Sunet Sysadmin wrote:
Hi,
i am trying to setup courier mail server. i am trying to figure out a
way that all mails comming will be filtered in by courier server and
forwarded to another mail server.
AFAIK, only global filtering does that. However, it cannot alter messages.
If the server
Hi all,
I currently can change html pages, but cannot easily add
features to webmail. By _easily_ I think I mean
- don't edit Courier's html templates,
- use PHP scripts if at all possible,
- integrated into Courier environment,
i.e. authentication, folders, etc.
One feature I'm thinking
Papis wrote:
Hi all,
I want create some mailling list in courier mta using couriermlm.
I create the directory of mailling liste :
couriermlm create directory [EMAIL PROTECTED]
I create the two required dot file :
.courier-list
.courier-list-default
Hi all,
the courierdeliver local module checks if the message
already contains a Delivered-To header with the same
recipient that it is about to deliver to. Is that needed?
That feature requires a wrapper around sendmail for resending
messages from command line. I use sed with something like:
Sam Varshavchik wrote:
Alessandro Vesely writes:
the courierdeliver local module checks if the message
already contains a Delivered-To header with the same
recipient that it is about to deliver to. Is that needed?
Yes.
Ok, thanks. Sorry for the silly question. I should have found a long
Sam Varshavchik wrote:
Alessandro Vesely writes:
Sam Varshavchik wrote:
Alessandro Vesely writes:
Sep 12 07:21:12 wmail courierlocal:
id=000CA6A9.450554B7.1F80,from=[EMAIL PROTECTED],addr=[EMAIL PROTECTED]:
Invalid address.
Sep 12 07:21:12 wmail courierlocal:
id
Hi,
my log has these:
Sep 12 07:21:12 wmail courierd:
started,id=000CA6A9.450554B7.1F80,from=[EMAIL
PROTECTED],module=local,[EMAIL
PROTECTED]/export/courier/tana.it/mylist!!,addr=[EMAIL PROTECTED]
Sep 12 07:21:12 wmail courierd: Waiting. shutdown time=none, wakeup time=none,
Sam Varshavchik wrote:
Alessandro Vesely writes:
Sep 12 07:21:12 wmail courierlocal:
id=000CA6A9.450554B7.1F80,from=[EMAIL PROTECTED],addr=[EMAIL
PROTECTED]:
Invalid address.
Sep 12 07:21:12 wmail courierlocal:
id=000CA6A9.450554B7.1F80,from=[EMAIL PROTECTED],addr=[EMAIL
Cheng Bruce wrote:
Would you please advise me if it is good enough to run in our
company(production) ?
There's a number of things I don't understand.
logfile /home/domains/maildrop.log
VERBOSE 9
DEFAULT=$HOME/Maildir
TEST=/usr/bin/test -f
CUSTOM_FILTER=$HOME/.mailfilter
if (
Wojtek Swiatek wrote:
A real life example would be: all emails sent to [EMAIL PROTECTED] should
be bounced to [EMAIL PROTECTED] I would like the messages, as seen on the
gmail.com account, to look like as if they were sent by the original sender
That may violate SPF: in case the original
Sam Varshavchik wrote:
Georg Lutz writes:
On 2006-09-06 04:37, Sam Varshavchik wrote:
Arturo 'Buanzo' Busleiman writes:
Sam Varshavchik wrote:
There's no point, whatsoever, in a pre-data hook. The only thing known
at that time is the sender's IP and return address. The return address
Shawn M. Jones wrote:
[...]
I also let 'error' for both FROM fields.
After you mentioned this I saw error for some legitimate mail in the
maillog. I've added that to the list too.
BOFHCHECKDNS should affect that too
Well, thank you Alessandro for clearning a lot of this up for me. I've
Shawn Jones wrote:
I've set the following options in /etc/courier/bofh:
opt BOFHSPFHELO=all
opt BOFHSPFMAILFROM=pass,neutral,none,softfail,unknown
opt BOFHSPFFROM=pass,neutral,none,softfail,unknown
opt BOFHSPFHARDERROR=fail,softfail
opt BOFHSPFTRUSTME=1
opt BOFHSPFNOVERBOSE=1
Issue #1:
Bernd Wurst wrote:
If the above is not true, you can use custom query string (have a look at the
end of authmysqlrc) and use something like
[...] WHERE account=LOWER('$(local_part)@$(domain)') [...]
The default query selects WHERE login_field = input
so it depends ho you defined that
Hi Jeremy,
Jeremy Davis wrote:
Alessandro Vesely wrote:
The default query selects WHERE login_field = input
so it depends ho you defined that field (MYSQL_LOGIN_FIELD)
in the MySQL table:
CHAR values are sorted and compared in case-insensitive fashion
according to the default
Sam Varshavchik writes:
Alessandro Vesely writes:
Sorry, I meant
Nope, I don't think you meant this either.
Oh my gosh, it must have been a hot afternoon...
So I just meant
--- courier/module.local/local.original.c 2006-05-28 17:29:52.0
+0200
+++ courier/module.local
--- courier/module.local/local.original.c 2006-05-28 17:29:52.0
+0200
+++ courier/module.local/local.c2006-08-07 15:54:15.495792252 +0200
@@ -693,13 +693,13 @@
if (pid == 0)
{
char*argv[11];
+ int fd0 = fd;
close(0);
-
Sorry, I meant
--- courier/module.local/local.original.c 2006-05-28 17:29:52.0
+0200
+++ courier/module.local/local.c2006-08-07 16:08:02.727499591 +0200
@@ -693,13 +693,16 @@
if (pid == 0)
{
char*argv[11];
+ int fd0 = fd;
I added the following paragraph to dnsheader.txt:
In case the original message was not yours, your e-mail address
is probably being abused by spammers. To protect your e-mail
address, ask your network administrator to publish SPF information
for your domain: that way you will not
Sam Varshavchik wrote:
Alessandro Vesely writes:
I added the following paragraph to dnsheader.txt:
In case the original message was not yours, your e-mail address
is probably being abused by spammers. To protect your e-mail
address, ask your network administrator to publish SPF
Since --with-transport isn't documented any more, I configured
Courier without it. Now I have a running courieruucp.
Should I comment it out from module.uucp, i.e.
#PROG=./courieruucp?
TIA
-
Take Surveys. Earn Cash.
--- courier/libs/comtrack.original.c2005-01-23 03:34:46.0 +0100
+++ courier/libs/comtrack.c 2006-08-03 21:23:45.008312657 +0200
@@ -112,8 +112,10 @@
fp=fopen(namebuf, a);
if (fp)
+ {
fprintf(fp, %s %c%s\n, buf2, (char)status, addrbuf);
-
Hi,
I keep getting this message
# Checking listening processes
NEW: --WARN-- [lin003w] The process `couriertcpd' is listening on socket
57123 (UDP on every interface) is run by courier.
It looks bogus (otherwise it would have been named `courierudpd', eh?)
I asked on other lists to no
Sometimes configuring gets complicated, e.g. the BLACKLISTS
item in esmtpd, which changes quite often. I set it with
something like
ALIST=-block=blah...
ANOTHER=-block=...
# and then
BLACKLISTS=$ALIST $ANOTHER
Sysconftool silently discards anything after the first line.
OTOH, the patch
Sam Varshavchik wrote:
Alessandro Vesely writes:
Hi,
I keep getting this message
# Checking listening processes
NEW: --WARN-- [lin003w] The process `couriertcpd' is listening on
socket 57123 (UDP on every interface) is run by courier.
It looks bogus (otherwise it would have been
Sam Varshavchik wrote on Thu, 25 Aug 2005 18:32:25 -0400:
Stefan Hornburg writes:
Hello,
I'm the Debian maintainer of Courier and try to build Debian packages
for courier-authlib 0.57 / courier 0.51 on Debian sarge.
courier-authlib 0.56 / courier 0.50 builds worked fine.
I appreciate
Sam Varshavchik writes:
Alessandro Vesely writes:
Sam,
AC_CHECK_FUNC looks for functions and explicitly avoids macros,
which fails with BDB 4.xx versions that decorate function names and
#define db_env_create db_env_create_40xx.
The patch attached apparently works fine, after installing
Sam,
AC_CHECK_FUNC looks for functions and explicitly avoids macros,
which fails with BDB 4.xx versions that decorate function names and
#define db_env_create db_env_create_40xx.
The patch attached apparently works fine, after installing both
Lindsay Haisley writes:
I can always query the database (authmysql) directly, but since I'm going to
publish this I'd rather at least use a method which will work with courier
across any auth mechanism supported by the current courier configuration.
I have a script doing that using
Ricardo Kleemann wrote:
Anyone have another one to add to cbl.abuseat.org?
It's included in spamhaus' XBL
http://www.spamhaus.org/xbl/
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Ricardo Kleemann writes:
So using
BLACKLISTS='-block=sbl-xbl.spamhaus.org,BLOCK'
should take care of it?
Yup! You can add a 3rd var as, e.g.,
BLACKLISTS='-block=sbl-xbl.spamhaus.org,BLOCK,550 Rejected - see
http://www.spamhaus.org/query/bl?ip=@;'
I don't know how they do. Sometimes I
Ricardo Kleemann wrote:
Thanks.
the '@' returns an ip with a ::: prefix. Is there any way to print out
the reject with only the IP?
Ooops, I didn't mind it... I disable IPv6 at build time.
BLACKLISTS='-block=sbl-xbl.spamhaus.org,BLOCK,550 Rejected - see
[EMAIL PROTECTED] wrote:
Please !!! Help ... I am tired of compiling with make , gmake , dmake , sun's
and gnu LD's ...
GNU make is a perfectly working, POSIX compliant make.
Version 3.81 compiles and works out of the box even on Windows.
(I've only seen one package, ipf, suggesting not to
Sam Varshavchik wrote:
Alessandro Vesely writes:
However, why shouldn't webmail just discard unwanted cookies,
rather than storing them in the environment?
All cookies get stuffed into a single header. The header needs to be
parsed first, but this can't be done at this early stage
Sam Varshavchik wrote:
Alessandro Vesely writes:
Sam Varshavchik wrote:
Alessandro Vesely writes:
However, why shouldn't webmail just discard unwanted cookies,
rather than storing them in the environment?
All cookies get stuffed into a single header. The header needs to be
parsed first
Tuomas Silen writes:
Hi,
There seems to be a bug in couriertls (~every version of courier-imap) that
causes it to segmentation fault if there are files with no dots in the
certificate directory (peer_cert_dir). Openssl likes to put files like Makefile
there for some strange reason.
Problem
Gordon Messmer wrote:
JP wrote:
Sam, if someone else were to write an acceptable patch that updated
SQWebmail to handle larger cookies, would you be willing to accept it
into the main trunk?
While making that argument, consider that allowing greater input means
that you may increase the
Josh Grebe wrote:
Hello,
I have been comtemplating writing a pre-DATA hook into courier to allow for more
efficient greylisting. I guess it would actually hook into the RCPT TO: handler.
That is much more lightweight for the receiving server.
The biggest problem with the current method (in
Sander Holthaus wrote:
[...]
Those md5's should be done carefully, as a sender might rewrite MIME
boundaries on the fly, assigning random values as required by the RFC.
Do those need to change on every retry as well? E.g. are they
rewritten / parsed as the message is send?
They _may_
Alessandro Vesely wrote:
Lloyd Zusman wrote:
Lloyd Zusman [EMAIL PROTECTED] writes:
[ ... ]
But now, I have a new challenge: coming up with a patch/enhancement to
the filtering mechanism which will cause a message to be accepted
without any further filtering. In other words, every filtering
Sam,
I don't use ldap, however ldapaddressbook failed to configure
and compile. Apparently, solaris has a problem with ldap.h:
it cannot be compiled unless lber.h is included before it.
Perhaps the patch attached may work...
diff -du courier-0.53.1.20060318/ldapaddressbook/configure.in
Lloyd Zusman wrote:
Lloyd Zusman [EMAIL PROTECTED] writes:
[ ... ]
But now, I have a new challenge: coming up with a patch/enhancement to
the filtering mechanism which will cause a message to be accepted
without any further filtering. In other words, every filtering step
would result in a
Sam Varshavchik wrote:
[...] All filters must
give a green light, so it's not important which filter rejects it.
It may make sense to order them for efficiency reasons?
---
This SF.Net email is sponsored by xPML, a groundbreaking
Bowie Bailey wrote:
Alessandro Vesely wrote:
Bowie Bailey wrote:
Lyndon Tiu wrote:
I wrote a maildroprc:
-
if \
( \
/^To:.*all.*/ \
\
( \
( /^From:.*user1.*/ ) \
\
( /^From:.*user2.*/ ) \
) \
)
{
to /mail/junk/Maildir
A few mumblings about that patch
Lloyd Zusman wrote:
[...]
+#define FILTER_LIST_INCREMENT 8
+
+static char **filterlist = NULL;
+static int filterlistsize = 0;
+static int nfilters = 0;
+
+static void free_filters()
+{
+ if (filterlist != NULL)
+ {
+ for (int
Bowie Bailey wrote:
Lyndon Tiu wrote:
I wrote a maildroprc:
-
if \
( \
/^To:.*all.*/ \
\
( \
( /^From:.*user1.*/ ) \
\
( /^From:.*user2.*/ ) \
) \
)
{
to /mail/junk/Maildir
}
-
Now the maildroprc regex
Why don't we use a default-deny approach and DNS whitelists?
That's the theme of a thought provoking article by Meng Wong:
http://www.circleid.com/posts/internet_governance_an_antispam_perspective/
---
This SF.net email is sponsored by: Splunk
Sam Varshavchik wrote:
Jay Lee writes:
Sam could say better why 4xx errors aren't possible with .smtpfilter.
A message may have multiple recipients.
Couldn't a filter be aware of the number of recipients
and have more actions granted in case the number is 1?
What for?
A possible example
Dirk Kulmsee wrote:
E.g. [EMAIL PROTECTED] fills out the mailform and the mailform generates a mail
to
the company's responsible recipient who has her/his mail account on some
other mail server on this planet. The mailform uses [EMAIL PROTECTED] as
sender
address to give the recipient a
Michael Mattess wrote:
[...]
First we looked at a plane installation, we limited our selves to mail
retrieval so we only used courier-pop and mounted the maildirs via NFS
[...] i modified Courier-pop to use a PostgeSQL DB instead of the
maildirs and the results were quite surprising.
It
[EMAIL PROTECTED] wrote:
Hi,all
does Courier MTA has a big picture like as qmail? http://www.nrg4u.com/.
If you mean A4-sized approx. 50K gifs, then no, not really. Courier doc
is cool readable html. In http://www.courier-mta.org/queue.html there is
an ascii crude diagram, though.
Chris Petersen wrote:
Yeah. VPN and just host the imap server on our dsl line.
It's fast enough for the few times that people will be
checking their mail from home.
Fine. That appears to be the best practice (with minor worryings about security)
See also thread Best practices for
Chris Petersen wrote:
[...]
Our goal is to set up one server at our colo for fast handling of
incoming mail. This server would do spam/virus scanning/blocking to
reduce the download load on our in-office DSL connection. A second
on-site server would be set up to receive mail from the first
Sam Varshavchik wrote:
Ricardo Kleemann writes:
However, isn't the mail server supposed to accept for DSNs ?
Correct.
Is
there a
way to get around this?
No, there's no way to get around an idiot who is placed in charge
Sam Varshavchik wrote:
Alessandro Vesely writes:
E.g. checking xxx entities terminate with `;'?
But there's no _hot_ security concern, is there?
Just being proactive here -- in case MSIE, or something else --
misparses it in a way that could be exploited:
fooscript...
Thank you so
Sam Varshavchik wrote:
Although in the rest of the world, any !-- comment -- in HTML gets
ignored, with MSIE a specially formatted HTML comment can get processed
as regular HTML code, with scripting, et al:
Sam Varshavchik wrote:
This release of SqWebMail filters out certain MSIE-only scripting
constructs that could be used for malicious purposes. As an
alternative: a cumulative patch [...]
Apparently, that patch is not related with the downloading of viral
attachments that has been recently
Sam Varshavchik wrote:
Alessandro Vesely writes:
Sam Varshavchik wrote:
PCRE will be able to support all of maildrop's functionality except
for the w pattern option. With PCRE, the w option would require
the entire message to be loaded into memory. Maildrop does not load
large messages
Sam Varshavchik wrote:
PCRE will be able to support all of maildrop's functionality except for
the w pattern option. With PCRE, the w option would require the
entire message to be loaded into memory. Maildrop does not load large
messages into memory, and this would not be desirable.
Bernhard Lukas wrote:
[...]
BLACKLISTS=-block=blackholes.mail-abuse.org,BLOCK
-block=relays.mail-abuse.org,BLOCK
[...]
Nevertheless I am still getting much SPAM and often from the same
address - does the blacklist really work?
No. To check that, you should try the query by hands, e.g.
dig
Vlad,
a similar question was discussed last year, you may
want to check the thread
http://sourceforge.net/mailarchive/message.php?msg_id=7110106
Vlad Dinulescu wrote:
Thanks for the quick answer!
How can I modify a message from the filter? Like
adding a header, or modifying the body?
Sam
Rodrigo Severo wrote:
[...]
I'm wondering if it might be better to:
a) not to deal with it at all because there might be so few cases
Appealing as this option can be, to make an informed decision we still
need a sharp view of what we don't want to deal with. Your clear analysis
of the
Sam Varshavchik wrote:
A real DNS server returns records in random order. That's a fundamental
function of DNS: load balancing.
Is it random or round robin?
For multihoming, load balancing should imply that querying from net A
results in (at least averagely) different addresses than
Malcolm Weir wrote:
[...]
The second MX is _strictly_ a fallback system. It will store and forward
messages for the primary, but it doesn't do the SMTP-level message
rejection. So it will accept pretty much anything sent to it, and then
shovel that on to my primary when the primary
uchu wrote:
If someone wants to run a mailing list they should use a mailing list
management software that's designed for that purpose, instead of
trying to shoehorn something out of an ordinary E-mail client.
Totally agreed on that.
The lack of functionality that had been objected
Julian Mehnle wrote:
Alessandro Vesely wrote:
Julian Mehnle wrote:
pureperlfilter is the bootstrapping executable for Courier::Filter.
C:F cleans up its socket when shutting down (e.g. when told to by
courierfilter), but refuses to start if the socket it wants to create
already
Julian Mehnle wrote:
Alessandro Vesely wrote:
Ben Kennedy wrote:
In this example, pureperlfilter is cleaning up its socket (though
sometimes for whatever reason the socket stays around causing it to
refuse to launch subsequently); perlfilter is not removing its.
IMHO, the latter
Ben Kennedy wrote:
Hey all,
I've been having a slightly annoying issue with mail filters.
Specifically, doing 'courierfilter stop' or otherwise having courier stop
or restart leaves sockets kicking around, leading to socket already
exists log messages and/or 432 Mail filters temporarily
Bowie Bailey wrote:
[...] Since there are no DNS records for the
domain, it is impossible to deliver directly to this address from
anywhere else.
H... except for purposely built malaware?
---
SF email is sponsored by - The IT Product
Lucio wrote:
Alessandro Vesely wrote:
Courier already has a big-brother option.
Yes, I know. But I need the users to be aware of the fact that their messages
are being sniffed, so I have to force them to send CCs of thier messages to
the big-brother address.
If you want users to type
/
http://www.catb.org/~esr/fetchmail/
Good Luck!
Original Message
Date: Tue, 22 Feb 2005 12:55:29 +0100 (CET)
Subject: Migrating to courier IMAP (Re: [courier-users] Filtering messages)
From: Martijn Lievaart [EMAIL PROTECTED]
To: Alessandro Vesely [EMAIL PROTECTED]
Alessandro
Lucio wrote:
[...]
Rejects all outgoing messages if they do not specify
[EMAIL PROTECTED] in one of the headers.
Courier already has a big-brother option. Check it out in etc/courierd
before you proceed. If you are going to test a filter, setting an
ARCHIVEDIR is a convenient way to debug
Gordon Messmer wrote:
Sam Varshavchik wrote:
I think that ctlfile should be opened for append mode.
...capital idea. Am I wrong to assume that all of the calls to open()
in the openctl() function should have the same flags?
I'm not sure what you mean. And I wonder what sense may have
Julian Mehnle wrote:
Jay Lee [EMAIL PROTECTED] wrote:
Julian Mehnle said:
If it's a cost issue, run just one (i.e. drop the differently
configured secondary MX). That's what I actually do.
Have you ever had issues with downtime or lost mail?
Nothing where a secondary MX would
Gordon Messmer wrote:
Sam Varshavchik wrote:
It's always safe to append more stuff to the control file. Whether
it's safe to rewrite it from scratch is something I need to think about.
I thought it was, too. However, when I started actually testing that
theory, I found that it's
this mean you are missing the crypt function?
Would you mind compiling the attached test program to check that?
(As usual, you may compile it with/without, say, -D_XOPEN_SOURCE=400
see man standards -shouldn't really change much in this case.)
Alessandro Vesely wrote:
David K. wrote:
I have
Hi David,
David wrote:
Hi Alessandro,
I just realized that my email client was translating some of the test
program source as smileys.
:-)
I have saved the code as an attachment and it
compiles fine as you sent it.
gcc crypttest.c
crypttest.c: In function `main':
crypttest.c:8:
and try rebuilding
with that. To reconfigure you should amend the top level configure so that
it reads
case `$LD -V 21` in
instead of
case `$LD -V` in
at line 25767.
Good luck!
Ale
- David
Alessandro Vesely writes:
Hi David,
David wrote:
Hi Alessandro,
I just
David K. wrote:
[...]
pstack core
core 'core' of 17227: /usr/local/libexec/courier-authlib/authdaemond
d26cf539 strcmp () + 115
d2793d3d authcheckpassword (804d234, d254bba6, d277bb7b, d253e724) + 101
d253e73d ()
Possibly that is a strcmp in do_authcheckpassword, comparing
', `print' and `quit' as needed. See e.g.
http://sources.redhat.com/gdb/current/onlinedocs/gdb_9.html
- David
Alessandro Vesely wrote:
David K. wrote:
[...]
pstack core
core 'core' of 17227: /usr/local/libexec/courier-authlib/authdaemond
d26cf539 strcmp () + 115
d2793d3d
chester c young wrote:
does anyone have any idea on how to have ssl with multiple domains?
Must use different ip addresses. That's rooted in the ssl handshake:
the client does not send the name of the (virtual) host before it
receives the server's certificate.
alekosr wrote:
2) Is it possible to forward a copy of all emails to a single email account ?
There is an ARCHIVEDIR variable you can set in courierd (the big-brother
option).
---
The SF.Net email is sponsored by: Beat the post-holiday
Hi Sam,
I was browsing the docs when I found
auth_sasl initializes *authtype_ret and *authdata_ret.
They will be set to a malloc(3)-ed buffers that can be
directly passed as arguments to auth_generic(3). It is
the application's responsibility to free(3) these
buffers when it's done with
Toby Heywood wrote:
[...]
MySQL Log
- --
050101 12:52:23 5 Query SELECT id, crypt, , uid, gid,
home, , , name, FROM passwd WHERE id = [EMAIL PROTECTED]
Is mandc the correct domain name? I get a strage MX record for it.
Sure the MySQL log lines originated from incoming
Sam Varshavchik writes:
Alessandro Vesely writes:
replacing
-Wl,--whole-archive with -Wl,-z -Wl,allextract, and
-Wl,--no-whole-archive with -Wl,-z -Wl,defaultextract.
That's a job for libtool, as (not much extensively) exemplified in
http://www.gnu.org/software/automake/manual
Hi Sam,
I hope that requiring GNU ld is a temporary feature of courier-authlib.
The GNU loader has traditionally been buggy on solaris-sparc
and requiring it may jeopardize the portability of Courier.
It is only the line that builds libcourierauthcommon.la that fails.
I rewrote that command line
Julian Mehnle wrote:
In the old times of the Internet where a.com (everybody) could send a
message and claim it to come from x.com, forwarders would have to take no
responsibility for what domains are used as the sender addresses of the
mail they forward. As a result, everybody could simply
Lloyd Zusman wrote:
% deliveryloc [EMAIL PROTECTED]
/home/uname/Maildir
What about
# authtest [EMAIL PROTECTED]
Authenticated: module authdaemon
Home directory: /home/uname
UID/GID: 12345/12345
[EMAIL PROTECTED]
AUTHFULLNAME=User Full Name
Not (always) installed, no
The maxlongsize.h header used to define the max size of
a decimal representation of the largest long int
using the size of its string representation as given in
limits.h, i.e. (simplified slightly)
#define MAXLONGSIZE MAXLONGSIZE1(ULONG_MAX)
#define MAXLONGSIZE1(x) MAXLONGSIZE2(x)
#define
Sam Varshavchik writes:
Alessandro Vesely writes:
I think authtest might also be good as a generic configuration tool
as it delivers name, home directory, uid and gid of a given user.
(Suppose a configure script takes an option like [EMAIL PROTECTED]
for installing some data
Sam Varshavchik wrote:
[...]
Overview of the planned courier-authlib package:
1. Uses the existing configure options.
2. Uses standard FHS install paths.
3. Installs:
A) Various authdaemond builds, and the authdaemond startup script
B) Configuration files
C) Test
Isn't this the same as
Problem (and patch for) 534 Message header size, or ... error
22 Jan 2004 05:51:42 +0100
http://sourceforge.net/mailarchive/message.php?msg_id=7102594
[EMAIL PROTECTED] wrote:
Hello!
I would like to know why the real maximal length of To: header is
only 5000
That's an interesting idea!
Jason L. Buberel wrote:
After reading through the on-line man pages of courierfilter,
localfilter, and maildrop, I'm left wondering how I would go about
implementing a mail filter that processes all outbound messages.
AFAIK maildrop is only used for messages
Michelle Konzack wrote:
[...]
OK, now I know it is /usr/sbin/courierfilter which produce the error
Unable to reserve file descriptor 3.
I'd be curious as to why that happens. As fd's 0, 1, and 2 are in use,
the sequence
close(3);
if (open(/dev/null, O_RDONLY) != 3
Michelle Konzack wrote:
Am 2004-05-25 10:04:38, schrieb Alessandro Vesely:
cannot fail. Can you reproduce the failure running the program by itself?
yes
If that's possible, truss can reveal what's going on.
Sorry, I said truss but I meant strace. If you run a line like
strace -f
Lindsay Haisley wrote:
Thus spake Alessandro Vesely on Fri, May 14, 2004 at 01:42:12AM CDT
Pardon my intrusion, but this option looks like requiring two servers:
one to accept incoming mail from any stupid client and the other for
filtering e-mail according to varying recipients
Sam Varshavchik writes:
Alessandro Vesely writes:
Sam, may I ask if filters are going to be redesigned one day? In this
case it would seem more polite to sprout two control files, similarly
to what is being done with 200 recipients...
I don't follow you. Splitting the message
Sam Varshavchik wrote:
[...]
Now, one change that can be made is to disable all filtering completely if
RELAYCLIENT is set. This is going to disable all content filtering for all
senders with relaying privileges. I'm just not sure if we want to do that.
Pardon my intrusion, but this option
Sander Holthaus - Orange XL wrote:
I've been semi-succesfull in setting up Courier and Amavisd-new. Everything
works, except that amavisd-new (running as a courierfilter) cannot seem to
add any headers to the message. I don't have a clue where to look for a
solution, besides the config-file
Sam Varshavchik wrote:
It looks to me like your patch reinvents the wheel.
I'm not sure why wheels have to be reinvented every now and then, but
on the ChangeLog I found the following:
http://cvs.sourceforge.net/viewcvs.py/courier/courier/courier/ChangeLog?rev=1.600view=markup
: 2002-08-01 Mr.
501 - 600 of 648 matches
Mail list logo
