Printable View
Hello,
I wrote a little NFS server as a proxy-server...
The problem is, when registering the NFS server on a TCP transport, it
comes to retransmits of reply (!) messages and the server gets very slow.
In UDP, everything works correctly...
Anybody has a hint for me?
Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.06.19.14.47.34.182800@alpha-lab.net>...[color=blue]
> Hello,
>
> I wrote a little NFS server as a proxy-server...
>
> The problem is, when registering the NFS server on a TCP transport, it
> comes to retransmits of reply (!) messages and the server gets very slow.
> In UDP, everything works correctly...
>
> Anybody has a hint for me?[/color]
You don't provide much to go on.
Why is your server retransmitting responses? They should be
sent once. Of course, at teh TCP level they could
be re-sent if the client's TCP end point
is not responding. What does a packet trace show you?
In the trace just the duplicate reply shows up. On the TCP level however,
I see an ACK from the server before and an ACK from the client immediatly
after the retransmitted reply (reproducible).
Am Sun, 20 Jun 2004 06:44:22 -0700 schrieb Mike Eisler:
[color=blue]
> Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.06.19.14.47.34.182800@alpha-lab.net>...[color=green]
>> Hello,
>>
>> I wrote a little NFS server as a proxy-server...
>>
>> The problem is, when registering the NFS server on a TCP transport, it
>> comes to retransmits of reply (!) messages and the server gets very slow.
>> In UDP, everything works correctly...
>>
>> Anybody has a hint for me?[/color]
>
> You don't provide much to go on.
>
> Why is your server retransmitting responses? They should be
> sent once. Of course, at teh TCP level they could
> be re-sent if the client's TCP end point
> is not responding. What does a packet trace show you?[/color]
Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.07.05.19.32.52.70882@alpha-lab.net>...[color=blue]
> In the trace just the duplicate reply shows up. On the TCP level however,[/color]
if all you see ia a duplicate and not the riginal, then how do you know
it is a duplicate?
:-)
[color=blue]
> I see an ACK from the server before and an ACK from the client immediatly
> after the retransmitted reply (reproducible).[/color]
if your server is really sending multiple replies and the client isn't
sending multiple requests, then there's a bug in either your
RPC library or your server (are you calling svc_sendreply()
multiple times?)
[color=blue]
>
>
> Am Sun, 20 Jun 2004 06:44:22 -0700 schrieb Mike Eisler:
>[color=green]
> > Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.06.19.14.47.34.182800@alpha-lab.net>...[color=darkred]
> >> Hello,
> >>
> >> I wrote a little NFS server as a proxy-server...
> >>
> >> The problem is, when registering the NFS server on a TCP transport, it
> >> comes to retransmits of reply (!) messages and the server gets very slow.
> >> In UDP, everything works correctly...
> >>
> >> Anybody has a hint for me?[/color]
> >
> > You don't provide much to go on.
> >
> > Why is your server retransmitting responses? They should be
> > sent once. Of course, at teh TCP level they could
> > be re-sent if the client's TCP end point
> > is not responding. What does a packet trace show you?[/color][/color]
Am Tue, 06 Jul 2004 17:18:37 -0700 schrieb Mike Eisler:[color=blue]
> Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.07.05.19.32.52.70882@alpha-lab.net>...[color=green]
>> In the trace just the duplicate reply shows up. On the TCP level however,[/color]
>
> if all you see ia a duplicate and not the riginal, then how do you know
> it is a duplicate?
>
> :-)
>[color=green]
>> I see an ACK from the server before and an ACK from the client immediatly
>> after the retransmitted reply (reproducible).[/color]
>
> if your server is really sending multiple replies and the client isn't
> sending multiple requests, then there's a bug in either your
> RPC library or your server (are you calling svc_sendreply()
> multiple times?)
>[color=green]
>>
>>
>> Am Sun, 20 Jun 2004 06:44:22 -0700 schrieb Mike Eisler:
>>[color=darkred]
>> > Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.06.19.14.47.34.182800@alpha-lab.net>...
>> >> Hello,
>> >>
>> >> I wrote a little NFS server as a proxy-server...
>> >>
>> >> The problem is, when registering the NFS server on a TCP transport, it
>> >> comes to retransmits of reply (!) messages and the server gets very slow.
>> >> In UDP, everything works correctly...
>> >>
>> >> Anybody has a hint for me?
>> >
>> > You don't provide much to go on.
>> >
>> > Why is your server retransmitting responses? They should be
>> > sent once. Of course, at teh TCP level they could
>> > be re-sent if the client's TCP end point
>> > is not responding. What does a packet trace show you?[/color][/color][/color]
Well: I can identify duplicate replies out of the XID, which is the same
for all replies ;)
JOchen
Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.09.20.10.48.23.604578@alpha-lab.net>...[color=blue]
> Am Tue, 06 Jul 2004 17:18:37 -0700 schrieb Mike Eisler:[color=green][color=darkred]
> >> > Why is your server retransmitting responses? They should be
> >> > sent once. Of course, at teh TCP level they could
> >> > be re-sent if the client's TCP end point
> >> > is not responding. What does a packet trace show you?[/color][/color]
>
> Well: I can identify duplicate replies out of the XID, which is the same
> for all replies ;)[/color]
I suspect the TCP sequence number is the same too. :-)
Mike Eisler <spamisevi1@yahoo.com> wrote:[color=blue]
> Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.09.20.10.48.23.604578@alpha-lab.net>...[color=green]
>> Well: I can identify duplicate replies out of the XID, which is the
>> same for all replies ;)[/color][/color]
[color=blue]
> I suspect the TCP sequence number is the same too. :-)[/color]
And if it is not, it means the NFS stuff is indeed resending things.
I guess that would indeed be a bug if it was the reply.
rick jones
--
a wide gulf separates "what if" from "if only"
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to raj in cup.hp.com but NOT BOTH...
Rick Jones <foo@bar.baz.invalid> wrote in message news:<lAZ3d.11326$MV4.9646@news.cpqcorp.net>...[color=blue]
> Mike Eisler <spamisevi1@yahoo.com> wrote:[color=green]
> > Jochen Witte <jwitte@alpha-lab.net> wrote in message news:<pan.2004.09.20.10.48.23.604578@alpha-lab.net>...[color=darkred]
> >> Well: I can identify duplicate replies out of the XID, which is the
> >> same for all replies ;)[/color][/color]
>[color=green]
> > I suspect the TCP sequence number is the same too. :-)[/color]
>
> And if it is not, it means the NFS stuff is indeed resending things.
> I guess that would indeed be a bug if it was the reply.[/color]
Unprecedented in my experience, but yes it would be a bug.