On Wed, Oct 25 2000, Tom Holroyd wrote:
> Alpha DP264 UP.
>
> Doing a dump to a SCSI MO disk (aic7xxx, Fujitsu Gigamo, ro ext2 fs), and
> simultaneously writing a megabyte to a floppy (tar -cf /dev/fd0, "FDC 0 is
> a post-1991 82077") produced a ten second freeze. I did this twice and
> it's reproducible. The system froze including the mouse and returned to
> normal afterwards. vmstat looked like this:
>
> procs memory swap io system cpu
> r b w swpd free buff cache si so bi bo in cs us sy id
> 2 1 2 0 4440 598848 37344 0 0 719 492 1308 1081 97 3 0
> 2 1 2 0 2344 601000 37344 0 0 1480 481 1386 1685 95 5 0
> 1 1 2 0 2464 600880 37344 0 0 778 433 1368 1197 98 2 0
> 1 1 2 0 4568 598928 37344 0 0 345 419 1477 1206 98 2 0
> 1 4 2 0 4960 596392 38936 0 0 2027 725 1405 1825 94 6 0
> *2 4 2 0 5000 596264 39032 0 0 115 3252 6173 106 0 100 0
> 1 2 2 0 5064 597032 39032 0 0 6251 9 1308 3332 85 15 0
> 1 2 2 0 5064 597032 39032 0 0 0 9 1049 223 97 3 0
> 1 2 2 0 5056 597032 39032 0 0 0 12 1122 372 96 4 0
>
> The 100% sys cpu line is the one (this one only lasted about 6 seconds).
This is a known problem that I've been meaning to look into. It's not just MO
drives, basically any "slower" type device will provoke this behaviour.
-- * Jens Axboe <axboe@suse.de> * SuSE Labs - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 21:00:16 EST