Ok, I have spent the night analysing.

No change brought:
- reinstalling kmods
- manual installation of NVidias most recent driver from NVIDIA
- using nouveau for a while, but it is 10 times slower (glgears) and the fan disturbs
- trying akmod

This morning I found out, that the rsyslogd is was not active. I upgraded the beta from rawhide, by the way. This had the bug, that it did not install/activate/update the syslog daemon. So I had no messages in my log file...

This morning I got some new weird messages:
Kernel tried to execute NX protected page
BUG: Unable to hangle paging request
SMP
Plymouth fb-release

Maybe this helps?

What shall I do next?

I have traces thanks to rsyslog is running again..
Last freeze:
 28 07:53:38 duter kernel: [ 5664.550221] 
May 28 07:53:38 duter kernel: [ 5664.550221] Pid: 945, comm: jbd2/dm-5-8 Tainted: G      D     2.6.38.6-27.fc15.x86_64 #1 FOXCONN RD790A01/A79A-S
May 28 07:53:38 duter kernel: [ 5664.550221] RIP: 0010:[<ffffffff812377ae>]  [<ffffffff812377ae>] __list_del_entry+0x4e/0x98
May 28 07:53:38 duter kernel: [ 5664.550221] RSP: 0018:ffff88021cecfcb0  EFLAGS: 00010207
May 28 07:53:38 duter kernel: [ 5664.550221] RAX: dead000000200200 RBX: ffff88021b67e000 RCX: ffff88021b67e000
May 28 07:53:38 duter kernel: [ 5664.550221] RDX: 3261656233343633 RSI: 3137653162343234 RDI: ffff88021b67e000
May 28 07:53:38 duter kernel: [ 5664.550221] RBP: ffff88021cecfcb0 R08: ffff8800cfd97ae8 R09: 0000000000000002
May 28 07:53:38 duter kernel: [ 5664.550221] R10: 0000000000000002 R11: 000000000000310d R12: 0000000000000014
May 28 07:53:38 duter kernel: [ 5664.550221] R13: ffff880182344c40 R14: ffff88021b970020 R15: ffff88021e3b8070
May 28 07:53:38 duter kernel: [ 5664.550221] FS:  00007f6f807a6700(0000) GS:ffff8800cfd80000(0000) knlGS:00000000f76e6870
May 28 07:53:38 duter kernel: [ 5664.550221] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
May 28 07:53:38 duter kernel: [ 5664.550221] CR2: 00007f6f98068090 CR3: 000000019390c000 CR4: 00000000000006e0
May 28 07:53:38 duter kernel: [ 5664.550221] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
May 28 07:53:38 duter kernel: [ 5664.550221] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
May 28 07:53:38 duter kernel: [ 5664.550221] Process jbd2/dm-5-8 (pid: 945, threadinfo ffff88021cece000, task ffff88021ebf5c80)
May 28 07:53:38 duter kernel: [ 5664.550221] Stack:
May 28 07:53:38 duter kernel: [ 5664.550221]  ffff88021cecfcd0 ffffffff81237806 0000000034653234 ffff8802104ea800
May 28 07:53:38 duter kernel: [ 5664.550221]  ffff88021cecfd30 ffffffff811cb2a4 ffff880182344c40 ffff88021acc4f00
May 28 07:53:38 duter kernel: [ 5664.550221]  ffff88021b67e000 0000000200000191 ffff88021cecfd30 ffff8802104ea800
May 28 07:53:38 duter kernel: [ 5664.550221] Call Trace:
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff81237806>] list_del+0xe/0x2d
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff811cb2a4>] jbd2_journal_write_revoke_records+0x11e/0x172
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff811c88ed>] jbd2_journal_commit_transaction+0x35f/0x1052
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8100804e>] ? load_TLS+0x10/0x14
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff81008714>] ? __switch_to+0xc6/0x220
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff810609eb>] ? lock_timer_base+0x2c/0x52
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8147588c>] ? _raw_spin_unlock_irqrestore+0x17/0x19
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff81060a84>] ? try_to_del_timer_sync+0x73/0x81
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff811ccd30>] kjournald2+0xc9/0x20a
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8106f212>] ? autoremove_wake_function+0x0/0x3d
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff811ccc67>] ? kjournald2+0x0/0x20a
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8106eb53>] kthread+0x84/0x8c
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8100a9e4>] kernel_thread_helper+0x4/0x10
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8106eacf>] ? kthread+0x0/0x8c
May 28 07:53:38 duter kernel: [ 5664.550221]  [<ffffffff8100a9e0>] ? kernel_thread_helper+0x0/0x10
May 28 07:53:38 duter kernel: [ 5664.550221] Code: 94 87 7c 81 be 2f 00 00 00 eb 51 48 b8 00 02 20 00 00 00 ad de 48 39 c2 75 11 49 89 d0 be 32 00 00 00 48 c7 c2 d9 87 7c 81 eb 31 
May 28 07:53:38 duter kernel: [ 5664.550221] RIP  [<ffffffff812377ae>] __list_del_entry+0x4e/0x98
May 28 07:53:38 duter kernel: [ 5664.550221]  RSP <ffff88021cecfcb0>
May 28 07:53:38 duter kernel: [ 5664.601514] ---[ end trace c9afe3ac2116924f ]---

This crash was with nouveau in GNOME3 trying to start digikam. 

So my problem seems to be not related to nvidias driver, right?

There is a last message behind that trace:

[drm] nouveau 0000:01:00.0: EvoCh 1 Mthd 0x0080 Data 0x00000000 (0x0006 0x05)

I have more traces in my log:
While booting I get many errors like:

nouveau 0000:01:00.0: VM: trapped write at 0x000112babc on ch 0 [0x00000070] BAR/PFIFO_WRITE/IN reason: PAGE_NOT_PRESENT

and finally, short before end of boot:


May 28 08:00:06 duter kernel: [   41.013618] general protection fault: 0000 [#1] SMP 
May 28 08:00:06 duter kernel: [   41.013636] last sysfs file: /sys/devices/system/cpu/cpu3/cache/index2/shared_cpu_map
May 28 08:00:06 duter kernel: [   41.013644] CPU 0 
May 28 08:00:06 duter kernel: [   41.013647] Modules linked in: nfsd lockd nfs_acl auth_rpcgss exportfs sunrpc ipt_LOG xt_iprange xt_multiport act_nat ebt_dnat ebtabl
e_nat ebtables ebt_snat nf_nat_proto_dccp nf_nat_tftp nf_nat_proto_udplite nf_nat_ftp nf_nat_proto_sctp libcrc32c nf_nat_pptp nf_nat_proto_gre nf_nat_amanda iptable_n
at nf_nat_snmp_basic nf_nat_sip nf_nat_h323 nf_nat_irc nf_nat ts_kmp nf_conntrack_amanda nf_conntrack_sip nf_conntrack_sane nf_conntrack_proto_sctp nf_conntrack_netbi
os_ns nf_conntrack_ftp nf_conntrack_pptp nf_conntrack_irc nf_conntrack_tftp nf_conntrack_proto_udplite nf_conntrack_netlink nfnetlink nf_conntrack_h323 nf_conntrack_p
roto_gre nf_conntrack_proto_dccp ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables isl6421 cx24123 cx88_dvb cx88_vp3054_i2c videobuf_dvb dvb_cor
e ir_lirc_codec lirc_dev ir_sony_decoder rc_hauppauge ir_jvc_decoder ir_rc6_decoder nouveau cx8800 ir_rc5_decoder snd_hda_codec_realtek cx8802 snd_hda_intel ir_nec_de
coder ttm cx88xx snd_usb_audio
May 28 08:00:06 duter kernel: drm_kms_helper snd_hda_codec rc_core drm r8169 tveeprom v4l2_common btcx_risc videobuf_dma_sg i2c_algo_bit snd_hwdep snd_usbmidi_lib vid
eo snd_rawmidi sp5100_tco pwc videodev videobuf_core v4l2_compat_ioctl32 snd_seq snd_seq_device snd_pcm snd_timer edac_core i2c_piix4 mii snd edac_mce_amd i2c_core k1
0temp soundcore serio_raw snd_page_alloc wmi microcode joydev ipv6 firewire_ohci firewire_core pata_acpi ata_generic crc_itu_t pata_atiixp pata_jmicron [last unloaded
: scsi_wait_scan]
May 28 08:00:06 duter kernel: [   41.013919] 
May 28 08:00:06 duter kernel: [   41.013923] Pid: 204, comm: plymouthd Not tainted 2.6.38.6-27.fc15.x86_64 #1 FOXCONN RD790A01/A79A-S
May 28 08:00:06 duter kernel: [   41.013938] RIP: 0010:[<ffffffff8125e73f>]  [<ffffffff8125e73f>] fb_release+0x26/0x57
May 28 08:00:06 duter kernel: [   41.013951] RSP: 0018:ffff88021b6ebec8  EFLAGS: 00010292
May 28 08:00:06 duter kernel: [   41.013957] RAX: 6e410a426b203020 RBX: ffff88021b980000 RCX: 0000000000000001
May 28 08:00:06 duter kernel: [   41.013964] RDX: 0000000000000000 RSI: 0000000000000055 RDI: ffff88021b980008
May 28 08:00:06 duter kernel: [   41.013971] RBP: ffff88021b6ebed8 R08: 0000000000000000 R09: 0000000000000000
May 28 08:00:06 duter kernel: [   41.013978] R10: ffff88021b428d90 R11: 0000000000000246 R12: ffff88021b980008
May 28 08:00:06 duter kernel: [   41.013986] R13: ffff88021b98ed70 R14: ffff88021b98ed70 R15: ffff88021e0f2cc0
May 28 08:00:06 duter kernel: [   41.013993] FS:  00007f5a49477720(0000) GS:ffff8800cfc00000(0000) knlGS:0000000000000000
May 28 08:00:06 duter kernel: [   41.014001] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
May 28 08:00:06 duter kernel: [   41.014006] CR2: 00007f5a4947d000 CR3: 000000021b709000 CR4: 00000000000006f0
May 28 08:00:06 duter kernel: [   41.014006] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
May 28 08:00:06 duter kernel: [   41.014006] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
May 28 08:00:06 duter kernel: [   41.014006] Process plymouthd (pid: 204, threadinfo ffff88021b6ea000, task ffff88021b630000)
May 28 08:00:06 duter kernel: [   41.014006] Stack:
May 28 08:00:06 duter kernel: [   41.014006]  ffff88021b428d80 0000000000000008 ffff88021b6ebf28 ffffffff81122b15
May 28 08:00:06 duter kernel: [   41.014006]  ffff88021b428d90 ffff88021b9ee200 ffff88021b4bd398 ffff88021b428d80
May 28 08:00:06 duter kernel: [   41.014006]  ffff88021b6382c0 0000000000000000 0000000000000000 0000000000000000
May 28 08:00:06 duter kernel: [   41.014006] Call Trace:
May 28 08:00:06 duter kernel: [   41.014006]  [<ffffffff81122b15>] fput+0x121/0x1bb
May 28 08:00:06 duter kernel: [   41.014006]  [<ffffffff8112007f>] filp_close+0x66/0x70
May 28 08:00:06 duter kernel: [   41.014006]  [<ffffffff81120126>] sys_close+0x9d/0xda
May 28 08:00:06 duter kernel: [   41.014006]  [<ffffffff81009bc2>] system_call_fastpath+0x16/0x1b
May 28 08:00:06 duter kernel: [   41.014006] Code: 5c 89 f8 5d c3 55 48 89 e5 41 54 53 66 66 66 66 90 48 8b 9e a0 00 00 00 4c 8d 63 08 4c 89 e7 e8 35 64 21 00 48 8b 8
3 b8 03 00 00 
May 28 08:00:06 duter kernel: [   41.014006] RIP  [<ffffffff8125e73f>] fb_release+0x26/0x57
May 28 08:00:06 duter kernel: [   41.014006]  RSP <ffff88021b6ebec8>
May 28 08:00:06 duter kernel: [   41.017564] ---[ end trace 237011e4926b1837 ]---

Now I have to care about our garden and my family. Please tell me, if I can perform a test for you or provide additional information.

Last time, as I tried to re-install akmod-nvidia I got really weird complaints about a missing kernel. I could not install it anymore. Finally I found out that there was a real orphan kmod-nvidia package installed.
After I removed it I could reinstall akmod-nvidia again and after it built and installed the package it runs.
I am using my system for two hours now and had no freezes nor crashs anymore.

Please don't close this bug yet, as I am not sure, why it is not happening right now. Maybe you can use the traces, I have posted...

If it is gone "for ever", I will come back and tell.

System frooze again. This time without any log entry. So my problem is still unsolved...

I worked with running top in a visible window and catched a crash. In top systemd and kthreadd where the topmost, but this might mean nothing as top refreshes the view delayed.
There was no message in the log file.

I continued to analyse and update everything that is available. Up to now, my problem is not fixed and I habe nothing new: The problem seems to be not connected to a specific user interaction. Ok, it did not occur over night, this morning I could continue to work. I notice it most of the times, in a mouse that stops working, then the harddisk starts running for a second or two and then everything is unresponsive. The LEDs on keyboard do not change anymore and non of the buttons, beside power and reset do not work anymore.

After an uptime of two days I got two freezes within ten minutes...

I'm having the same crashes with F15, they are hard crashes in that keyboard doesn't work (num/capslock lights won't go on or off when pressed) and i can't ping from another machine. all i can do is yank the power cable.

crashes are very intermittent, and don't seem to be related to load at all - mprime ran for 8 hours overnight and memtest86 ran for 4 hours without a single failure.

i've played urbanterror and compiled android and haven't seen a crash, then i do something mundane like switch firefox4 window or switch to another workspace in gnome3 and it crashes.

xorg-x11-drv-nvidia-libs-270.41.06-1.fc15.i686
kmod-nvidia-2.6.38.6-27.fc15.x86_64-270.41.06-2.fc15.x86_64
akmod-nvidia-270.41.06-2.fc15.x86_64
xorg-x11-drv-nvidia-libs-270.41.06-1.fc15.x86_64
kmod-nvidia-2.6.38.6-26.rc1.fc15.x86_64-270.41.06-1.fc15.x86_64
xorg-x11-drv-nvidia-270.41.06-1.fc15.x86_64

nvidia 9400GT graphics card.

i had similar issues in f14 but not in f13 with the same hardware.

I am not sure, if this issue is related to X or the grafics drivers.

Summarized:
- Completely sporadic, action independent freezes
- Even network is not responding anymore and PING dies normally last
- happends 5 times in a row, then hours not
- seems to happen, if computer is in low load use
- running by itself it runs for hours
- AMD Phenom IIx4 810 Processor stepping 2


User actions:
- Happends in Gnome more often than in KDE
- Browsing using firefox, but not systematic to anything, browsing mainly reading, scrolling and going in history for- and backwards
- Writing emails via Evolution (3 frozes in a row)
- Reading system logs to find the source of this problem

System actions:
- No specific repeating last message in logs before freeze
- often last message is activation of a service by dbus

I think this is something that has to do with the kernel, something "deep inside" like process scheduling, IRQ sharing etc. But I might be completely wrong...

Error messages I found lately:
hda-intel: IRQ timing workaround is activated for card #1. Suggest a bigger bdl_pos_adj

EDAC_amd64: ECC disabled in the BIOS or no ECC capability, module will not load...
(haven't found that setting in bios (yet))

Your BIOS doesn't leave a aperture memory hole. Please enable IOMMU option in the BIOS setup!

Mh. Maybe I have found something: In a thread somewhere in a forum I heard about a fix that had to do with the power saving mode of nvidia cards. The guy passed some mystic parameter values to the driver to switched of the frequency changing in low load.
This was worth a test: I use nvidia-settings - PowerMizer - preferred mode to "prefer maximum performance" and had no freezes anymore. (24h)

As the freezes are so sporadic I cannot tell if this is the real source of the problems, but I have other hints that support it: In the past I used cpuspeed to throttle my cpu down, when my computer is used for office task.
In this time I had similar freezes of my computer. They disappeared permanently when I disabled the cpuspeed service.

If this proofs right, the system seems to have problems with changing clock speed of the CPU, RAM and graphic card.

I leave my system as it is (cpuspeed of, nvidia performance on max) until Sunday, 5th. Then I will enable the nvidia clock speed again and check if the freezes re-appear.
If I have freezes earlier, I will report the failure of my theory.

i've had crashes with nouveau and nvidia-proprietary drivers so its not a gfx *driver* issue at least. the powermizer stuff doesn't work with noeveau so not convinced that's the issue though.

i don't use cpuspeed (or intel turbo boost) so i don't think its to do with changing cpu frequencies.

my next step is buy a new 9500GT, new motherboard and new RAM, if that doesn't fix it i move to debian with an ATI card!

Sorry, my point was that I used nvidia-settings to disable it. I am not sure if this feature is driver controlled, and if, if it is not yet supported by nouveau.

I also had the crashed with Nouveau. If your card has the power saving feature and if this is supported by nouveau or if this feature is driver independent, you still might have the same problem. Otherwise you have found another issue.

Information should be here: http://nouveau.freedesktop.org/wiki/PowerManagement
It is not very specific about, what is supported and what not. 

You can test it if you install akmod-nvidia and disable the performance levels as I did. Otherwise it might be good if you could open a separate thread to avoid mixing problems and confusing developers...

Crash, forget the powermizer theory...

(I have installed the latest Xorg, after a day of uptime, rebooted and got a freeze while I was away from keyboard)

you don't happen to have a realtek gigabit network card do you?

as i'm starting to wonder if its this old bug again with the r8169 drivers:

https://bugzilla.redhat.com/show_bug.cgi?id=642861

as today i had what i thought was a crash but it recovered briefly for me to look at dmesg which had a lot of link up/down messages as well as "NOHZ: local_softirq_pending 08".

i've reinstalled the nvidia drivers and added "pcie_aspm=off" to my grub to see if that helps (never did with f14).

I have the network card mentioned there...

03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 01)
04:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 01)

Now, that you say it, since the update I have sporadic problems with the network too. Firefox is slow/does not react, ekiga tells, it cannot determine network connection, situations where '/etc/init.d/network start' did help.

[    9.351761] r8169 Gigabit Ethernet driver 2.3LK-NAPI loaded
[    9.351783] r8169 0000:03:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
[    9.351875] r8169 0000:03:00.0: setting latency timer to 64
[    9.351932] r8169 0000:03:00.0: irq 44 for MSI/MSI-X
[    9.352089] r8169 0000:03:00.0: eth0: RTL8168b/8111b at 0xffffc90012e42000, 00:1f:e2:5c:cc:20, XID 18000000 IRQ 44
[    9.358083] r8169 Gigabit Ethernet driver 2.3LK-NAPI loaded
[    9.358103] r8169 0000:04:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[    9.358142] r8169 0000:04:00.0: setting latency timer to 64
[    9.358205] r8169 0000:04:00.0: irq 45 for MSI/MSI-X
[    9.358341] r8169 0000:04:00.0: eth1: RTL8168b/8111b at 0xffffc900117ae000, 00:1f:e2:5c:cc:21, XID 18000000 IRQ 45
[   22.212164] r8169 0000:03:00.0: eth0: link down
[   22.213161] r8169 0000:03:00.0: eth0: link down
[   25.272182] r8169 0000:03:00.0: eth0: link up

I have no NOHZ messages in dmesg log.


But, I have most crashes when i use gnome, less with kde and now I am trying xfce...

My logwatch agent tells me, I had 5 times:
ioremap error for 0xcffb0000-0xcffb1000, requested 0x10, got 0x0

whilst searching for a new motherboard i noticed that *every* consumer motherboard has a RTL8111D variant, so its very important that the r8169 driver works properly, and it seems its been broken since f14.

my onboard nic is:

04:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 03)

in the meantime i've bought an intel e1000 pcie nic to test.

pcie_aspm=off didn't solve it, i've just had a crash.

I have these sporadic freezes too, but I have an Atheros network card, so I think that that's probably not the problem. There are no tracks in my log files, but there was one kernel crash, reported by oops: 

WARNING: at net/sched/sch_generic.c:256 dev_watchdog+0x108/0x17c()
Hardware name: TW9/SW9
NETDEV WATCHDOG: em2 (atl1c): transmit queue 0 timed out
Modules linked in: fuse 8021q garp stp llc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device microcode snd_pcm uvcvideo videodev v4l2_compat_ioctl32 snd_timer intel_ips atl1c snd iTCO_wdt soundcore iTCO_vendor_support wmi snd_page_alloc i2c_i801 ipv6 xts gf128mul dm_crypt nouveau ttm drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: scsi_wait_scan]
Pid: 0, comm: kworker/0:1 Not tainted 2.6.38.6-27.fc15.x86_64 #1
Call Trace:
 <IRQ>  [<ffffffff8105510e>] warn_slowpath_common+0x83/0x9b
 [<ffffffff810551c9>] warn_slowpath_fmt+0x46/0x48
 [<ffffffff813db4c5>] ? netif_tx_lock+0x4a/0x7c
 [<ffffffff813db653>] dev_watchdog+0x108/0x17c
 [<ffffffff810609bd>] ? internal_add_timer+0xbc/0xbe
 [<ffffffff8106115c>] ? cascade+0x65/0x81
 [<ffffffff8106131c>] run_timer_softirq+0x1a4/0x266
 [<ffffffff8100fc99>] ? paravirt_read_tsc+0x9/0xd
 [<ffffffff813db54b>] ? dev_watchdog+0x0/0x17c
 [<ffffffff8105ae40>] __do_softirq+0xd2/0x19d
 [<ffffffff8100fc99>] ? paravirt_read_tsc+0x9/0xd
 [<ffffffff8100aadc>] call_softirq+0x1c/0x30
 [<ffffffff8100c101>] do_softirq+0x46/0x81
 [<ffffffff8105afc4>] irq_exit+0x49/0x8b
 [<ffffffff8147bf5b>] smp_apic_timer_interrupt+0x7e/0x8c
 [<ffffffff8100a593>] apic_timer_interrupt+0x13/0x20
 <EOI>  [<ffffffff8100fc99>] ? paravirt_read_tsc+0x9/0xd
 [<ffffffff81274687>] ? intel_idle+0xdb/0x100
 [<ffffffff81274666>] ? intel_idle+0xba/0x100
 [<ffffffff81398e14>] cpuidle_idle_call+0xe7/0x166
 [<ffffffff81008321>] cpu_idle+0xa5/0xdf
 [<ffffffff81464dfe>] start_secondary+0x20c/0x20e

It might not be relevant, I don't know. 

Moreover I observed that the cpu fan raises after most freezes.

i've got that (only once) in my logs too:

[   30.155718] ioremap error for 0xdf7e2000-0xdf7e3000, requested 0x10, got 0x0
[   32.545193] r8169 0000:04:00.0: eth0: link up
[   32.545715] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready

daniel, what graphics card, motherboard etc do you have?

(In reply to comment #0)
> Description of problem:
> Since I have installed Fedora 15 I experience sporadic freezes, in detail
> neither mouse, not keyboard nor network work anymore. I have to reset or
> shutdown my computer hard to get it running again.

I had exactly the same issue, using an x86_64 virtualbox guest (i.e. no funny driver involved). That was pretty bad: just messing with a few terminals, firefox and anjuta, the system would freeze within 30 minutes at best. This vbox guest had 4 cores allocated. Running it with one core only, I have been using it for the entire day without issue. Setting it back to four cores; I was back with the freezes. I am positive, at least for my setup which is pretty much f15 out of the box on a 'virtualbox hardware', that it is an smp race condition somewhere. Cant find no crash or errlog nowhere. Yum updated today (including the new xorg).

I wonder, why we get no request for detail from the developers, no feedback at all.
Dear kernel experts: Is there a way to get debug output from the moment it freezes? Use the debug enabled kernel? Is there a hidden switch? Is there something that we can test, can help?

Maybe we can get a hint, what the freezes cause, by running 'top' with extremely short delay.
I run it in a terminal window with option "Always in foreground' and set its refresh rate to 0.1s. So I can see the last working task after the freeze. It is important to set the rate to a very small value, to be close enough to the event.

In my last case, it showed 'gnome-shell'. With this I don't want to say, the other window managers don't have this issue, I had freezes in KDE and XFCE too, but I had them only a very few times.

Please the 'top' test on your systems too, maybe we can trace the problem down a bit...

gnome-shell is pretty much constantly at the top of my "top" list.

Today:
13 min. uptime: gnome-shell on top, no user actions
50 min. uptime: Xorg on top, no user actions
22 min. uptime: firefox, then Xorg, then gnome-shell, freeze after small mouse movement
~50 min uptime: freeze while monitor was in power save, no output of top available

Then I gave up, trying to work something and enjoyed the hot weather..

Hopefully someone finds the problem very soon. I need a reliable system...

For my part, the freezes outside of gnome shell are gone. Seemingly.

I am working now with xfce again and had no freezes since the last kernel update anymore.
My attempt with gnome mail ended in a freeze. I will retry it every time I see an important update comming in and update this issue.

So at least for me, this issue is now one duplicate to the many of gnome 3...

I have the same issue on my T61. Fedora 15 64 bit. My video card is 
01:00.0 VGA compatible controller: nVidia Corporation G84M [Quadro NVS 140M] (rev a1)

Can't notice anything particular in var log messages. The only applications running is normally firefox. The system totally freezes with a hard reboot as the only option.

We are running a pilot of fedora15 at our workplace, and are experiencing lockups which sound identical to these; i.e. no logs, no mouse move, no option but to reboot.  It's occurred under both nvidia and nouveau.  We have seen it on all 3 of the PCs that are running F15.  It also happened on my Dell XPS1210 laptop today, which has Intel 945 integrated graphics.

It does *not* sound like this gnome bug, because the mouse cursor freezes too unlike the gnome bug:

https://bugzilla.gnome.org/show_bug.cgi?id=650857

We cannot correlate the freeze based on app use, system load, etc.  Seems to be random.  Can go for a couple of days w/o a lockup, then encounter 3 over the next hour.

I have very similar freezes on my laptop. It's an intel gm965 chipset, gma 3100 video with the stock intel driver, and intel iwl3945 wifi. Freezes mainly happens during video playback (flash video or gnome mplayer) but sometimes after reboot it freezes after gdm login. I've tried to run gnome in fallback mode, but it doesn't help. Nothing is in the logs. Has anybody any idea about this, how to debug this?

I also reverted back to fallback mode, and have experienced freezes.

These threads from the forums seem to report the same thing:

http://www.fedoraforum.org/forum/showthread.php?t=264892
http://www.fedoraforum.org/forum/showthread.php?t=264191

(In reply to comment #30)
> I have very similar freezes on my laptop. It's an intel gm965 chipset, gma 3100
> video with the stock intel driver, and intel iwl3945 wifi. Freezes mainly
> happens during video playback (flash video or gnome mplayer) but sometimes
> after reboot it freezes after gdm login. I've tried to run gnome in fallback
> mode, but it doesn't help. Nothing is in the logs. Has anybody any idea about
> this, how to debug this?

I've found out that one of my RAM modules went wrong, and that had caused the freezes on my system. Sorry for the false report, please ignore my previous comment.

An update from me. Last night I ran memtest and it discovered NO errors. So my RAM seems to be fine.
After the last updates my problem has changed: The freezes are rare now. In the last two weeks or so I had no non 3D related freezes anymore. I can do my office work again without hard reboots.
BUT when I use flightgear or any other 3D game, I still get freezes after some varying minutes of gameplay.

I have noticed, that nvidia has corrected some freeze errors in their driver I think that some of them will have caused my former problems. And as I seem to have 3d issues only the remaining issue will be in this area too.

Getting these too, and it looks like it might be a local graphics (Nvidia driver/desktop?) problem in my case.

If I boot, login and then do nothing locally it's steady as a rock.  I can run servers (I use BIND, Sendmail, Apache, OpenSSH, VSFTPd and Samba) and hammer all of those without any lockups.  I even tried running a remote X11 desktop from a Windows client running XMing and was able to use KDE, Firefox etc. without any problems.

Once I start manipulating the local desktop however (I've tried KDE and Gnome) then I get the lockups.  It might take a day or so, or it might happen in less than a minute after the desktop appears.  Tried installing the proprietary NVidia driver from RPMFusion today but got a script error on install, so I'm back on Nouveau and random crashes for now...

Hate to add to the problem, but I have an ATI HD3650 card and I am experiencing these freezes, where the only way out is hit reset, or power off. I am usually listening to music when it happens and the player will keep on going with the current track but will stop after the track completes. Same thing with the printer, I was printing a 200 page document which did complete the job but did not print the next print job. 

I was experiencing this originally with Gnome3 Shell and then did a reinstall of F15 with KDE. I have tried disabling Desktop Effects and that seemed to have stopped it until today, I've froze 3 times today. It seem to have started around the 27th of August after the KDE update to 4.6.5-5 There isn't any indication of a problem in any of the logs and it does not seem to be any specific app or load factor.

Also I too have the RTL 8111/8168B Ethernet controller.

Everything is stable again for me. I had no crashes for some weeks now. If there wouldn't be so many other reports, I would agree to close this bug.

I am experiencing the same freezes with F15 KDE on a Thinkpad T510. I am using the kmod-nvidia drivers, and the freezes are exactly like mentioned above, they seem to happen independent of the applications that I run. KDE Desktop effects are turned off.

Any hints on how to contribute with log files or tests to find this bug would be appreciated.

The original reporter has said he no longer gets freezes.  I'm closing this bug out.

For everyone else that reported a "me too" problem, none of the information is really usable.  If you are still seeing hangs with the latest 2.6.40.6 F15 kernel, and you are NOT using the nvidia modules, or the vbox modules, please open a new bug.