https://bugzilla.novell.com/show_bug.cgi?id=336538#c1 Matej Horvath changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |alexv@unizar.es --- Comment #1 from Matej Horvath 2007-10-26 08:08:39 MST --- Please attach all information as described here: http://en.opensuse.org/Bugs:Suspend_Failure . -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 Matej Horvath changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |kernel-maintainers@forge.provo.novell.com AssignedTo|bnc-team-screening@forge.provo.novell.com |seife@novell.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 Stefan Seyfried changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|seife@novell.com |kernel-maintainers@forge.provo.novell.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538#c5 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |astarikovskiy@novell.com, len.brown@intel.com Severity|Major |Critical Status|NEW |ASSIGNED --- Comment #5 from Thomas Renninger 2007-11-06 09:43:37 MST --- There already was a problem with fans breaking away after s2disk/ram. I thought this got fixed and the kernel does not rely anymore on the fan and power resource state it had before suspend... Some people had the idea of a watchdog, another thread or whatever mechanism to watch whether the thermal management came out of sync and at least be able to throw a warning before the disk starts to melt. Adding some people who might know ... You might also want to try a more recent kernel from here to verify whether this already got fixed mainline: ftp://ftp.suse.com/pub/projects/kernel/kotd/HEAD/ Raising severity, those thermal issues are critical... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO Info Provider| |astarikovskiy@novell.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538#c10 Alejandro Marino Vaquero Avilés-Casco changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|alexv@unizar.es | --- Comment #10 from Alejandro Marino Vaquero Avilés-Casco 2007-11-27 15:56:40 MST ---

Alejandro: Best you also keep acpi_listen running in a several console to catch ACPI events. Does top show you kacpid running at 100%, maybe some other process?

No, the computer is almost idle

If you have not suspended and you have acpi_listen running you should see a thermal (maybe another) event when 60 degrees are exceeded? Can you check that you do not see this event after s2ram/disk, pls.

No, I've just tried and I don't see the thermal event in acpi_listen neither before nor after suspending to disk, but the fan only works before

Maybe the problem only exists, if the machine is still very hot (above 60 degree after resume, hmm unlikeley...)?

No, in fact, I have to push a bit the machine to reach 60ºC (overload both processors and things like that) but I've reached temperatures higher than 80ºC, and that's a bit dangerous I have not tried a new kernel, for I use this computer everyday to work, and there are some kernel related packets that are not so up-to-date (like ipw3945, nvidia drivers and so). Now I'm using 2.6.22.5-31-default. So tell me if you need anything else -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538#c12 --- Comment #12 from Alejandro Marino Vaquero Avilés-Casco 2007-11-29 12:44:54 MST --- Hey guys, I've just found something interesting that may help you. I tried "s2ram -f" (I've never tried a suspend-to-ram before because it didn't work out-of-the-box, but now, with the -f parameter, works flawlessly) and the fan works fine. Now comes the crazy crazy thing, ok? There it goes: 1. I suspend to disk... 2. ...awake the computer... 3. ...and run some simulations. The processor will go above 60ºC without turning on the fans. 4. So I kill the simulations... 5. ...the processor cools below 60ºC (it takes less than a second) 6. Then I suspend to ram... 7. ...awake the computer again... 8. ...run the simulations, and the fan... works!! Without rebooting the computer, a s2ram just will do! Why is that? What changes??? I really don't catch it acpi_listen didn't return any info on the matter. In fact, it didn't return anything, but it really works, for when I press a hotkey, it says something like this hotkey ATKD 0000007e 00000000 So no thermal info is given anyway Hope it helps... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538#c14 --- Comment #14 from Stefan Seyfried 2007-11-30 05:47:58 MST --- Another thing to try: edit /etc/suspend.conf and put the following line in there (simply at the end): shutdown method = shutdown and try if suspend to disk now leaves your fans functional. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538#c16 --- Comment #16 from Alejandro Marino Vaquero Avilés-Casco 2007-12-01 16:10:25 MST --- (In reply to comment #14 from Stefan Seyfried)

Another thing to try: edit /etc/suspend.conf and put the following line in there (simply at the end):

shutdown method = shutdown

and try if suspend to disk now leaves your fans functional.

Yeah, it works!!! With the old kernel. I have not tried the new one, but I will when I get a break from work. Thank you guys! -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538#c18 --- Comment #18 from Alejandro Marino Vaquero Avilés-Casco 2007-12-02 02:56:52 MST --- Created an attachment (id=185541) --> (https://bugzilla.novell.com/attachment.cgi?id=185541) /etc/sysconfig/kernel configuration file -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User alexv@unizar.es added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c20 --- Comment #20 from Alejandro Marino Vaquero Avilés-Casco 2007-12-09 05:05:27 MST ---

Can you please replace the line

INITRD_MODULES="processor thermal ata_piix fan jbd ext3 edd"

in this file with the following one:

INITRD_MODULES="ata_piix jbd ext3 edd"

(just comment out the old one, so that you can restore it later), recreate the initrd image and see if hibernation works correctly in this configuration _without_ the "shutdown method = shutdown" line in /etc/suspend.conf ?

It doesn't change the behaviour of the laptop at all. The fan won't work after resuming from disk, but it will after a s2ram -f. I'll try to update the kernel without making a mess with the modules and see what happens... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c22 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |hare@novell.com, teheo@novell.com --- Comment #22 from Thomas Renninger 2008-01-11 10:03:42 MST --- Last comment sounds sever: You mean you updated from 10.3 GM to a 10.3 maintenance kernel and since then your hard disk does not get detected anymore? Can you boot the rescue system and install with rpm -ivh (--force) the original kernel from CD/DVD again. Does this one boot again? Please compare the boot parameters of both in /boot/grub/menu.lst, are they the same? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User alexv@unizar.es added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c24 --- Comment #24 from Alejandro Marino Vaquero Avilés-Casco 2008-01-11 11:13:28 MST --- Created an attachment (id=190274) --> (https://bugzilla.novell.com/attachment.cgi?id=190274) Boot "log" with broken kernel Reconstruction, not a real log -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User alexv@unizar.es added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c26 --- Comment #26 from Alejandro Marino Vaquero Avilés-Casco 2008-01-11 11:15:54 MST --- It would be great if you could tell me what happened to my kernel. I don't have any logs, for the kernel couldn't find the harddisk, but I tried to remember what I saw, and I have written a sort of a "log file" (bootlog) just to show you where exactly appeared the problem during the boot. In the "log file", the version numbers of the libraries and kernel may be wrong, and comments are placed between *, it is just intented to explain what happened, for no log was recorded. I also have uploaded a (piece of a) normal boot.msg file, just for comparison. Don't read it thoroughly, just wanted to remark the loads of things that are missing in bootlog, and that may tell you what went wrong during the update. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c28 --- Comment #28 from Thomas Renninger 2008-01-13 15:33:44 MST --- I also slipped into sd missing somehow and Hannes told me to add it. Ahh, looking at comment #19, sd probably has been ripped out there. You need to add it to INITRD_MODULES in /etc/sysconfig/kernel and then invoke mkinitrd (so that the module gets added to initrd and can get loaded early) Do you still have a booting kernel? If not, you probably have to boot the rescue system, mount the partition, do a chroot into it, mount /proc, mount /sys and then edid /etc/sysconfig/kernel and invoke mkinitrd. Hope it works that way and I haven't forgotten anything... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User len.brown@intel.com added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c30 Len Brown changed: What |Removed |Added ---------------------------------------------------------------------------- CC|len.brown@intel.com |acpi-bugzilla@lists.sourceforge.net --- Comment #30 from Len Brown 2008-01-13 22:17:53 MST --- upstream has a similar report: http://bugzilla.kernel.org/show_bug.cgi?id=9169 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User alexv@unizar.es added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c32 --- Comment #32 from Alejandro Marino Vaquero Avilés-Casco 2008-01-14 09:02:55 MST --- New test performed: Excluding/Including processor, thermal and fan in the INITRD_MODULES field doesn't mend things. In fact, with "mkinitrd -A" (which creates a 15 mb file, full of modules that I'll never use) the same happens, and it won't boot. I tested it for both kernels, 2.6.22.5-31 and the 2.6.24_rc6_git11. But I'm sure there is something wrong with initrd, for to restore the old kernel (2.6.22.5-31) I just overwrite the new initrd file with the old one, and everything works again. Could someone explain me why my old and working initrd file is around 500000 bytes smaller than the new ones I make with mkinitrd (same kernel, the old one)? I'm afraid that I won't be able to test another kernel until I fix this problem... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User alexv@unizar.es added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c34 --- Comment #34 from Alejandro Marino Vaquero Avilés-Casco 2008-01-14 17:27:52 MST --- Good news guys, I finally found it. This is one of the most annoying bugs I've ever come across. I unpacked my one hundred initrd images filling my disk, to see which where the differences between the working one and the others, et voila, it was the udev binaries (in /sbin). When I updated the kernel, I updated the udev package as well, and now there is one binary, udevadm, doing several functions, instead of udevcontrol, udevsettle or udevtrigger, for instance. The last 3 were executed in the working initrd, for the 3 udev binaries were included in the image, using old udev scheme. In the new scheme, these three binaries are substituted by symbolic links to one binary, udevadm, and someone (who deserves hell) forgot to put the udevadm file inside the image, so the three symbolic links were broken, and I suppose that everybody updating a kernel after updating udev should face the same problem. mkinitrd doesn't put the right files in the image (I think it believes the symbolic links udevcontrol, udevsettle and udevtrigger are binaries). Are the people in charge of mkinitrd aware of this????? I'll try hibernation with a newer kernel asap. Now is too late here, and I should sleep a bit... NB: My udev package is udev-118-4.6 My initrd package is mkinitrd-2.1-36.2 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User teheo@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c36 --- Comment #36 from Tejun Heo 2008-01-15 04:11:39 MST --- Can you please file a bug report on the initrd issue? We definitely need to get that fixed. Thanks a lot. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=336538 User trenn@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=336538#c38 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED --- Comment #38 from Thomas Renninger 2008-02-05 07:34:03 MST --- Works/fixed for current mainline kernel. Workaround for 10.3: suspend to disk shutdown method: "shutdown" (see comment #16). -> closing. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.