[OpenIndiana-discuss] NFS hang during copy

[gonczi at comcast.net]

Hello, 

When looking at (nearly) hung sytems, it maybe enlightening to take a few snapshots of all the kernel threads 
E.g.: fire up mdb, then enter a set of commands, like below: 

mdb -k 
::log /tmp/threads.txt 
::threadlist -v 
::quit 

On a busy system, (where you are burning cpu time), profiling with lockstat is a good first step. 
something like: `lockstat -kgIw sleep 5` ( off the top of my head, please double check the options 
via `man lockstat`) 

Note that lockstat just samples what is on cpu, so if you are spending your cycles waiting on locks or cv-s, 
lockstat will report this as " you are spending your time running some form of mwait" 


mpstat tends to be useful for coarse grain analysis, at lest you get an idea what type of 
activity is responsible for the poor performance. 

Steve 


----- Original Message ----- 
From: "Mateusz Pawlowski" <js at yllq.net> 
To: "Discussion list for OpenIndiana" <openindiana-discuss at openindiana.org> 
Sent: Monday, March 21, 2011 7:16:46 AM 
Subject: Re: [OpenIndiana-discuss] NFS hang during copy 

On Mon, 21 Mar 2011 10:39:51 +0100 (CET), Roy Sigurd Karlsbakk wrote: 
>> > I'm fighting a problem with an OpenIndiana 148 server and NFS3 
>> > mounts from Linux clients. A simple cron job is run that moves 
>> some 
>> > data files from another server to the OI box. This runs well for a 
>> > while, until at some point, the client hangs and reports NFS 
>> server 
>> > connection failure. The calltrace from linux is 
>> 
>> [...] 
>> 
>> What mount options is the Linux client using? 
> 
> urd:/dpool/nilu_wrk on /xnilu_wrk type nfs4 
> (rw,bg,clientaddr=192.168.0.120,addr=128.39.104.76) 
> 
> (I forgot I did some testing with nfs4 as well). I just tried to 
> change that to: 
> 
> urd:/dpool/nilu_wrk on /xnilu_wrk type nfs 
> (rw,sync,bg,soft,intr,vers=3,addr=128.39.104.76) 
> 
> same issue - after some time, client process hang in D state and a 
> (forced) reboot is required for this to be released. 
> 
> Vennlige hilsener / Best regards 
> 


I have same issues after upgrading from OS , any NFS operation on 
bigger that a few meg files halts server with kernel taking all 
resources when viewed via top. 
My clients are Gentoo with 2.6.36 kernels 

-- 
Regards, 
Mateusz 
using Generik Hosting 
www.generik.co 

_______________________________________________ 
OpenIndiana-discuss mailing list 
OpenIndiana-discuss at openindiana.org 
http://openindiana.org/mailman/listinfo/openindiana-discuss