On Thu, 6 Oct 2005, Daniel Tiefnig wrote:
> There's an error mismatch in test 2027:
>
> GnuTLS/2027 TLS server & client: no certificate in client
> ===============
> Line 6 of "test-mainlog-munged" does not match line 6 of "log/2027".
> ----------
> 1999-03-02 09:44:33 10HmaY-0005vi-00 TLS error on connection to
> ip4.ip4.ip4.ip4 [ip4.ip4.ip4.ip4] (gnutls_handshake): A TLS packet with
> unexpected length was received.
> ----------
> 1999-03-02 09:44:33 10HmaY-0005vi-00 TLS error on connection to
> ip4.ip4.ip4.ip4 [ip4.ip4.ip4.ip4] (gnutls_handshake): Error in the push
> function.
Hmm. I wonder if that is something to do with different GnuTLS versions?
I am using GnuTLS 1.2.3.
> And do you remember the exim process, that was started with "-t" and
> "-bd"? This was caused by '-t' in $optargs when the runtest script is
> called with "-t <num>". I don't think that's intended, is it? :o)
Well no. It doesn't make sense to call the test script with -t. I
guessed that you had used -t instead of -d by mistake when I looked at
this, so I then ignored it. (The Exim option -t means "accept a message,
taking the envelope recipients from the To: and Cc: header lines".) And
neither "-t <num>" nor "-d <num>" are valid options sequences for Exim.
I should probably document what options are sensible to use; perhaps
only -d actually makes sense.
> There's a minor bug around line 1418 in the runtest script if it's
> called as "perl runtest":
>
> | $cwd = $0;
> | $cwd =~ s/\/runtest$//;
> | chdir($cwd) || die "** Failed to chdir to \"$cwd\": $!\n";
>
> Of course there's no '/' in 'runtest', so we try to cwd to 'runtest'. A
> quick and dirty fix would be to substitute like this:
>
> $cwd =~ s/(\/)?runtest$/$1./
Why doesn't everybody run things the way I want them to? <grin>
--
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
