> so I suggest that you supply a remote email address to exim.
> And please, remember to reply not quoting my whole message, however
> short, it's a good habit.
>
> g
>
Thanks for the hints.
I found out the problem is exim crash whenever it tries to deliver an out-going remote email.
When I restart exim, I got signal 11 in the mainlog:
2004-10-25 16:26:56 queue run: process 15878 crashed with signal 11 while delivering 1CM08y-00047I-AV
2004-10-25 16:26:56 queue run: process 15879 crashed with signal 11 while delivering 1CLzpu-00042K-AD
I sent two msg to remote email account, so got two crashes.
I rebuilt exim with the following options:
in the Makefile:
WITH_SASLAUTHD= yes
WITH_AUTH_SASL= yes
#WITH_BDB_VER?=1
WITH_BDB_VER?=42
(Note, I changed BDB version to 42 as suggested by exim FAQ:
http://www.tp4.ruhr-uni-bochum.de/SoftwareDocs/exim-3.30/FAQ.html#SEC146
)
I also switched back to saslauthd, the undelivered out-going email does sent successfully, BUT fall back to the old problem - failed to smtp authentication when I try to send another email.
So in summary, the use of cyrus_sasl as authenticaitor does allow smtp auth, but caused Signal 11 on each out-going email. With saslauthd authenticator, it can send email to non-local email but failed to do smtp auth.
Here is the debugging msg on an out-going email and authenticated with cyrus_sasl (instead of saslauthd):
# exim -d -M 1CM0R4-00051X-9q
Exim version 4.43 (FreeBSD 5.3) uid=0 gid=0 pid=19321 D=fbb95cfd
Probably Berkeley DB version 1.8x (native mode)
Support for: iconv() IPv6 PAM Perl OpenSSL
Lookups: lsearch wildlsearch nwildlsearch iplsearch cdb dbm dbmnz dnsdb dsearch nis nis0 passwd
Authenticators: cram_md5 cyrus_sasl plaintext spa
Routers: accept dnslookup ipliteral manualroute queryprogram redirect
Transports: appendfile/maildir/mailstore/mbx autoreply lmtp pipe smtp
Fixed never_users: 0
changed uid/gid: forcing real = effective
uid=0 gid=0 pid=19321
auxiliary group list: 0
configuration file is /usr/local/etc/exim/configure
log selectors = 00000ffc 00010400
trusted user
admin user
skipping ACL configuration - not needed
Cyrus SASL knows about: NTLM:LOGIN:ANONYMOUS:PLAIN:GSSAPI:OTP:DIGEST-MD5:CRAM-MD5
Cyrus SASL driver sasl_cram_md5: CRAM-MD5 initialised
Cyrus SASL knows about: NTLM:LOGIN:ANONYMOUS:PLAIN:GSSAPI:OTP:DIGEST-MD5:CRAM-MD5
Cyrus SASL driver sasl_plain: PLAIN initialised
set_process_info: 19321 delivering specified messages
set_process_info: 19321 delivering 1CM0R4-00051X-9q
reading spool file 1CM0R4-00051X-9q-H
user=root uid=0 gid=0 sender=sam.wun@???
sender_fullhost = ([192.168.4.235]) [192.168.4.235]
sender_rcvhost = [192.168.4.235] (helo=[192.168.4.235])
sender_local=0 ident=unset
Non-recipients:
Empty Tree
---- End of tree ----
recipients_count=1
body_linecount=9 message_linecount=26
Delivery address list:
sam.wun@???
locking /var/spool/exim/db/retry.lockfile
locked /var/spool/exim/db/retry.lockfile
no retry data available
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Considering: sam.wun@???
unique = sam.wun@???
no domain retry record
no address retry record
sam.wun@???: queued for routing
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
routing sam.wun@???
--------> dnslookup router <--------
local_part=sam.wun domain=remote.com
checking domains
remote.com in "localhost:mydomain.com:mydomain.net"? no (end of list)
remote.com in "! +local_domains"? yes (end of list)
calling dnslookup router
dnslookup router called for sam.wun@???
domain = remote.com
Segmentation fault
Thanks
Sam
