Fedora 10 - Random Lock up - Page 11

YaKillaCJ · #**151** 9th March 2009, 01:45 PM

Quote:

Originally Posted by MuadNu

YaKillaCJ, as I posted above, an alternative solution (and maybe better since it allows you to run F10 with acpi enabled) is to downgrade the kernel. I did it and hadn't have problems since. If you are interested I can tell you how to do it.

Well every kernel Ive worked wit since F9 would still freeze up. I couldnt get F8 disk to even boot in2 any menus. I upgrade the kernels regularly on F9 thru 10 and keep at least 2 previous 1s to see if 1 is more stable then the other. No go.

MuadNu · #**152** 9th March 2009, 01:52 PM

Quote:

Originally Posted by YaKillaCJ

Well every kernel Ive worked wit since F9 would still freeze up. I couldnt get F8 disk to even boot in2 any menus. I upgrade the kernels regularly on F9 thru 10 and keep at least 2 previous 1s to see if 1 is more stable then the other. No go.

I'm no expert, but if that is the case (and since as it seems you can't install other distros) then I think your issue may be different from what's happening to most of the people in this thread. In any case, do you remember getting freezes with F10 before running any updates? And what is the oldest kernel that you still have?

bbobbo · #**153** 15th March 2009, 05:58 AM

Quote:

Originally Posted by JonathanR

I'd like dmesg, /var/log/messages, and xorg logs. The time frame helps.

not sure if you're still interested in more examples, but here are my logs:

dmesg: http://pastebin.com/fbd4f1e6
messages: http://pastebin.com/f3eacae25
xorg: http://pastebin.com/f7c8e2740

basically, i think i can force a freeze if i leave firefox open overnight. here is the time sequence of events for the freeze represented in these logs:

1) reboot machine on Mar 14 02:38:14
2) start firefox
3) go to bed
4) when i look at my screen in the morning, everything seems fine--clock is still running normally
5) try clicking on firefox, click on kde menu, etc--screen and clock freeze, but mouse is still active
6) ctrl-alt-backspace, and then ctrl-alt-del to reboot on Mar 14 09:55:07
7) the reboot doesn't complete because partitions can't umount (see picture below)
8) at this point, i do a physical reboot

i will try regressing my kernel to see if that fixes the problem. however, since it seems to be an acpi problem, would disabling all power saving features in bios also temporarily fix the freezes? my motherboard (gigabyte ga-ex58-ud4p) has three different power savings options in the bios, two of which are enabled by default (i'm using the default settings):

Quote:

CPU Enhanced Halt (C1E) (Note)
Enables or disables Intel® CPU Enhanced Halt (C1E) function, a CPU power-saving function in
system halt state. When enabled, the CPU core frequency and voltage will be reduced during
system halt state to decrease power consumption. (Default: Enabled)

Quote:

C3/C6/C7 State Support (Note)
Allows you to determine whether to let the CPU enter C3/C6/C7 mode in system halt state. When
enabled, the CPU core frequency and voltage will be reduced during system halt state to decrease
power consumption. The C3/C6/C7 state is a more enhanced power-saving state than C1.
(Default: Disabled)

Quote:

CPU EIST Function (Note)
Enables or disables Enhanced Intel SpeedStep Technology (EIST). Depending on CPU loading,
Intel® EIST technology can dynamically and effectively lower the CPU voltage and core frequency
to decrease average power consumption and heat production. (Default: Enabled)

or is this completely unrelated to acpi...?

MuadNu · #**154** 19th March 2009, 05:30 PM

I received a PM asking about how to downgrade the kernel, but since it may be of interest to other people I'm posting the answer here. I did it following JonathanR's suggestions in a couple of the previous messages. Here it goes step-by-step, though I may be missing something since I don't remember exactly all the sequence (let me know if so):

Install Smart Package Manager:
Code:
```
#yum install smart-gui
```
Open smart-gui (Applications->System Tools->Smart...) and then use the search tool to look for all the packages containing "kernel". Then choose reinstall for the packages "kernel", "kernel-devel" and "kernel-firmware" matching the kernel version you want to use.
This should leave your system working with the downgraded kernel. But now every time you run yum-update it will offer you to upgrade back the kernel. To avoid this you need to install yum-versionlock:
Code:
```
#yum install yum-versionlock
```
Edit /etc/yum/pluginconf.d/versionlock.conf and make sure that it has a line saying "enabled = 1".
Finally edit /etc/yum/pluginconf.d/versionlock.list and add the packages you want to keep. In my case the file contains
Code:
```
kernel-2.6.27.5-117.fc10
kernel-devel-2.6.27.5-117.fc10
kernel-firmware-2.6.27.5-117.fc10
```

Hope it helps and it works!

bbobbo · #**155** 19th March 2009, 05:45 PM

Quote:

Originally Posted by MuadNu

I received a PM asking about how to downgrade the kernel, but since it may be of interest to other people I'm posting the answer here. I did it following JonathanR's suggestions in a couple of the previous messages. Here it goes step-by-step, though I may be missing something since I don't remember exactly all the sequence (let me know if so):

Install Smart Package Manager:
Code:
```
#yum install smart-gui
```
Open smart-gui (Applications->System Tools->Smart...) and then use the search tool to look for all the packages containing "kernel". Then choose reinstall for the packages "kernel", "kernel-devel" and "kernel-firmware" matching the kernel version you want to use.
This should leave your system working with the downgraded kernel. But now every time you run yum-update it will offer you to upgrade back the kernel. To avoid this you need to install yum-versionlock:
Code:
```
#yum install yum-versionlock
```
Edit /etc/yum/pluginconf.d/versionlock.conf and make sure that it has a line saying "enabled = 1".
Finally edit /etc/yum/pluginconf.d/versionlock.list and add the packages you want to keep. In my case the file contains
Code:
```
kernel-2.6.27.5-117.fc10
kernel-devel-2.6.27.5-117.fc10
kernel-firmware-2.6.27.5-117.fc10
```

Hope it helps and it works!

just to note, after installing smart-gui i had to do "File -> Update Channels" in order for the older kernels to be listed.

update: since regressing to kernel 2.6.27.5-117.fc10.x86_64, i've been up and running for over 4 days without a single freeze.

Code:

# uptime
 12:42:02 up 4 days, 10:13, 21 users,  load average: 0.51, 0.35, 0.20

Cassanova · #**156** 19th March 2009, 08:34 PM

Quote:

Originally Posted by tkoco

This probably is not the problem, but just the same, open a terminal window, and do a CLI "free" command. Make sure that you have allocated sufficient 'swap' space for the software loading (services with programs running). I had the same thing happen and it turned out that I did not have any swap space. Of course, I rectified the situation.

ummm how'd you do that... this often happens to me when I'm running multiple things at once. This may be my issue, but I'm a bit afraid to mess with partitions... last time I did, terrible things happened.

System (dual boot WIN Vista/Fedora)
Gigabyte Mobo: MA770DS3
AMD Athlon 64 X2 6400+ 3.2 GHz (dual core obviously)
Sapphire ATI Radeon HD 4850
4 GB DDR2 RAM @ 800 MHz

Perhaps overkill for Linux... but I built it with Windows games (Specifically Assassin's Creed) in mind

infernosoft · #**157** 20th March 2009, 05:56 AM

Quote:

Originally Posted by MuadNu

I received a PM asking about how to downgrade the kernel, but since it may be of interest to other people I'm posting the answer here. I did it following JonathanR's suggestions in a couple of the previous messages. Here it goes step-by-step, though I may be missing something since I don't remember exactly all the sequence (let me know if so):

Install Smart Package Manager:
Code:
```
#yum install smart-gui
```
Open smart-gui (Applications->System Tools->Smart...) and then use the search tool to look for all the packages containing "kernel". Then choose reinstall for the packages "kernel", "kernel-devel" and "kernel-firmware" matching the kernel version you want to use.
This should leave your system working with the downgraded kernel. But now every time you run yum-update it will offer you to upgrade back the kernel. To avoid this you need to install yum-versionlock:
Code:
```
#yum install yum-versionlock
```
Edit /etc/yum/pluginconf.d/versionlock.conf and make sure that it has a line saying "enabled = 1".
Finally edit /etc/yum/pluginconf.d/versionlock.list and add the packages you want to keep. In my case the file contains
Code:
```
kernel-2.6.27.5-117.fc10
kernel-devel-2.6.27.5-117.fc10
kernel-firmware-2.6.27.5-117.fc10
```

Hope it helps and it works!

Sweet tutorial! Thanks a lot. The downgrade worked, but for me, led to a series of other problems. I'll stay with what I had before, and hope F11 will be better!

Thanks though, I really appreciate it! :-)

KolorGuild · #**158** 20th March 2009, 07:21 AM

Hey,

I am a noob to Linux. I just recently installed Fedora 10 on my laptop. But I experience the same crashes that has been mentioned in this thread. I've been reading through this thread but I will admit it's all gone over my head.

The setup is as follows:
Intel 1.6GHz Celeron M processor 420
ATi Mobility Xpress 200M G-card
512MB RAM

Has there been a patch / update release that fixes this problem yet? Or could someone PM me the exact steps I have to take in trying to correct it. By exact I mean Terminal commands etc. Sorry to be a pain but I am a noob but very intrigued with Linux (I'm an Apple user that recently was forced to use PC's). So any help would be great!

MuadNu · #**159** 20th March 2009, 02:43 PM

Quote:

Originally Posted by KolorGuild

Hey,

I am a noob to Linux. I just recently installed Fedora 10 on my laptop. But I experience the same crashes that has been mentioned in this thread. I've been reading through this thread but I will admit it's all gone over my head.

The setup is as follows:
Intel 1.6GHz Celeron M processor 420
ATi Mobility Xpress 200M G-card
512MB RAM

Has there been a patch / update release that fixes this problem yet? Or could someone PM me the exact steps I have to take in trying to correct it. By exact I mean Terminal commands etc. Sorry to be a pain but I am a noob but very intrigued with Linux (I'm an Apple user that recently was forced to use PC's). So any help would be great!

There are two possible fixes that I'm aware of: disabling acpi and downgrading the kernel. I wouldn't be able to tell you what's the better option, you can read through the thread to get an idea, and different people have preferred different options.

Now, for the first option, you need to do the following. Open a terminal and become root (type

Code:

su -

and then enter your root password). Next type

Code:

gedit /boot/grub/menu.lst

Look for a section looking like

Code:

title Fedora (2.6.27.5-117.fc10.x86_64)
	root (hd0,6)
	kernel /boot/vmlinuz-2.6.27.5-117.fc10.x86_64 ro root=UUID=4df6f72d-bd36-40b6-963a-25b51e35a6f9 rhgb quiet
	initrd /boot/initrd-2.6.27.5-117.fc10.x86_64.img

(there may be more than one, if you have more than one kernel installed, if so go to the first such section). Then add the following at the end of the line starting with kernel: "noapic acpi=off", so that it looks like this:

Code:

kernel /boot/vmlinuz-2.6.27.5-117.fc10.x86_64 ro root=UUID=4df6f72d-bd36-40b6-963a-25b51e35a6f9 rhgb quiet noapic acpi=off

After you reboot you'll be running with acpi disabled, and hopefully with the lock ups gone.

If you want to go about downgrading the kernel, I think my previous post (#154) is more or less clear. Just open a terminal, become root as above, and then follow the instructions. And you can post back if you have any trouble.

KolorGuild · #**160** 20th March 2009, 06:09 PM

Thanks for the help. I tried the 'acpi disabled' technique first. It seemed to work but it disabled my USB devices (ie. iPod and iBurst Modem) at the same time. I only realised this after I tried to connect to the net and got this error... No idea what it means.

Quote:

Kernel failure message 1:
------------[ cut here ]------------
WARNING: at net/sched/sch_generic.c:219 dev_watchdog+0xda/0x12d()
NETDEV WATCHDOG: ib0 (): transmit timed out
Modules linked in: ipt_LOG ppp_synctty ppp_async crc_ccitt ppp_generic slhc fuse sco bridge stp bnep l2cap bluetooth sunrpc ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 dm_multipath uinput snd_hda_intel snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss snd_pcm arc4 snd_timer snd_page_alloc ecb crypto_blkcipher snd_hwdep snd yenta_socket ath5k sdhci_pci mac80211 cfg80211 tifm_7xx1 rsrc_nonstatic soundcore sdhci mmc_core tifm_core i2c_piix4 ib_usb sky2 ib_net pcspkr joydev serio_raw ata_generic pata_acpi pata_atiixp radeon drm i2c_algo_bit i2c_core [last unloaded: microcode]
Pid: 0, comm: swapper Not tainted 2.6.27.5-117.fc10.i686 #1
[<c042bb3a>] warn_slowpath+0x4b/0x6c
[<c0420900>] ? __enqueue_entity+0xa3/0xeb
[<c04223aa>] ? enqueue_entity+0x203/0x20b
[<c04223ed>] ? enqueue_task_fair+0x3b/0x3f
[<c041d92c>] ? resched_task+0x3a/0x6e
[<c06a76b3>] ? _spin_unlock_irqrestore+0x22/0x38
[<c0426237>] ? try_to_wake_up+0x221/0x22b
[<c042624c>] ? default_wake_function+0xb/0xd
[<c043cd65>] ? autoremove_wake_function+0xf/0x33
[<c06a74f1>] ? _spin_lock+0x8/0xb
[<c063e554>] dev_watchdog+0xda/0x12d
[<c06a779f>] ? _spin_lock_irq+0x1c/0x20
[<c043f35a>] ? run_hrtimer_pending+0xf3/0x105
[<c04341f5>] run_timer_softirq+0x14b/0x1bb
[<c063e47a>] ? dev_watchdog+0x0/0x12d
[<c063e47a>] ? dev_watchdog+0x0/0x12d
[<c043076f>] __do_softirq+0x84/0x109
[<c04306eb>] ? __do_softirq+0x0/0x109
[<c0405eec>] do_softirq+0x77/0xdb
[<c04303d6>] irq_exit+0x44/0x83
[<c0413ee9>] smp_apic_timer_interrupt+0x6e/0x7c
[<c0404759>] apic_timer_interrupt+0x2d/0x34
[<c04090e2>] ? mwait_idle+0x38/0x4b
[<c0402c4d>] cpu_idle+0x101/0x134
[<c0697406>] rest_init+0x4e/0x50
=======================
---[ end trace 07f7bebbcda56d94 ]---

infernosoft · #**161** 29th March 2009, 08:59 PM

Solved for now.......... i had to, SADLY, disable compiz, and now i'm using metacity.

Doesn't look as pretty, but i'm able to do EVERYTHING i wasn't able to before.

Hopefully, compiz fixes this b/c i'm 99% sure it's a compiz problem.

MuadNu · #**162** 3rd April 2009, 04:49 PM

I noticed there is a new kernel update available today (2.6.27.21-170). Does anyone know if the lock up bug was fixed?

bbobbo · #**163** 3rd April 2009, 05:20 PM

Quote:

Originally Posted by MuadNu

I noticed there is a new kernel update available today (2.6.27.21-170). Does anyone know if the lock up bug was fixed?

i noticed that too. i'm waiting for someone else to be the guinea pig since i'm going out of town for 2 weeks on sunday--i can't afford to have my machine lock-up and become inaccessible remotely while i'm away.

fortnern · #**164** 3rd April 2009, 07:48 PM

Does not fix the problem, at least on my machine :/ Out of curiosity, has anyone tried F11 beta?

MuadNu · #**165** 3rd April 2009, 07:58 PM

Quote:

Originally Posted by fortnern

Does not fix the problem, at least on my machine :/

Too bad...

Quote:

Originally Posted by fortnern

Out of curiosity, has anyone tried F11 beta?

I ran F11 for maybe half an hour from the live CD and had no problems, but that's definitely far from an exhaustive test. It would be interesting to know if it's been fixed in the beta though.