SEO

Hello all,
the command topas has the follwing output:


Topas Monitor for host: gaviao EVENTS/QUEUES FILE/TTY
Wed Dec 10 17:55:13 2003 Interval: 2 Cswitch 670 Readch 930062
Syscall 4355 Writech 92471
Kernel 45.8 |############# | Reads 444 Rawin 18
User 36.3 |########## | Writes 251 Ttyout 286
Wait 0.6 | | Forks 0 Igets 0
Idle 17.1 |##### | Execs 0 Namei 14
Runqueue 3.5 Dirblk 0
Interf KBPS I-Pack O-Pack KB-In KB-Out Waitqueue 3.0
en0 18.8 79.7 80.2 6.5 12.3
lo0 11.4 25.5 25.5 5.7 5.7 PAGING MEMORY
Faults 0 Real,MB 4095
Disk Busy% KBPS TPS KB-Read KB-Writ Steals 0 % Comp 54.5
hdisk3 3.0 36.1 6.0 0.0 36.1 PgspIn 0 % Noncomp 46.3
hdisk0 0.0 0.0 0.0 0.0 0.0 PgspOut 0 % Client 0.5
hdisk1 0.0 0.0 0.0 0.0 0.0 PageIn 0
hdisk2 0.0 0.0 0.0 0.0 0.0 PageOut 9 PAGING SPACE
Sios 6 Size,MB 6144
cxmnmain (161644 24.9% PgSp: 8.5mb lisline % Used 38.5
sxmnmain_(118362 24.9% PgSp:20.0mb lisline % Free 61.4
cxmnmain (99248) 24.6% PgSp:10.2mb lisline
apscmain (108172 1.8% PgSp: 6.3mb lisline
oracle (146650 1.4% PgSp: 1.3mb lisline Press "h" for help screen.
oracle (117288 1.1% PgSp: 1.9mb lisline Press "q" to quit program.


As you can see, there is something worng because there is to much kernel usage by a process and the Readch is to big too.

How can I see which process is responsable for this?

Thanks,
-vcf

Vitor Flausino <vflausino@dti.pga.pt> wrote in message news:<3fd75f3e$0$243$4d4efb8e@read.news.pt.uu.net> ...
> Hello all,
> the command topas has the follwing output:
>
>
> Topas Monitor for host: gaviao EVENTS/QUEUES FILE/TTY
> Wed Dec 10 17:55:13 2003 Interval: 2 Cswitch 670 Readch 930062
> Syscall 4355 Writech 92471
> Kernel 45.8 |############# | Reads 444 Rawin 18
> User 36.3 |########## | Writes 251 Ttyout 286
> Wait 0.6 | | Forks 0 Igets 0
> Idle 17.1 |##### | Execs 0 Namei 14
> Runqueue 3.5 Dirblk 0
> Interf KBPS I-Pack O-Pack KB-In KB-Out Waitqueue 3.0
> en0 18.8 79.7 80.2 6.5 12.3
> lo0 11.4 25.5 25.5 5.7 5.7 PAGING MEMORY
> Faults 0 Real,MB 4095
> Disk Busy% KBPS TPS KB-Read KB-Writ Steals 0 % Comp 54.5
> hdisk3 3.0 36.1 6.0 0.0 36.1 PgspIn 0 % Noncomp 46.3
> hdisk0 0.0 0.0 0.0 0.0 0.0 PgspOut 0 % Client 0.5
> hdisk1 0.0 0.0 0.0 0.0 0.0 PageIn 0
> hdisk2 0.0 0.0 0.0 0.0 0.0 PageOut 9 PAGING SPACE
> Sios 6 Size,MB 6144
> cxmnmain (161644 24.9% PgSp: 8.5mb lisline % Used 38.5
> sxmnmain_(118362 24.9% PgSp:20.0mb lisline % Free 61.4
> cxmnmain (99248) 24.6% PgSp:10.2mb lisline
> apscmain (108172 1.8% PgSp: 6.3mb lisline
> oracle (146650 1.4% PgSp: 1.3mb lisline Press "h" for help screen.
> oracle (117288 1.1% PgSp: 1.9mb lisline Press "q" to quit program.
>
>
> As you can see, there is something worng because there is to much kernel usage by a process and the Readch is to big too.
>
> How can I see which process is responsable for this?
>
> Thanks,
> -vcf


i would use nmon and use the "t" flag for processes, then hit "5" to
give you i/o channels, this iwll list the processes in order of
highest data throughput.

rgds
Mark Taylor