. In the second case we often do not get any crash report at all. This issue, nvidia driver instability, is both. The next viewer version, 1.21, has a great new feature, called a "watchdog". It basically oversees those other threads, and when it notices that one of them crashes it can get info out as to what the main thread was doing. This will enable us to convert many lockups / freezes into ordinary crashes that give us good information and can be readily addressed. In the meantime I will work more directly with some of the technically savvy readers on this thread (as other JIRA are opened). We may give one of you a special debug viewer and ask you to go crash it, for example. We are going to keep hammering away at the driver instability and the other JIRA until we resolve them. --Periapse Toysoldier – The Linden who checked out your problem with the email notifications has concluded that the problem is likely on the hosting end (our JIRA, public and internal, are hosted by another company). He tried to contact them about the issue last week, but I didn't follow up to see if he was successful (apparently not). I'll circle back around and see what's going on.
The temporary freezes you describe (animations continue, you can move the camera, but you can't move your av and chat doesn't update) sounds like a disconnect from a simulator. This can be due to local networking issues, heavy or overloaded sim, or, well, other bugs. The simulator and your viewer will continually try to reconnect until a timeout is reached after which the simulator logs you out. Please note if this behavior continues, and if so open a new JIRA for it. Sounds like the watchdog needs to be ported to 1.20...without it, the dump reports don't tell you what you need to know to build a releaseable 1.20. It just scares me that you folks want to release this viewer in its current state without even getting proper debug information until the next version
I'd be very willing to run your debug viewer and crash it. I'd love to see the other nVidia JIRA numbers posted here. I still get crashes in the NVIDIA driver in 1.20 RC 5. The usual "black screen, then driver has stopped responding and recovered and then SL crashes hard" sort of error. What I do not get anymore is the slowdowns, but I dont know if that is related to my having changed my video card. I had trouble with the previous video card I had (a 9600GT which had a lot of problems of its own outside of SL - so could have been a bad card, since replaced with the 8600 GTS) which slowed down over time in SL, slower... slower ... stop responding crashed, with sometimes the driver doing the black screen and recovered routine, sometimes not. Now performance is normal, just I will get the sudden crash in the nvidia driver. Sometimes go for hours not having any problem, its really erratic.
CPU: Intel Pentium 4 (Unknown model) (2799 MHz) < Pentium D 915 (sorry I entered this comment before I read the latest comments above me. I need to find the new JIRA for the failing video driver. I will leave this comment here and repeat it on the new JIRA)
OK... I spoke too soon. The RC5 viewer did give me the video driver not responding error last night. It is a lot less than the previous versions but itss there. But what has really become annoying is the FREEZING!!! God its terrible to freeze up all the time. And the 2 minute freezes where I cant move could possibly be that the network driver is freezing (being i have an NVidia driver on a laptop which I believe controls my network function as well). It may be the reason why I get the video driver not responding and the freezing of the viewer (similar to losing the network). Anyway... I cant provide any more diagnostic info as it just seems a lot more random. The SHE DEVILS club seems to be a sore spot where i encounter the problem more than i dont. The good news is that when the video driver not responding ocurred - it did seem to send a crash report. LL Team - take a look. It happened last night. I wish my laptop was older Toy! People reporting the FREEZING effect, where the viewer (usually) recovers after 30-120 seconds, might want to check to see if it's accompanied by wild, continuous disk activity. If so, it might actually be VWR-3878 at work. I see a LOT more of these freezes, and a lot longer, since I went to an nVidia 8600M with 512MB, it seems like. Perhaps because there's just that much more cache to purge.
Unfortunately, there doesn't seem to be a lot of Linden interest in VWR-3878. Ran RC6 for almost 15 minutes before the first "Display driver nvlddmkm stopped responding and has successfully recovered"
crash. thanks Coyote about VWR-3878, didn't know about that one. I think that is what was happening to my previous card (a 9600GT). And may explain why it never happened with my other computer, which is running a 7600GS with only 256 MB over an AGP bus, a far slower card with less memory, as well as slower graphics bus... could it be that the freezing bug is possibly worse with faster graphics cards? as I am not having it anywhere near as bad with the 8600GTS as it was with the 9600GT, and it has a 128 bit bus rather than the 256 bit PCI-E 2.0 bus that the 8800 and higher cards have.
could it be the faster the card, the worse that bug is? dunno. Sorry for hijacking, probably should copy this exchange in the other thread. I had tried removing my Nvidia 8800GT video card and going back to my ATI x1900XT video card, but that proved to be a TOTAL disaster in Leopard (OS X 10.5.2) - So, ended up putting the Nvidia card back in. Have been running 1.19.0.5 but tried the new RC 1.20.6 today. Same problem, intermittant lockups and freezes, with occasional total system lockups, and then after about an hour it suddenly gets REALLY bad and starts freezing HARD, to where it's nearly impossible to even log out. I have the "Activity Monitor" up and keep seeing the CPU use maxing out and all the bars for the core being used by SL turning from green to red when the freezes happen, except for the times when the freezes manage to lock up the whole system, including Activity Monitor.
Since LL seems unable to re-create and/or reproduce and then fix this issue, I suggest they go ahead and release this as the main viewer while keeping 1.19.0.5 available as an option. That way, they should end up getting a LOT of error reports from people's systems freezing up, and that will hopefully help them resolve the issue. Either that, or maybe Nicholaz will figure it out and post a patched version that doesn't freeze up. As far as theories regarding it being an issue with local servers or sims or whatever, I strongly suggest those theories are false. Why? Because when I use 1.19.0.5 things work just fine (except for never having a friends list, I've learned to live without that now). But the moment I use anything newer, like 1.19.1.4 or later, the freezes happen over and over. Updated to GeForce 174.74 nVidia driver ( http://www.nvidia.com/object/winvista_x86_174.74.html
).... have been very stable on RC6 for about four hours.
Second Life 1.20.6 (86925) May 6 2008 20:43:54 (Second Life Release Candidate) Still crashing with RC6 which is supposed to be the last RC. The crash causes the SL-screen to go black, and and error message that the graphics driver crashed.
Second Life 1.20.6 (86925) I'm running the newest official ForceWare-Version 169.25 (same happens with the previous release), haven't tried the beta version yet. Update: I'm seeing the same hard-crash to black Selavy describes.
Is there a JIRA for it, anybody know? Is there a tracking bug for 1.20 nVidia problems? Hey Periapse,
You mentioned that a new issue has been opened that fucuses on the NVidia "driver not responding" error that originally was the original reason for this issue being opened. What is this new issue #. I have been using Please advise. I'm unassigning this issue because devs just can't use it anymore.
People still suffering from the issue that began this JIRA (Windows Vista, nvidia graphics card, message "driver stopped responding and has recovered", then system hangs) should go to If you are suffering from some other crash, like the hard crash to black Selavy and Maggie describe, please do the following: Thanks, Periapse:
Your filter currently returns 335 issues. Do you seriously expect me to "look to see if my crash is already reported"? Try this driver: v175.63 with 1.20 RC6:
http://www.laptopvideo2go.com/forum/index.php?showtopic=18537
Seems to have fixed all crashes, SmartHeap errors, and silent viewer disappearances under XP. Haven't tried it under Vista. Back to stability and productivity! Thank you, MB!
The v175.63 beta driver seems to solve the problem. Hope nVidia will fix the problem in it's official version some day. CPU: Intel Core 2 Series Processor (2666 MHz) Update to my earlier post: I'm now using driver version 175.16 (a beta version) and everything seems very stable so far!
my config again: I've been having the same problem for once. I have Windows XP. In February my system crashed and did not recover. I had to do a PC Restore to get my OS up and running again. I have been getting hard blue screen crashes related to the Nvidia Graphic Driver, but this was only happening when I played SL.
Now, as of yesterday, I cannot use SL at all. When I open the viewer it comes up for a second then crashes. Then the crash log fails. I do not know what to do at this point. I've tried everything. I have a Nvidia Geforce 7300 graphic card. I'm going to try to close this issue again. Not because "the issue is fixed", but because there is no single issue in this JIRA, but a laundry list that grew over time. As I observed before, the devs who are trying to fix viewer/render bugs need useful information on particular crashes. This JIRA is just to muddied.
I let it stay re-opened so that people would have a chance to respond with comments. That has happened, but more troublesome to me is that new people are continuing to add their crashes to this JIRA. They probably haven't read this whole thread, and don't know that nothing here is actionable any more. I really do want all these different crashes to be addressed, and the only way for that to happen is for individual, focused JIRA – one for each crash. The original issue for this JIRA is being tracked in Please, if you haven't already, look for or submit JIRA with your specific crash symptoms. Notes: |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
Bug
Closed
Showstopper
1.20 causes fatal driver instability with some nVidia drivers
You might want to pass the following to the internal developers. It looks like LLRenderPass::pushBatch() may be sending corrupt/bad data to the GL library. On Solaris, we're using Nvidia's GL library and that basically a recomiple from their Linux drivers so I don't think the crash I'm encountering is local to Solaris.
In any case, once I get logged in, the viewer immediately tips over. Here's the tail end of the log:
2008-04-10T12:42:00Z WARNING: dispatch: Ignoring unknown message EstablishAgentCommunication
2008-04-10T12:42:03Z INFO: setStartupState: Startup state changing from STATE_PRECACHE to STATE_WEARABLES_WAIT
2008-04-10T12:42:03Z INFO: setStartupState: Startup state changing from STATE_WEARABLES_WAIT to STATE_CLEANUP
2008-04-10T12:42:04Z INFO: setStartupState: Startup state changing from STATE_CLEANUP to STATE_STARTED
2008-04-10T12:42:04Z INFO: idle_startup: Doing first audio_update_volume...
2008-04-10T12:42:04Z INFO: idle_startup: Done first audio_update_volume.
2008-04-10T12:42:04Z INFO: updateGeometry: WL Skydome strips in 26 batches.
Program received signal SIGSEGV, Segmentation fault.
Here's the stack trace:
0xfd0c1760 in _nv001457gl () from /usr/lib/libGLcore.so.1
(gdb) where
#0 0xfd0c1760 in _nv001457gl () from /usr/lib/libGLcore.so.1
#1 0xfd13cfbe in _nv001457gl () from /usr/lib/libGLcore.so.1
#2 0x00000008 in ?? ()
#3 0x00000441 in ?? ()
#4 0x00030101 in ?? ()
#5 0x0c7eaea0 in ?? ()
#6 0x41e756f5 in ?? ()
#7 0x3e6ac6f6 in ?? ()
#8 0xfd0c1740 in _nv001457gl () from /usr/lib/libGLcore.so.1
. . .
#167 0x085291e4 in LLRenderPass::pushBatch (this=Cannot access memory at address 0x449
)
at ... /newview/lldrawpool.cpp:440
Previous frame inner to this frame (corrupt stack?)
Since I don't have the source for libGL I'm stuck here.