[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-bugs-dist
Subject:    Bug#24349: disk accesses not reported
From:       Chris Schlaeger <cs () kde ! org>
Date:       2001-04-30 17:51:46
[Download RAW message or body]

On Mon, Apr 30, 2001 at 10:27:13AM -0700, LA Walsh wrote:
> Here is just the disk_io part first over scan and create tar of a large dir:
> 
> law> while :; do grep disk /proc/stat; sleep 5 ; done
> disk_io: (3,0):(2805360,2523364,24880668,281996,6858624)
> ...
> 
> Here is a complete /proc/stat listing:
> 
> law> more /proc/stat
> cpu  517538 79487 5576451 20348848
> cpu0 517538 79487 5576451 20348848
> page 12569745 3608532
> swap 692 1095
> intr 45015187 26522324 56872 0 3 3 664503 3 959917 5813786 0 3 7370820 686894 0
> 2816478 123581 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 \
>                 0 0 0 0 0 0
> disk_io: (3,0):(2820884,2532577,25060428,288307,7217064)
> ctxt 98482078
> btime 988385608
> processes 40489
> 
> I ran kysguard over the time these figures were generated with 1 min load
> and all 5 disk params.  The load went up to about 1.7, the disk windows
> all remained blank.

This all looks fine. Now we have to see where this information gets
lost. Now start ksysguards backend and monitor the disk usage
manually. Simply type ksysguardd in a shell and then enter
'disk/3:0/wio' or 'disk/3:0/rio' while causing some disk IO. You
should get numbers greater than 0 once in a while.

Chris


[prev in list] [next in list] [prev in thread] [next in thread]