System (new computer) Motherboard: Asus P7P55D LE BIOS updated: P7P55D-LE-ASUS-1408 (Same problems with original version) CPU: Intel Core i5 661 3.33Ghz 4 Gb Ram Graphics card: Asus Gforce GT 220, 1Gb 2* SATA Hd Western Digital Caviar Green 2 Terabytes OS: Ubuntu10.04 2.6.32-22-generic kernel 64 bits (Same problems with kernel 2.6.31) Desktop: Gnome Problems (only occasionally) Temporary freezes: Opening applications Closing applications Opening application's dialogs (specially opening/saving documents) Opening and saving (gedit) Refreshing nautilus Audio playing stops temporarily Disk operations seem to take a long time specially 'umount' Sometimes shutdown takes 10..15 minutes apparently waiting for 'umount' Freezes can be produced by copying big files from an encryped Fs (mounted loop) The slowdown of the system seems to be exponential, the longer it takes to copy the more the system slows down, the more the system slows down the longer it takes to copy... etc Checked: Memory checked: Ok All cpus (4: 2 cores + hyperthreading) have low loads (10..20%) when problems appear. temperature (sensors) low: 32°C Changed some settings in BIOS without any improvement Drives set to AHCI or IDE Intel SpeedStep (CPU speed controlled by OS) enabled or disabled Hard disks are Ok: Tested them with palimpsest / disk always shows a choppy benchmark graph (Have not found any documentation on palimpsest :-( ) Same problem booting from both disks (both have Linux) Problems happen with both free and non-free nvidia drivers No error entries in logs except for blocked process warnings (not always tracker-indexer, sometimes umount): INFO: task tracker-indexer:1878 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. ......... Call Trace: ......... acpi does not seem to be functional acpi -V Cooling 0: Processor 0 of 0 Cooling 1: Processor 0 of 0 Cooling 2: Processor 0 of 0 Cooling 3: Processor 0 of 0 acpidump >/tmp/table.txt Wrong checksum for OEMB!
Today there was a particularly bad freeze (It usually happens soon after booting when anacron runs /etc/cron.daily). The system is still misbehaving as I write this but in the previous 10 minutes it was almost completely unresponsive. I could barely run 'ls' and 'top' in terminals. Cpu load was very low. I could also check temperatures: Motherboard : 24.00°C Main Cpu : 30.00°C Graphics Cpu : 28.00°C Hd /dev/sda : 27.00°C Hd /dev/sdb : 27.00°C This time I got some kernel messages that include attached. Any help heartily welcome! Thanks in advance.
Created attachment 26451 [details] kern.log May 20, 2010 after a bad system freeze. I had to wait a long time until the attachment/browser dialog window opened :-(
I have checked 'umount -l' with strace: a) invoked from a gnome Linux terminal: strace -r -t -T -o strace.reference umount -l /mnt/EmergencyLinuxStore umount took about 12 seconds (strace.reference attached): 0.000024 umount("/mnt/EmergencyLinuxStore", MNT_DETACH) = 0 <11.912094> b) invoked from a cron job in conditions which induce freezes: Same line as above included in a shell script (log file: strace.21513) umount took about 512 seconds!! (strace.21513 attached): 0.000024 umount("/mnt/EmergencyLinuxStore", MNT_DETACH) = 0 <511.296337>
Created attachment 26557 [details] strace 'umount -l' log file: strace.reference
Created attachment 26558 [details] strace 'umount -l' log file: strace.21513
Please close this bug as I have created a new one: https://bugzilla.kernel.org/show_bug.cgi?id=16219 Best regards
*** This bug has been marked as a duplicate of bug 16219 ***