Re: RPC clnt_create() adress already in use

["PAULUS, Raimund, TI-ABN" <Raimund dot Paulus at dillinger dot biz>]

Hi Mark,

in my email (https://sourceware.org/ml/cygwin/2017-12/msg00194.html) i described 2 approaches. I prefer  nr 1.
Here the part of the source in bindresvport.c:
------------------------------------------------------------------------------------------------------------------
        if (port == 0) {
                port = (getpid() % NPORTS) + STARTPORT;
        }
        res = -1;
        errno = EADDRINUSE;

/* fix for bind() */
port = 0;
                again:
        for (i = 0; i < nports; ++i) {
                *portp = htons(port++);
                 if (port > endport)
                        port = startport;
                res = bind(sd, sa, salen);
                if (res >= 0 || errno != EADDRINUSE)
                        break;
        }
        if (i == nports && startport != LOWPORT) {
            startport = LOWPORT;
            endport = STARTPORT - 1;
            nports = STARTPORT - LOWPORT;
            port = LOWPORT + port % (STARTPORT - LOWPORT);
            goto again;
        }
        mutex_unlock(&port_lock);

        return (res);
}
-------------------------------------------------------------------------------------------------------

This causes bind() to search an unused port. I use libtirpc with this fix since several weeks and it works for me. I don't know an other way (fixing Cygwin) to success.
The RPC-client on my pc is started every few minutes and has to connect to the RPC-server.  Without the fix libtirpc is not usable and I have to use Cygwin 1.5.18 with the old librpc.

Raimund


-----Ursprüngliche Nachricht-----
Von: cygwin-owner@cygwin.com [mailto:cygwin-owner@cygwin.com] Im Auftrag von Mark Geisert
Gesendet: Dienstag, 30. Januar 2018 10:05
An: cygwin@cygwin.com
Betreff: Re: RPC clnt_create() adress already in use

PAULUS, Raimund, TI-ABN wrote:
> Hi Mark,
> as Corinna wrote in the thread " Cygwin socket option SO_REUSEADDR operates unlike Linux", the cause for the problem in libtirpc is Microsofts implementation of socket handling. If i implement a simple socket client, that is not so much of an issue, because I call the functions separately and can handle the error. Moreover in this case bind() is not necessary.
> But in the case of libtirpc the implementation of clnttcp_create() determines the sequence of bind() and connect() for the local port and you cannot correct the error of bind() after the call of connect(). Therefore I think best is to set the port number to zero (port = 0) before the call to bind().
>
> Raimund
>
>
> -----Ursprüngliche Nachricht-----
> Von: cygwin-owner@cygwin.com [mailto:cygwin-owner@cygwin.com] Im 
> Auftrag von PAULUS, Raimund, TI-ABN
> Gesendet: Dienstag, 9. Januar 2018 09:14
> An: cygwin@cygwin.com
> Betreff: WG: AW: RPC clnt_create() adress already in use
[...]

Hi Raimund,
Where, exactly, would you set the port number to zero?  I can imagine how this would solve the problem you're seeing but I'm concerned about other use cases.

It also seems a little odd to me we'd be making a Cygwin-specific patch rather than "fixing" Cygwin itself so the patch would not be necessary.  But I don't feel strongly either way here.  libtirpc already has Cygwin-specific patches.

..mark


--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

Attachment: bindresvport.c
Description: bindresvport.c

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple