Messages in this thread | | | From | Ian Campbell <> | Date | Mon, 25 Aug 2008 10:59:31 +0100 | Subject | Re: NFS regression? Odd delays and lockups accessing an NFS export. |
| |
On Sun, 2008-08-24 at 18:15 -0400, Trond Myklebust wrote: > The server should ideally start to close the socket as soon as it > receives the FIN from the client. I'll have a look at the code.
The server is a 2.6.25 kernel from Debian backports.org. I was planning to downgrade to the 2.6.24 "Etch and a half" kernel but was holding off while this thread was in progress. If you think it would be a useful data point I can do it sooner rather than later.
It's changing the client kernel which seems to make a difference since running 2.6.24 against this server has given me no trouble. I guess the newer client could expose a bug in the server though.
Ian. -- Ian Campbell
I can't understand why a person will take a year or two to write a novel when he can easily buy one for a few dollars. -- Fred Allen [unhandled content-type:application/pgp-signature] | |