RE: nfs: task 37637 can't get a request slot

From: Bernard Dautrevaux (Dautrevaux@microprocess.com)
Date: Tue Mar 28 2000 - 03:27:51 EST


> -----Original Message-----
> From: Trond Myklebust [mailto:trond.myklebust@fys.uio.no]
> Sent: Monday, March 27, 2000 10:06 PM
> To: Bernard Dautrevaux
> Cc: 'linux-kernel@vger.rutgers.edu'
> Subject: Re: nfs: task 37637 can't get a request slot
>
>
> >>>>> " " == Bernard Dautrevaux <Dautrevaux@microprocess.com> writes:
>
> > As long as it is working I get occasional complaints like
> > these:
> > svc: unknown program 100227 (me 100003) svc:
> unknown version
> > (3)
> > however I also get these messages in single processor mode.
>
> This is a completely harmless message. It is triggered by Solaris
> clients trying to test for their proprietary ACL code.

Thanks for the clarification; I was thinking so but I appreciate to be sure.

>
> > When freezed you from time to time I see the following message
> > on the system console:
> > nfs: task 37637 can't get a request slot
> > where the task number may change from message to message (I've
> > seen at least 37638 and 37639)
>
> This just means that the RPC layer is congested for some reason:
> Either the network is too slow, and some requests are timing out
> because they can't get shovelled out in time, or you have noise
> causing blockage.
>
> Use tcpdump on both client and server in order to determine whether or
> not the client is sending out requests, and whether the server is
> replying to them (each and every request should normally be receiving
> a reply if all is well).
>

I'll do that next time it crashes ;-( and keep you informed; however the
network is still working OK on this station: pings from the station to the
server and vice versa pass as fast as usual, xloads or xosview displayed on
X terminals work correctly, although the load average on the Linux (NFS
client) box is climbing to very unusual heights and quickly stabilize at
13-15 while CPU is 95% idle... (the load average on the Solaris server is
normal and it seems to continue serving other machines).

In all cases thank you for the prompt answer.

        Bernard.

PS: I'm not subscribed to the list, so please CC. me of any answer; Thanks.

--------------------------------------------
Bernard Dautrevaux
Microprocess Ingéniérie
97 bis, rue de Colombes
92400 COURBEVOIE
FRANCE
Tel: +33 (0) 1 47 68 80 80
Fax: +33 (0) 1 47 88 97 85
e-mail: dautrevaux@microprocess.com
                b.dautrevaux@usa.net
--------------------------------------------

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Fri Mar 31 2000 - 21:00:21 EST