RH 4 long halts while doing i/o stress test to san

[Yard, John]

Another odd feature of this system is there are 16 cpu s....

I almost wonder if there is some spin-lock issue....

Any suggestions who to get more information on this issue...

the RH 4.8 sysctl parms seem pretty bare ...

JYard
UCLA


-----Original Message-----
From: redhat-list-bounces at redhat.com
[mailto:redhat-list-bounces at redhat.com] On Behalf Of Yard, John
Sent: Tuesday, September 15, 2009 10:52 AM
To: General Red Hat Linux discussion list
Subject: RE: RH 4 long halts while doing i/o stress test to san

Yes], but there is conflicting information jyard

-----Original Message-----
From: redhat-list-bounces at redhat.com
[mailto:redhat-list-bounces at redhat.com] On Behalf Of hike
Sent: Tuesday, September 15, 2009 5:05 AM
To: General Red Hat Linux discussion list
Subject: Re: RH 4 long halts while doing i/o stress test to san

our san vendor provides the tuning parameters.
have you contacted the vendor for help?


On Mon, Sep 14, 2009 at 6:04 PM, Yard, John <jyard at ais.ucla.edu> wrote:

>
> On RH 4 I have a fiber multipath connection
> to an emc san.
>
> If I do a stress test of i/o to the san -
> an 8G cp from the san to the internal drive and
> visa versa - I see halts of simple
> montoring programs . The monitoring pgm
> writes the date/time to a file and sleeps.
> 30 seconds after cron starts the stress test
> the monitor script will halt for 20 seconds,
> then continue.
>
> RH support suggests this is a tuning issue.
> There is not a lot of docs on 4.8 tuning parms,
> is there any comments or suggestions where I
> need to start
>
> JYard
> UCLA
>
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
>
-- 
redhat-list mailing list
unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list

-- 
redhat-list mailing list
unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list