Bug#597226: severity: Grave
During the week, I tried to use the two kernels: 2.6.35.7 (compiled by myself) and the 2.6.32-23 (from debian oficial mirror). Using 2.6.35, no error occurred. But using 2.6.32, the error still happens. And there was one time I used the 2.6.32, I turned off the computer and turned it on using the 2.6.35 and the error occurred. So, I think that there is something wrong in the shut down process of the 2.6.32. Summarizing: the kernel 2.6.35 works fine and the kernel 2.6.32 might has a problem in the shut down process. My problem is fixed, but if you want, I could do some more tests to find out the problem in the kernel 2.6.32, just tell me what I should do. Thanks, Gilberto
Bug#597226: severity: Grave
On Sun, 2010-10-03 at 15:32 -0300, Gilberto Segundo wrote: Hi! Sorry for the late. I had never compiled the kernel before. I compiled the kernel 2.6.35.7 I followed the steps: 1 - Download and untar the kernel in /usr/src. 2 - # make localmodconfig 3 - INSTALL_MOD_STRIP=1 CONCURRENCY_LEVEL=3 fakeroot make-kpkg --initrd --append-to-version=-i386-edps --revision=+1.0 kernel_image kernel_headers 4 - # dpkg -i linux-headers-2.6.35-i386-edps_+1.0_i386.deb # dpkg -i linux-image-2.6.35-i386-edps_+1.0_i386.deb I shut down and turned on the computer 6 times and no error occured. I'm going to use the computer this week and after I put here the results. It is now a week later. What are your results? Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse. signature.asc Description: This is a digitally signed message part
Bug#597226: severity: Grave
Hello. I tested without the nvidia driver and the problem continues to happen. I did 3 tests: 1 - With Nvidia driver installed and using nvidia driver 2 - With Nvidia driver installed but using nv driver instead of nvidia 3 - With nvidia drver uninstalled and using nv driver. The lsmod does not show nvidia modules loaded. In all tests, when I rebooted the system (I rebooted the system 13 times), no error occured, but when I shut down the computer and after turn it on, the problem occurs in 70% of times, more os less. I tried to reboot the system and shut down the computer before the grub show up, but doing this, the problem occurs in next boot. Do you have more suggestions?
Bug#597226: severity: Grave
On Sun, Oct 03, 2010 at 12:08:21PM -0300, Gilberto Segundo wrote: Hello. I tested without the nvidia driver and the problem continues to happen. I did 3 tests: 1 - With Nvidia driver installed and using nvidia driver 2 - With Nvidia driver installed but using nv driver instead of nvidia 3 - With nvidia drver uninstalled and using nv driver. The lsmod does not show nvidia modules loaded. In all tests, when I rebooted the system (I rebooted the system 13 times), no error occured, but when I shut down the computer and after turn it on, the problem occurs in 70% of times, more os less. I tried to reboot the system and shut down the computer before the grub show up, but doing this, the problem occurs in next boot. Do you have more suggestions? did you try newer experimental linux-images? if you can still reproduce with 2.6.36-rcX or 2.6.35, please inform upstream ext4 devs and let us know the bug nr so that we can track it. by the way nouveau is replacing nv. thanks. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#597226: severity: Grave
On Sun, 2010-10-03 at 12:08 -0300, Gilberto Segundo wrote:
Hello.
I tested without the nvidia driver and the problem continues to
happen.
I did 3 tests:
1 - With Nvidia driver installed and using nvidia driver
2 - With Nvidia driver installed but using nv driver instead of nvidia
3 - With nvidia drver uninstalled and using nv driver. The lsmod does
not show nvidia modules loaded.
In all tests, when I rebooted the system (I rebooted the system 13
times), no error occured, but when I shut down the computer and after
turn it on, the problem occurs in 70% of times, more os less.
[...]
From this I would conclude that (some of) the last disk write operations
are being lost. What model of hard drive is in this computer? You
should be able to check that with:
cat /sys/bus/scsi/devices/target0:0:0/0:0:0:0/{vendor,model}
Ben.
--
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.
signature.asc
Description: This is a digitally signed message part
Bug#597226: severity: Grave
HD: ATA SAMSUNG HD250HJ I'm compiling the kernel 2.6.35.7 now. After I tell here the results.
Bug#597226: severity: Grave
Hi! Sorry for the late. I had never compiled the kernel before. I compiled the kernel 2.6.35.7 I followed the steps: 1 - Download and untar the kernel in /usr/src. 2 - # make localmodconfig 3 - INSTALL_MOD_STRIP=1 CONCURRENCY_LEVEL=3 fakeroot make-kpkg --initrd --append-to-version=-i386-edps --revision=+1.0 kernel_image kernel_headers 4 - # dpkg -i linux-headers-2.6.35-i386-edps_+1.0_i386.deb # dpkg -i linux-image-2.6.35-i386-edps_+1.0_i386.deb I shut down and turned on the computer 6 times and no error occured. I'm going to use the computer this week and after I put here the results. Thank you everybody
Bug#597226: severity: Grave
Today, I got the same error after the boot, more or less with three minutes in use. The filesystem is remounted read-only and the messages like that are showed (Dmesg did not save the log because the filesystem become read only) Aborting journal on device sda5-8. EXT4-fs (sda5): Remounting filesystem read-only EXT4-fs (sda5): delayed block allocation failed for inode 527890 at logical offset 0 with max blocks 1 with error -30 This should not happen!! Data will be lost Once data will be lost, I think the severity should be changed to grave, shouldn't? The system stopped and I had to restart the system using a live CD and fix the partition. There is a similar bug in Ubuntu: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/528981
Bug#597226: severity: Grave
On Fri, 2010-10-01 at 10:37 -0300, Gilberto Segundo wrote: Today, I got the same error after the boot, more or less with three minutes in use. The filesystem is remounted read-only and the messages like that are showed (Dmesg did not save the log because the filesystem become read only) Aborting journal on device sda5-8. EXT4-fs (sda5): Remounting filesystem read-only EXT4-fs (sda5): delayed block allocation failed for inode 527890 at logical offset 0 with max blocks 1 with error -30 This should not happen!! Data will be lost Once data will be lost, I think the severity should be changed to grave, shouldn't? [...] This message means that data in memory will not be written to disk; we do not consider that grave unless the failure is silent. See http://kernel-handbook.alioth.debian.org/ch-bugs.html. Please can you test without using the nvidia driver? Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse. signature.asc Description: This is a digitally signed message part
