8.0-RC1 NFS client timeout issue

[Olaf Seibert]

On Sun 01 Nov 2009 at 17:17:15 -0500, Rick Macklem wrote:
> On Thu, 29 Oct 2009, Olaf Seibert wrote:
> 
> >
> > Thanks, it looks like it should do the trick. I can't try it before
> > monday, though.
> >
> Although I think the patch does avoid sending the request on the
> partially closed connection, it doesn't fix the "real problem",
> so I don't know if it is worth testing?

Well, I tested it anyway, just in case. It seems to work fine for me, so
far.

I don't see your extra RSTs either. Maybe that is because in my case the
client used a different port number for the new connection. (Usually,
this is controlled by the TCP option SO_REUSEADDR from <sys/socket.h>).

Here is a new packet trace. I had to cut out some packets since I forgot
to kill some (failing) mount attempts of another directory on the same
server.
(sorry again for the long lines)

No.     Time        Source                Destination           Protocol Info
    486 60.438406   xxx.xxx.31.43         xxx.xxx.16.142        NFS      V3 LOOKUP Call (Reply In 487), DH:0x61b8eb12/date
    487 60.438629   xxx.xxx.16.142        xxx.xxx.31.43         NFS      V3 LOOKUP Reply (Call In 486) Error:NFS3ERR_NOENT
    488 60.538796   xxx.xxx.31.43         xxx.xxx.16.142        TCP      hello-port > nfs [ACK] Seq=36477 Ack=44701 Win=8192 Len=0 TSV=228817 TSER=1575935

last real action on old connection (client port "hello-port")

    537 420.437763  xxx.xxx.16.142        xxx.xxx.31.43         TCP      nfs > hello-port [FIN, ACK] Seq=44701 Ack=36477 Win=49232 Len=0 TSV=1611935 TSER=228817
    538 420.437805  xxx.xxx.31.43         xxx.xxx.16.142        TCP      hello-port > nfs [ACK] Seq=36477 Ack=44702 Win=8192 Len=0 TSV=588734 TSER=1611935

server ends connection

    563 605.334262  xxx.xxx.31.43         xxx.xxx.16.142        TCP      hello-port > nfs [FIN, ACK] Seq=36477 Ack=44702 Win=8192 Len=0 TSV=773641 TSER=1611935

some time later, client now ends connection before sending its request on new connection
(port 875)

    564 605.334303  xxx.xxx.31.43         xxx.xxx.16.142        TCP      875 > nfs [SYN] Seq=0 Win=65535 Len=0 MSS=1460 WS=5 TSV=773641 TSER=0
    565 605.334440  xxx.xxx.16.142        xxx.xxx.31.43         TCP      nfs > hello-port [ACK] Seq=44702 Ack=36478 Win=49232 Len=0 TSV=1630424 TSER=773641
    566 605.334564  xxx.xxx.16.142        xxx.xxx.31.43         TCP      nfs > 875 [SYN, ACK] Seq=0 Ack=1 Win=49232 Len=0 TSV=1630424 TSER=773641 MSS=1460 WS=0
    567 605.334588  xxx.xxx.31.43         xxx.xxx.16.142        TCP      875 > nfs [ACK] Seq=1 Ack=1 Win=66592 Len=0 TSV=773641 TSER=1630424

new connection set up

    568 605.334605  xxx.xxx.31.43         xxx.xxx.16.142        NFS      V3 ACCESS Call (Reply In 570), FH:0x008002a2
    569 605.334828  xxx.xxx.16.142        xxx.xxx.31.43         TCP      nfs > 875 [ACK] Seq=1 Ack=141 Win=49092 Len=0 TSV=1630424 TSER=773641

and in use

> I'm hoping that the "Help TCP Wizards..." thread I just started
> on freebsd-current comes up with something.
> 
> At least I can reproduce the problem now. (For some reason, I have
> to reboot the Solaris10 server before the problem appears for me.
> I can't think why this matters, but that's networking for you:-)

Maybe it depends on server load or something. This particular server is
a central file server at a university, it may have some more pressure to
terminate unused connections.

> rick
-Olaf.
--