Götz Reinicke wrote:

:-) maybe now people understand, why i don't like to "test" test releases on a productiopn server.... ;-)

Regards

Götz

Timo Sirainen wrote:

...

http://dovecot.org/test/

Sorry :) The last fix I made a few minutes before test54 release didn't turn out to be exactly correct.

I totally undersand, which is why I'm still using 0.99.11 (and anxiously awaiting a stable release of 1.0 series)

alan

John Peacock wrote:

Still doesn't fix the SIG11 problem with logging in using secure methods (plaintext login works fine).

I attached to the dovecot-auth process in gdb and tried to log in and this was the backtrace I got:

(gdb) bt #0 0x00000000 in ?? () #1 0x08053e7a in mech_cram_md5_auth_continue (auth_request=0x80911b0, data=0x80780b0 "testing@example.com 86feff0ff45af61accb41facb997740e", data_size=52, callback=0x8075134 ) at mech-cram-md5.c:145 #2 0x08050c58 in auth_client_input (context=0x808b9a8) at auth-client-connection.c:340 #3 0x0805e02a in io_loop_handler_run (ioloop=0x807cb90) at ioloop-poll.c:184 #4 0x0805d749 in io_loop_run (ioloop=0x807cb90) at ioloop.c:218 #5 0x0805281e in main (argc=1, argv=0x8075134) at main.c:296

Let me know if there is anything else I can do to help debug this...

John

-- John Peacock Director of Information Research and Technology Rowman & Littlefield Publishing Group 4501 Forbes Boulevard Suite H Lanham, MD 20706 301-459-3366 x.5010 fax 301-429-5748