I am sandboxing 12.2 and have had a number of freeze/lockups, mouse still moves about, but nothing else responds, I have to crash reboot… Or do I? Anyone know why this might happen? (I know I haven’t given much info!) Any ideas why this might be happening, all hardware is the same as a reliable system which was running 11.4, except a different HDD, This install is 12.2 KDE, KDE upgraded to 4.9.1 with zypper…
Hard to speculate on what you’ve provided.
Recommend for starters…
Inspect your syslog for errors
$ tail -n 200 /var/log/messages
If the incidents are happening intermittently,
- Note what apps are running at that time. Try to notice whether a specific app or combination of apps <may> be related.
- Install real-time monitoring apps. If your system has a fairly large amount of resources, then running realtime reporting apps which change as little as possible while reporting small graphs or text only) should not impact your performance.
- You can run something like top or htop continuously.
I’m current running on my KDE desktop the following widgets
Memory Status (immediately available)
nVidia (downloaded from KDE)
Drive I/O (requires finding, compiling from source)
If you decide to install Drive I/O
Drive I/O System Monitor Plasmoid KDE-Apps.org
Skimpy build instructions are in the notes. the 4.6 source will work for all KDE 4.6 and later. If you can’t figure out the instructions, I should be posting a better set of instructions soon (PM me if you can’t wait or people can’t find my instructions in a few days)
HTH,
TSU
I might suggest loading the nVIDIA proprietary video driver you can find here as most lockups are hardware related with video being one of the worst I think. If 11.4 is doing find, I might suggest a new video driver:
Installing the nVIDIA Video Driver the Hard Way - Blogs - openSUSE Forums
AND
LNVHW - Load NVIDIA (driver the) Hard Way from runlevel 3 - Version 1.46 - Blogs - openSUSE Forums
Thank You,
TY to you both, I am running 12.2 as provided, on my old set-up I used the nVidia drivers… So maybe nouveau does not like my HW…
Tsu2 I will have a climb through that log, running HTOP would not help, as the whole system AFAICS freezes apart from mouse, although last time I managed to drop to a console and kill bits of kde and got some function back and then logged out and back in… only really running FF, konversation, and mebbe an empty konsole when this happens, glad this is only a spare HDD and not my whole ’ iLife!’
Have you tried running the live kde cd? If this works without freezing then I would say that there is no compatibility issues with the Nouveau driver. This would lead to perhaps my next suggestion of doing a backup and installing afresh from the cd. It’s much quicker in this release (or seems to be for me) and it may cur your issues. I am not a fan of upgrading from within the OS as imho it can cause strange unexpected issues.
TY penguinclaw, I should have mentioned (phew Henk, the forum cop is not on duty!) that this is 100% fresh install from the live iso, no import of old home or ~/. files. I installed fresh to an empty HDD, then added KDE 4.9.1 repos and upped to 4.9.1
@ tsu2
As I thought this is mostly in greek to me (apologies to any Greek people watching, nothing personal!)
If you don’t mind I will paste this, seems that memory and ‘backlight’ are perhaps causing some issues?
SUSE Paste
Actually I was forced to switch back to nouveau exactly because of lockups similar to described by @wakou. In extreme cases only power button would help. Symptoms are the following message in kernel log:
Sep 27 19:11:07 opensuse kernel: [13356.883971] NVRM: Xid (0000:01:00): 13, 0003 00000000 00008297 00000f10 44480000 00000040
As well as messages similar to these in Xorg.log:
829.042] (WW) NVIDIA(0): WAIT (1, 6, 0×8000, 0x0000a4c8, 0x0000c408)
829.042] [mi] Increasing EQ size to 512 to prevent dropped events.
832.044] (WW) NVIDIA(0): WAIT (2, 6, 0×8000, 0x0000a4c8, 0x0000e348)
839.044] (WW) NVIDIA(0): WAIT (1, 6, 0×8000, 0x0000a4c8, 0x0000e348)
842.047] (WW) NVIDIA(0): WAIT (2, 6, 0×8000, 0x0000a4c8, 0x000002a8)
849.047] (WW) NVIDIA(0): WAIT (1, 6, 0×8000, 0x0000a4c8, 0x000002a8)
852.048] (WW) NVIDIA(0): WAIT (2, 6, 0×8000, 0x0000a4c8, 0x000062d8)
It looks like nVidia driver stops processing further events.
I do not care about pretty boot splash screen, so I am not affected by “login screen corrupted with nouveau” issue; but what is annoying, is very noisy fan with nouveau compared to nVidia drivers.
By the sound of it, it is hopefully fixed in 304.51 drivers
Fixed a bug that caused the X server to sometimes hang in response to input events.
That’s a topic that interests me because I’ve experienced exactly the same symptoms before and was really hoping to see them gone with all new 12.2.
I’m pretty sure this situation dates from a while back and isn’t related only to opensuse. It must be isolated cases though because the topics about it just come and go from time to time.
For me the crashes started with 12.1 when it was brand new(like a year ago). It was my first time trying linux so i didn’t feel like I can do too much about it. Thankfully I decided to give 11.4 a shot and I’ve been using it since then without any problems.
The things I’ve noticed or wondered about are:
It should be nvidia related. I haven’t seen any amd/intel users mention the same type of crashes. Nouveau worked totally fine for me all the time so it’s something in the proprietary drivers.
I personally own a fermi card(gtx 570). It’s possible that older hardware might work without any issues.
I feel like it would be helpful in tracking whether this is the case if you can mention what’s the card you’re using when sharing information.
It looks like there is something happening only with newer kernels.
11.4’s one(2.6.37) is working fine for me with the same hardware setup(and proprietary drivers) I was seeing the crashes with in 12.1. And now it seems that with 12.2 (3.4 kernels) it’s still crashing.
Does anyone know if there was some significant change in xorg from 11.4 to 12.1 that might contribute to this? I’ve read that from 12.1 to 12.2 in fact there was a change, but the crashes persist.
KDE should be fine because I’ve tried pretty much any major and minor release from 4.6 to 4.8.4 with opensuse 11.4 without any problems.
When I was trying to get my installation of opensuse 12.1 fixed I tried many of the drivers that were available then - from the repo, newest ones from the nvidia site and even some older ones. Nothing worked for me. Some of the release notes pointed to “crash issues” been fixed, but it wasn’t the case. Sadly I don’t remember any of the versions I tested with but there have been many new releases since then so it shouldn’t be relevant.
With my current 11.4 installation I’ve been just upgrading to whatever driver is in the standard repo. They all worked fine.
Right now I need my system to be stable so I can’t really help track the issue down, but on the other hand I really want to see the problem solved.
If someone is willing to do some tests and can invest the time I would suggest trying the newest nvidia driver available and then if needed grab some of the new kernels(3.5 or 3.6) and try with them.
If there’s still something wrong probably trying 2.6.37 kernel with opensuse 12.2 might solve it. Some useful results may come from trying kernels 3.4/3.5 on opensuse 11.4.
The other solution, at least for me was to stick with nouveau. Does anyone know if there have been some nice improvements with “fermi” cards(and newer - kepler and stuff)?
Lets get this thing fixed. 
Thank you. We have been experiencing these random crashes… required Power Off Re-Boot. We use VMware and just hate it when the System would lock up in the middle of one of our Homesite editing operations. Hopefully this will fix it.
Thanks again,
Chuck
Its not, i still have random freezes under Gnome 3.4 while using openSUSE 12.2
Well, we tried it in hopes… and they were dashed. The System locked up with:
2 Dolphins opened
1 Firefox with 3 tabs
1 VMware with Win2k and Homesite
2 xterms
Locked up and could not get response to anything. No keyboard, no mouse … nothing except the Power Switch.
No. We haven’t looked at the logs yet … shame on us… just to frustrated cause we were working on something important… rooting our Smartphone. Anyway, we don’t look at the logs to much … but will now. Hopefully they will have times so we can centralize the problem. After we cool down.
Thanks and take care,
Chuck
So not wishing to dash even more hopes, but I wonder if the kernel version would make any difference? Meaning, perhaps that there is more than one issue at play here. I have been using kernel 3.5.4 and even found it to be faster that 3.4 is. What if you tried a newer kernel just to see what difference it might make?
S.A.K.C. - SUSE Automated Kernel Compiler - Version 2.78 - Blogs - openSUSE Forums
Thank You,
Ok James, we’ll give it a try. Gona be a little slow at it due to other requirements -but- should be within the next 48. By the way, thanks for Grub2CMD. We used it to install this latest nVidia Driver. Our first dealings with Grub2 in full. During our testing we were still using Grub legacy. Went Grub2 full boat when we made openSuSE 12.2 our “new” Main System. Didn’t like it at first -but- now with Grub2CMD it is better.
Once again thanks,
Chuck
I was very happy to help and as always, if you have comments, problems or suggestions for grub2cmd, please let me know in my blog.
Thank You,
Yeah, same here …
9808.175] (WW) NVIDIA(0): WAIT (2, 6, 0x8000, 0x0000f92c, 0x00006bd4)
9809.098] [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
9809.098]
9809.098] Backtrace:
9809.146] 0: /usr/bin/Xorg (xorg_backtrace+0x36) [0x564616]
9809.146] 1: /usr/bin/Xorg (mieqEnqueue+0x26b) [0x5458db]
9809.146] 2: /usr/bin/Xorg (0x400000+0x4c3f2) [0x44c3f2]
9809.146] 3: /usr/bin/Xorg (xf86PostMotionEvent+0xc0) [0x482cb0]
9809.146] 4: /usr/lib64/xorg/modules/input/synaptics_drv.so (0x7fbcedccd000+0x5395) [0x7fbcedcd2395]
9809.146] 5: /usr/lib64/xorg/modules/input/synaptics_drv.so (0x7fbcedccd000+0x72b4) [0x7fbcedcd42b4]
9809.146] 6: /usr/bin/Xorg (0x400000+0x732d7) [0x4732d7]
9809.146] 7: /usr/bin/Xorg (0x400000+0x97710) [0x497710]
9809.146] 8: /lib64/libpthread.so.0 (0x7fbcf58db000+0xf140) [0x7fbcf58ea140]
9809.146] 9: linux-vdso.so.1 (__vdso_gettimeofday+0xe7) [0x7fff0d1ff9b7]
9809.146] 10: /usr/lib64/xorg/modules/drivers/nvidia_drv.so (0x7fbcef8da000+0x1043be) [0x7fbcef9de3be]
9809.146] 11: /usr/lib64/xorg/modules/drivers/nvidia_drv.so (0x7fbcef8da000+0x8f9be) [0x7fbcef9699be]
9809.146] 12: /usr/lib64/xorg/modules/drivers/nvidia_drv.so (0x7fbcef8da000+0x10832b) [0x7fbcef9e232b]
...
And as I did not have any other system for network logon - power button.
Pity … while fan speed could be tolerated, with nouveau I get random screen corruption (in small areas like menu or title bars, not full screen).
As an alternative to that, one may use the 3rd party stable kernel repo found at this address Index of /repositories/Kernel:/stable/standard , i has the latest and stable version of the kernel for opensuse. I will upgrade later on today an see if things are better 
openSUSE 12.2, Gnome 3.4, nvidia driver 304.51, Nvidia GeForce 9400GT, a few minutes ago the desktop has frozen again, even with the newest kernel available for openSUSE.
Weird thing is that nothing shows up in /var/log/messages or in Xorg.0.log about this freeze.
I have 2 different computers:
- laptop with AMD A8-3500 APU
- desktop with Intel i7-2600 and ATI FirePro card
Both system randomly freeze with 12.2 installed. And doesn’t matter whether I’m using ATI drivers that come with the kernel package or install a proprietary drivers. Therefore the crash is not video - related.
I downgraded SuSE on laptop to 12.1 and it is working without any issue.
On a desktop I tried to install a new kernel from
Index of /repositories/Kernel:/openSUSE-12.2/standard (Current version is 3.4.12-1.1)
and the system sometimes working for couple of days or may crash every 10 minutes in the row.
The best way to crash the system is to open google chat settings and select “Verify your settings”. That operation turn on web camera. Usually system freezes right away or if you unlucky enough you need to leave camera on for a little while. I saw some complaints related to web camera, however seems to be that is related to USB stack, not to the camera driver. One time I’ve got a kernel panic screen with the stack started in USB device usb_hcd. It explains why it is easy to crash the system with some USB device.
My assumption here is that you may try to unplug ALL (!) USB devices and see what the result would be. I simply cannot do that because I don’t have ps/2 keyboard and mouse at the moment and just enjoy a system freeze.