On 5/10/12 2:56 PM, "scameron@xxxxxxxxxxxxxxxxxx" <scameron@xxxxxxxxxxxxxxxxxx> wrote: > >I have this script, which just cranks up the scsi logging level >a bit, then writes 1 block out to a disk, sleeps a bit, then >cranks the logging back down. There's no other i/o going >to the disk (not mounted, no filesystem) > >-----8<---- -----8<---- -----8<---- -----8<---- -----8<---- >#!/bin/sh > >logger >vvvvvv==============================================================vvvvvv >LOGGING_LEVEL=14683648 > >sysctl -w dev.scsi.logging_level=$LOGGING_LEVEL >dd if=/dev/zero of=/dev/sdb count=1 bs=512 seek=5000 oflag=direct >sleep 1 >sysctl -w dev.scsi.logging_level=0 >logger >^^^^^^==============================================================^^^^^^ >-----8<---- -----8<---- -----8<---- -----8<---- -----8<---- > > >This script produces the log below. It does the write, as >expected, but then it also does a test unit ready, then inquiries and a >bunch of reads. > >Where are these coming from? I had a similar question once and found that when I closed the disk handle, udev was invoking blkid and that was generating some read noise. blktrace was helpful there. > -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
