this may sound kinda dumb, but aren't we just talking about a webmail
client for what you want to do here? is the need to have a local client
a requirement or a preference?
b
Serge Knystautas wrote:
David Schwartz wrote:
Unless you have some fancy authentication happening with your email,
for starters, i would suggest a '2b' for *IX users:
/*
2. Extra the gz or zip file to a local directory.
2b. (IX users) Make run.sh and phoenix.sh executable (`chmod +x ruh.sh
phoneix.sh`)
3. Start bin/run.sh (unix) or bin\run.bat (windows) You will see
something like the following:
*/
also
Ok, I can tell them to check that locally this resolution matches.
Usually the problem seemed to be the outside world not having problems,
but I would like to catch as many of these as possible.
I guess if you have NAT-issues you could have problems because that
external result won't match
been playing with the latest [binary] build. you might want to kill this
file:
/opt/james/apps/dummy.txt
since james.sar is in that dir.
b
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL
it seems like this would make more sense in the phoenix tree since you
are technically starting and stopping phoenix, correct? (which is why i
think that there is always confusion as to how one 'stops' james).
b
Danny Angus wrote:
Is there any reason why this can't replace run.sh ?
having one
Danny Angus wrote:
actually, james wouldn't have to 'break' the rfc. passing the
rcpto (mailfrom while you're at it :o) information to mailets as
they are invoked
it is passed as the Mail.recipients, how do you think we manage to send mail onwards??!
kinda what i thought (but i don't write
One suggestion to James is that they break the RFC rules (I'm not sure
if it's due to some RFC conformity) and break up the [EMAIL PROTECTED] into
two user and domain parts. Once this is done, virtual hosting would
be pretty neat. Mailets can then choose to ignore the domain, or take
them into
not directly. completely different code base.
b
Gary L. Harris wrote:
Is James affected by this?
CERT Advisory CA-2003-07 Remote Buffer Overflow in Sendmail
Gary Harris
wvinternet.com
-
To unsubscribe, e-mail: [EMAIL
doesn't md5 use a salt?
b
Javier Storni wrote:
Hi Vincenzo,
I've exported my Linux passwords to James, setting pwdAlgorithm to MD5.
But doesn't works
If anyone did that (export Linux passwords to James user table (JDBC)), cand
send me some hints ?
Thanks in advance.
Javier Storni
[EMAIL PROTECTED] wrote:
Does Linux use a salt? I don't know. If the answer is yes, then there is no solution to Javier's problem:
which is what i was angling at. it kinda looks like it to me:
php
echo md5(password4me);
/php
yields: 9a74675cc48f209ef0f90d9a2d6f6e7a
# grub-md5-crypt
Password:
as much as i would like to go undercover :o), the problem is that open relays are really a small part of the spam that is sent. true, they represent some of the lower forms of life, but in terms of being an annoyance to end users they are but a fraction of the overall volume. here are some mail
James currently touts a Bayesian mailet, but employs only an
overall data source and is not concerned with individual
preference; to be an effective SPAM blocker, a relationship
needs to be established between a specific user and her [sic]
Bayesian lists.
and of course the necessary functionality
can you expand upon what you consider pattern data?
b
alan.gerhard wrote:
back up a bit -
my point differs in that the pattern data collected is
individual and i do not see too much need for sharing.
other than that, the outstanding issue is, as a james user,
how to go about setting up and
alan.gerhard wrote:
This boils down to a collection of 'good mail' and a
collection of 'bad mail', that in my opinion needs to
reflect the users' interests, therefore I am a bit leery in
'sharing' this data, but am not dismissing it's potential.
collection of 'bad mail'? as in sending out a list
Noel J. Bergman wrote:
I was going to say that same thing to him. In fact, I had written it in my
note, but then I saw his comment about sending the mail to null, so I think
that he knows not to actually be an open relay.
--- Noel
i saw the same. however, how is he going to be an 'open relay'
unless the spammer is only looking at the SMTP codes (not going into *that* discussion again :o) the machine is going to have to actually *deliver* the note. at that point it will be an open relay and will be part of the problem. also, any spammer worth a darn will have a handful of 'feedback'
a good place to start is to post an 'uninteresting' note to a variety of USENET lists using a 'clean' e-mail address. (alt.sex is one i have used in the past, but the more you spread around the more likely you are going to get hits). this gets the real bottom feeders since anyone using that
You don't need to do anything to attract spammers; they just show up. You
don't need to do anything to be probed for being an open relay other than
have an available SMTP port on the internet. Your IP will be probed. I
recently installed a computer on broadband for my uncle. Within 5 minutes
i have witnessed that first hand, but unfortunately it is almost impossible to have any concrete proof. the only possibility that i can think of is trying to unsubscribe using a 3rd [clean!] address that is not yet on their list. if they are legit they should come back and say that it wasn't
yep, which is kinda how the whole rbl thing works (via dns lookups)...
b
[EMAIL PROTECTED] wrote:
If it were possible to create addresses that were known to receive only
spam, then you could set up these servers in a bunch of domains and have
them all update a central database with info on they
?
bill parducci To: James Users List [EMAIL PROTECTED]
[EMAIL PROTECTED] cc: (bcc: Kevin Bedell/Systems/USHO/SunLife
One thing, though. It seems to me as a Java programmer that I could put
together a mailet that contained much more sophisticated analysis than just
a reverse-dns lookup. If I were to write a mailet that could reliably
figure out spam based on more than just the sending host then it seems like
take a look at this:
http://nagoya.apache.org/wiki/apachewiki.cgi?James/SmartOrSecondaryHost
b
Greg Steuck wrote:
Hello,
could someone suggest the best/easiest way to configure James to have it
send most of the mail out using ordinary MX lookups but send mail for a
particular domain (say
i belive there may be a solution to this by adding the concept of
'direction' to the mail flow analysis. this only works if your mail
server is in a protected area where IP spoofing is not possible (you
cannot trust your ISP to check for spoofing, but a well configured
firewall or router does
yes indeed. there are potential holes in the solution (as with all
possible solutions). it really comes down to the level of certainty you
wish to achieve. since e-mail is just about the most unsecure method of
communication in the universe :o) this has been sufficient for most of
the stuff i
yes, but i use the concept of direction for other things when i evaluate
mail against my policies (branched logic: if external, do X; if
internal do Y), so this is a natural extension of my setup.
b
Noel J. Bergman wrote:
1. you define those ip addresses that are considered 'internal'. in my
testing tidbit:
for those who want to keep the console running, but don't want to stay
logged in i would suggest considering screen. not only does this keep
the session alive after you detach but it allows you to fire up numerous
'screens' where one can observe various log files, edit
anyone know what happened to the avalon http proxy stuff? i went to the
dirs to pull down the code and they are empty. from the description it
sounded like there was *something* going on as late as last summer. has
it been abondoned?
thanks
b
doh! didn't even think to look there. thanks!
(wonder why not accessible form website -- or links not removed?)
b
Steve Short wrote:
Not sure, but I found it in CVS at
http://cvs.apache.org/viewcvs.cgi/avalon-apps/
Steve
-
unless someone figures out how to spoof 127.0.0.1 :o)
b
Noel J. Bergman wrote:
Bryan,
Is james configured correctly/securely out of the box?
It should be, yes. We don't know what changes he made in the first place,
nor which one he found that he had to make to correct his configuration.
attached are the first swacks at the graphics. thoughts?
b
Noel J. Bergman wrote:
Bill,
If you are willing to do the graphics, and keep them relatively small, I
don't see any reason why we wouldn't add them to the site. Right now we are
doing the collaborative editing in the Wiki, but
if you are referring to protocol level rejection, the answer is that
james doesn't work that way; it accepts all messages and then processes
them via the mailets (match=RemoteAddrNotInNetwork=[] determines which
host will be allowed to relay).
in the default config relayed mail is dumped in
perhaps the term 'downstream' would be better suited to our purposes
than 'internal'? i was just using this since it is a common installation
configuration.
there are many ways we can expand on the implementational issues if, as
below, you wish to include dns entries. in that situation i would
i'll start with a few simple graphics and we can go from there. i'll try
to get something out tonight.
Noel J. Bergman wrote:
Note about load balancing. James doesn't have to refuse connections to
cause load balancing to occur. By setting up a number of MX records with
the same priority, the
: bill parducci [mailto:[EMAIL PROTECTED]]
Sent: Saturday, January 25, 2003 11:03
To: James Users List
Subject: Re: james and SMART_HOST (DH)
ok, after much doinking around i believe that i have a workable config
that allows a james server to act as a bidirectional MX (receives mail
on behalf
whilst watching the gobs of spam fly by on my james server i noticed some queuing behavior that i found odd. at first i thought that it was just me, but i have seen some posts ('why is this taking so long to deliver?' kinda stuff) that indicates that maybe it is something worth asking about.
that would explain it. sorry, i didn't even notice the thread count in the RemoteDelivery config (the rest of the thread settings seem to be bunched up at the bottom of the config file).
still, might be worth a noob note for those seeing 'long delivery' issues.
thanks
b
Noel J. Bergman wrote:
ok, i'll give it a shot.
b
Noel J. Bergman wrote:
no problemo... now, what is the wiki site? :o)
See my other message. :-)
I've given you a head start on the Smart Host / Secondary MX page by pretty
much copying your message, and doing some minor edits with the Wiki's text
formatting
ok, after much doinking around i believe that i have a workable config
that allows a james server to act as a bidirectional MX (receives mail
on behalf of internal servers and relays mail out for same) without
being a spam ho. i figured that i would post this for reference should
another noob
Serge Knystautas wrote:
This sounds like just a case of configuring two instances of
RemoteDelivery (one that does gateway to the other box) and one that
sends using standard MX record lookups. Then just put the appropriate
matchers in front of them so the right emails use the appropriate
well, this is sorta working so i think that i have an 'order' issue to
resolve. per your suggestion i have the following configuration:
mailet match=HostIs=internal.myfoo.com class=RemoteDelivery
outgoing file://var/mail/outgoing/ /outgoing
delayTime 2160 /delayTime
delayTime 2160
i have been trying to setup james as a mail exchange similar to some existing sendmail servers i have running. in a nutshell, i am looking for the equivalent of sendmail's 'SMART_HOST' capabilities ('forward all incoming mail to host X'). since i don't see anything that would indicate a way to do
great, thanks.
b
Serge Knystautas wrote:
http://james.apache.org/provided_mailets_2_1.html
Check out the RemoteDelivery mailet... you add the optional gateway
parameter.
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
i have been trying to setup james as a mail exchange similar to some
existing sendmail servers i have running. in a nutshell, i am looking
for the equivalent of sendmail's 'SMART_HOST' capabilities ('forward
all incoming mail to host X'). since i don't see anything that would
indicate a way
not to be pessimistic (i can't help myself :o), but the other problem i
think you are going to run into is multiple [James] MXs servicing a
single site. unless they all share a common database the 'state'
information may not be available upon receipt of a bounce.
the problem with a common
Yes, I think it's a well-established practice to use a common database
as a store in a distributed system.
provided that the database can handle the volume. since mail is well
suited to load balancing across many, simple systems it is easy to
envision an implementation whereby you do not have
i have been playing with james (2.1) and am wondering if i have
misconfigured my server because i don't see protocol level reject of
[open] relay requests, but rather it forwards such requests to the spam
folder after accepting them (250). i am concerned about this since most
automated open
None of the major open relay tests do that, in fact. James is far from
alone in this regard. My James server has been checked numerous times by
open relay testers, and passed each time.
--- Noel
interesting. so the behavior described is then normal and i can assume
there isn't a
48 matches
Mail list logo
