headers
Odhiambo G. Washington | 1 Nov 2003 09:32

Re: vexim errors

* Michael Ludwig <listuser <at> swanni-und-michi.de> [20031101 01:33]: wrote:
>
>
> Hi Avleen,
>
> "Avleen Vig" <lists-exim <at> silverwraith.com> wrote:
>
> > Take a look at vexim, if you don't mind running a mysql (or other)
> > database to store this data in.
> > The people on the mailing list are very helpful too.
>
> As I subscribes to an mailinglist (vexim <at> silverwraith.com and I do not know
> if that's the right one) and postet to it, noone did answer and I still got
> no other mail from this list.
> So I think this list must be user-dead, as well as it's archives on silverwraith.com ?

Hmm, I wonder if Avleen has a list there. But if you "subscribed" then it is
really there. Never saw it though.

> > I wrote my own solution (vexim) :-)
>
> As you, Avleen, claim to be one of the maintainers of vexim, I'm going to give
> this one here a try, maybe I can reach you on this list.

Avleen is really a maintainer, not just a claim. However, for now he's busy with
some "very important" matters and so I don't quite think he'll respond sooner.
I'll try to respond on his behalf ;)

> So my question is why vexim does not work at all.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kevin Reed | 1 Nov 2003 18:41
Favicon

W32/Mimail.c <at> MM ??

This looks basically like an non-issue so far as I can tell.  I've been
watching for, but have not really seen any of these.

The only ones I saw was at work and those were bounces before the Exchange
system got updated to see it.  They arrived via a sendmail system in
another division and went directly to the exchange server.  I saw traces
of them when it bounced them back to the hub server as non-deliverable but
I didn't get a chance to actually look at one.  I was off yesterday when
it hit the Exchange servers. By the time I knew about it, there was
nothing to look at.

My guess is that there is something in them that is getting stomped on
close to connection time like perhaps no PTR or a stupid HELO of our
system name type thing.   Those types of catches don't provide me with an
indication of who the mail was supposively for etc..

Anyone having to deal with these?

--
Kevin W. Reed - TNET Services, Inc.
Unoffical Exim MTA Info Forums - http://exim.got-there.com/forums

--

## List details at http://www.exim.org/mailman/listinfo/exim-users Exim details at
http://www.exim.org/ ##
Permalink | Reply | 0 comments |
headers
Avleen Vig | 1 Nov 2003 20:13

local_part in exiscan

Obviously $local_part cannot be expanded in
acl_check_content/acl_smtp_data, because there might be multiple
recipients.

The workaround to this I'm considering is using exiscan to pass app
messages to spamassassin and clamav, and later use routers to accept or
reject messages.

But I'm not too sure how to set these routers up - I have a look through
the archives, but I can't seem to find any examples. Does anyone have
suggestions?

--
Avleen Vig
Systems Administrator
Personal: www.silverwraith.com
EFnet:    irc.mindspring.com (Earthlink user access only)

--

## List details at http://www.exim.org/mailman/listinfo/exim-users Exim details at
http://www.exim.org/ ##
Permalink | Reply | 0 comments |
headers
Craig Jackson | 2 Nov 2003 03:03

ldap aliases

I am having some trouble finding the right solution to setting up user
aliases in virtual domains.

In the ACL, I verify aliases recipients like so......
${lookup ldapm {user="cn=admin,dc=localsurface,dc=com" \
pass=PASS ldap:///ou=${quote_ldap:$domain},dc=localsurface,\
dc=com?mailAlias?sub?}}

(which looks to me like an expensive look up)

But then this email fails all conditions in subsequent routers so
bounces. In my ldap directory structure, mailAlias is an attribute
string that is the local part of the email address alias. How then can
this address be routed properly?

I was thinking address rewriting, but how? I also thought about ldap
symbolic link but found no good documentation about that.

BTW, I did implement aliases another way that worked where the aliases
were not attributes but were their own RDNs. That looks good on paper
until you get 500 users with 2 aliases each and quickly it's
unmanageable. The aliases need to be attributes to the cn where all of
the other user specific data is kept AND needs to be able to handle
multiple returns, I guess using ldapm lookup.

Any  help deeply appreciated,
Thanks,
Craig

--
(Continue reading)

Permalink | Reply | 0 comments |
headers
Avleen Vig | 1 Nov 2003 06:42

Re: vexim errors

On Fri, Oct 31, 2003 at 09:26:32PM -0800, Avleen Vig wrote:
> > As I subscribes to an mailinglist (vexim <at> silverwraith.com and I do
> > not know if that's the right one) and postet to it, noone did answer
> > and I still got no other mail from this list.  So I think this list
> > must be user-dead, as well as it's archives on silverwraith.com ?

Nope, I've just been busy with life, and I haven't set up the archives
yet.

> > I don't mind changing anything in my php.ini, but vexim is very very
> > badly documented.

Most peopel find it sufficiently documented to get started, but OK. I'll
try harder.

I'm sorry you had problems with the interface. A much updated version
will be out, probably this weekend.

For the archives, I would like to request that people mail vexim issues
to the vexim list, rather than exim-users, as probably only a small
handful of people currently use the interface.
I do my best to reply, but with a baby arriving next week, I might be
slow to do so :-)

--
Avleen Vig
Systems Administrator
Personal: www.silverwraith.com
EFnet:    irc.mindspring.com (Earthlink user access only)
(Continue reading)

Permalink | Reply | 2 comments |
headers
Avleen Vig | 2 Nov 2003 16:45

Re: local_part in exiscan

On Sat, Nov 01, 2003 at 11:13:15AM -0800, Avleen Vig wrote:
> Obviously $local_part cannot be expanded in
> acl_check_content/acl_smtp_data, because there might be multiple
> recipients.
> The workaround to this I'm considering is using exiscan to pass app
> messages to spamassassin and clamav, and later use routers to accept or
> reject messages.
> But I'm not too sure how to set these routers up - I have a look through
> the archives, but I can't seem to find any examples. Does anyone have
> suggestions?

Replying to my own message.. hmmm :)

Ok to make this a little more sane, I have the following router:

ditch_spam:
  driver = redirect
  allow_fail
  data = :blackhole:
  condition = ${if >{$spam_score_int}{${lookup mysql{select sa_refuse * 10 from users,domains \
                where localpart = '${quote_mysql:$local_part}' \
                and domain = '${quote_mysql:$domain}' \
                and users.spamassassin = '1' \
               and users.domain_id=domains.domain_id }}} {yes}{no}}
  local_part_suffix = -*
  local_part_suffix_optional

Unfortunately this seems to :blackhole: mail where the mysql lookup has
NOT returned a result. To make matters worse, it only :blackhole:'s
sometimes, not all the time.
(Continue reading)

Permalink | Reply | 2 comments |
headers
Adam Thornton | 2 Nov 2003 18:22

Forgetting delivery attempts....

Well, the reason I wanted expanded port strings was so that I could do
on the fly mappings from ports on VM's stack to arbitrary host/port
combos.

However, consider the following scenario--local ports I map are in the
range 50000-50049--I get a mapping of vm:50002 for a remote host and
port that has a transient error.

Now, when I retry that host, Exim is going to retry vm:50002--but 50002
is no longer hooked up to anything.  What I want to do is forget that
mapping and just run the transport again, so I get a new host (still vm)
and a new port (let's say it's 50000) this time.

I *think* that setting retry_data_expire to something ridiculously
short, like 1s, will fix this for me.  Is that a correct reading of the
spec?  So the next time that message gets looked at, we say, "oh, our
cached data has expired, let's ask again," at which point I get the same
host (because it's fixed in $hosts, and hosts_override is set) and a new
port (because generated with a ${run {...}} expansion)?

Adam

--

## List details at http://www.exim.org/mailman/listinfo/exim-users Exim details at
http://www.exim.org/ ##
Permalink | Reply | 3 comments |
headers
Dean Staff | 2 Nov 2003 19:04
Picon

Broken Aliases after upgrading from Exim3 to Exim4

Hi,
I recently upgraded from Exim 3 to Exim 4.2x
my problem is I have 1 e-mail account for my domain I treat as an
alias. It for a user who never needed POP access. I just added his
name to the /etc/aliases files to redirect his e-mail to his e-mail
address at his ISP. I never needed to create a physical account for
him.

Under Exim 3 it worked like a charm. But I just noticed that under
Exim 4 it is broken.
Messages for root are redirected to my personal account, and those
work, but every message addressed to the aliased e-mail fails.

I get the following error.
Delivery has failed on the enclosed message for the following
reasons reported either by the mail delivery system on the mail
relay host or by the local TCP/IP transport module:

   451 Temporary local problem - please try later

The Exim4 log has this entry

2003-11-02 12:25:21 H=(win95) [192.168.62.73] F=<dean <at> domain.ca>
temporarily rejected RCPT <john <at> domain.ca>: host lookup did not
complete

I have checked my exim4 conf.d and I have the following router entry
in /etc/exim4/conf.d/router/400_exim4-config_system_aliases
system_aliases:
  debug_print = "R: system_aliases for $local_part <at> $domain"
(Continue reading)

Permalink | Reply | 3 comments |
headers
Donald Thompson | 3 Nov 2003 00:37

Pam, smtp auth, and exim-3.36

I'm trying to get smtp authentication working on a new machine, without
much luck. I've got the same situation setup on an older machine (read:
older libraries, older pam, older exim version), but the same setup is not
working as expected. Seems pam won't let the exim user authenticate user
passwords, even if its authenticating off a password file that is readable
by the exim user.

Heres the relevant lines from the exim pam file:

auth       required     pam_pwdfile.so pwdfile /etc/mail/exim.passwd
account    required     pam_unix_acct.so

Heres the relevant lines from the exim config file:

plain:
        driver = plaintext
        public_name = PLAIN
        server_condition = "${if pam{$2:$3}{1}{0}}"
        server_set_id = $2
login:
        driver = plaintext
        public_name = LOGIN
        server_prompts = "Username:: : Password::"
        server_condition = "${if pam{$1:$2}{1}{0}}"
        server_set_id = $1

Now when I run exim as non-root, I get 'Authentication service cannot
retrieve authentication info'. When I run as root, it authenticates
correctly.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Andreas J Mueller | 3 Nov 2003 00:46

Re: Exim 4.24 lsearch lookup problems

Hi Michael!

>> condition      = ${lookup {$sender_address}
>> lsearch{/etc/exim/senders_without_disclaimer}{yes}{no}}

> thanks for your tip, that solved my problem.
> But why did my condition not work?
> (I'm trying to get this behaviour understood).

Your condition did not work, because there is nothing to look up in
your file.  It only contains keys, not values.  lsearch'ed files
usually look like this:

webmaster <at> swanni-und-michi.de: value1
listuser <at> swanni-und-michi.de: value2

value1 or value2 is being returned as the result of a lookup.

Since you only wanted to run a router if the sender address matches a
specific list of addresses, you could have simply used the "senders"
precondition, like this:

  senders = /etc/exim/senders_without_disclaimer

No need to do a lookup when there's nothing to look up. ;-)

Andy

--
(Continue reading)

Permalink | Reply | 0 comments |

Gmane