On Sun, 31 Aug 2003, Tom Daly wrote:
> 2003-08-29 13:41:09 19sl4v-000MP6-RJ => ingo@??? R=handle_relay
> T=port_redir_smtp24 H=gimpy2.xxxxxx.com [xx.yy.zz.126]
> 2003-08-29 13:41:09 19sl4v-000MP6-RJ Completed
> 2003-08-29 13:41:10 19slQn-000ORH-5U => cliff@??? R=handle_relay
> T=port_redir_smtp24 H=gimpy2.xxxxxx.com [xx.yy.zz.126]*
> 2003-08-29 13:41:10 19slQn-000ORH-5U Completed
> 2003-08-29 13:41:10 19slKg-000OJz-BB send() to gimpy2.xxxxxx.com
> [xx.yy.zz.126] failed: Socket operation on non-socket
> 2003-08-29 13:41:10 19slKg-000OJz-BB == lenk@??? R=handle_relay
> T=port_redir_smtp24 defer (38): Socket operation on non-socket: send() to
> gimpy2.xxxxxx.com [xx.yy.zz.126] failed
>
> That happens in that order when there's > 2 deliveries to one host
> consistently, regardless of their mail server.
Mysterious. But to find out more, you'll have to set up a test with
debugging enabled (the -d option, catch the stderr output). You'll need
to put in three messages with the -odqs option so that they get queued,
but routed. Then run
exim -d -M id-of-one-of-them
and see what the debugging output says.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
Get the Exim 4 book: http://www.uit.co.uk/exim-book
