Processes stuck in nfsreq

[Kris Kennaway]

On Thu, Feb 03, 2005 at 10:29:16AM -0800, Mohan Srinivasan wrote:
> Hi,
> 
> In the info you have below, the NFS request was not sent (R_SENT
> is not set). We didn't send out the request because possibly nm_cwnd 
> didn't allow us to send the request. If nfs_request() could not send 
> the request, then typically, the request would be retransmitted from 
> nfs_timer(). 
> 
> If nm_cwnd is full, then there must be a lot of outstanding requests 
> that haven't been replied to yet (on this mount), which is the fundamental
> issue.

FYI, this is truly hung and not just backlogged - it hasn't proceeded
any further since last night.

> Can you force a core dump and send me a pointer to it ?

Unfortunately not, because twed dumping is still broken.  The only way
I was able to obtain this information is because kan has patches that
fix kgdb enough to use it on the running kernel.

> Also, after you force a core, can you also try a quick workaround - someone
> else also reported NFS client hangs and said that things were fine
> after they set mpsafenet to 0. It would be good to see if there's a
> correlation there.

I can try that.

Kris
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-current/attachments/20050203/690e50bb/attachment.bin