Hi, On 2012-06-06 20:50:17 +0100, Steven Chamberlain wrote: > This could be a symptom of bad RAM in the machine. (Although from the > system info, it sounds like you might have ECC RAM).
When there are RAM problems, there are error messages in the /var/log/kern.log file, like: ECC/ChipKill ECC error. But here, the sysadmin said that there were no such error messages. > Have you considered checking it, e.g. with memtest86+? If you can't > take the machine offline, memtester might give you some idea (as root, > test an amount approx. equal to the amount reported as free+cache). I'll ask the sysadmin. > Otherwise, you could maybe disable your swap partitions and run > badblocks on them in case something gets corrupted after being written > out / read back in from disk? The machine has 80 GB (currently only 3 GB are used). I suppose that swap is never used. At least, according to atop: MEM | tot 78.8G | free 51.8G | cache 23.6G | buff 831.9M | slab 1.8G | SWP | tot 9.8G | free 9.8G | | vmcom 650.4M | vmlim 49.2G | MEM | tot 78.8G | free 51.9G | cache 23.7G | buff 832.0M | slab 1.8G | SWP | tot 9.8G | free 9.8G | | vmcom 546.3M | vmlim 49.2G | at 16:02 and 16:12 respectively (the problem occurred at 16:08). Note: I'm running the same test in a loop on 3 machines (including this one where the problem occurred). -- Vincent Lefèvre <vinc...@vinc17.net> - Web: <http://www.vinc17.net/> 100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/> Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon) -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
