To introduce: I have a machine with the following specs:
- Intel Core i7 930
- 12GB Ram Corsair
- 2xSamsung HDD 320gb (No Raid, just partitions)
- Asus P6TD Deluxe
The machine has been at Datacenter just a few days. The average load is 0.50 and I have the following partitions:
/ ext4 noatime,barrier=0,errors=remount-ro 0 1 /datos ext4 noatime 0 2
Now the problem: At random intervals, the machine locks up, SSH lags heavily, and viewing htop, it says that all cores are being hammered by system processes. http://korrupzion.com/htop.png
Iostat output during freeze (got this when i finally managed to execute a command during freeze):
iostat Linux 2.6.35-22-server (charizard) 25/10/10 _x86_64_ (8 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 4,33 0,00 10,38 1,21 0,00 84,07 Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn sda 2,68 269,23 0,01 906918 24 sdb 52,30 897,99 1896,08 3024878 6386976
Vmstat output:
vmstat 1 procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 0 1 0 5680460 128056 1907340 0 0 115 131 1091 2621 4 9 86 1 4 0 0 5676360 128064 1909036 0 0 1872 52 4606 18143 10 33 57 0
I suspect that ext4 is related to those freezes, but i'm not sure anyway. Ubuntu was installed a week ago, before sending the machine to datacenter, before that, I had windows 7 to test performance and didn't have any sort of freezes
Please if you know another command to track the source of this freezes let me know, I've thinking about formatting again to debian lenny, which was the S.O I used to install to other machine w/o problems
Thanks.
EDIT 1: I remounted "/datos" with barrier=0, now i'm monitoring if problems raise again.
EDIT 2: Remounting /datos with barrier=0 didn't work u.u. Still trying to find solution