July 28, 2008  
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31

[00:03:14] *** maqr has joined #postfix
[00:04:39] *** cipherz has joined #postfix
[00:06:14] *** Juspion has joined #postfix
[00:16:16] *** Draecos has left #postfix
[00:18:10] *** cilly has joined #postfix
[00:20:48] <higuita> penrod: its that postfix still have sendmail, mailq, and any other sendmail commands, any app that is working with sendmail will work without change in postfix
[00:21:29] <higuita> even some internal "interface" that sendmail uses are avaiable in postfix, like support for milters
[00:21:55] <sysmonk> except support for bugs
[00:22:04] <sysmonk> postfix tries not to support those
[00:22:36] <higuita> i have literally shutdown old sendmail, remove the binaries, install postfix, reconfigure and start it
[00:23:02] <higuita> no ones see the difference other that is faster and works better :)
[00:23:09] <higuita> yep, that too sysmonk :)
[00:25:28] *** Juspion has quit IRC
[00:43:12] <penrod> higuita: I am trying to troubleshoot the "vacation" program and I see the following error in the logs. error: to submit mail, use the Postfix sendmail command
[00:58:56] *** madrescher has joined #postfix
[01:06:07] *** Nappy has joined #postfix
[01:13:21] *** Draecos has joined #postfix
[01:19:20] *** hever has quit IRC
[01:35:16] <penrod> higuita: are you still there ?
[01:36:39] *** m1n3s6 has left #postfix
[01:37:21] *** madrescher has quit IRC
[01:37:37] *** cilly has quit IRC
[01:46:21] <higuita> penrod: yes
[01:47:01] <higuita> how is your vacancies trting to send the email? what command
[01:51:34] *** growltiger_ has joined #postfix
[01:52:08] *** toytoy has joined #postfix
[01:52:21] *** Fallenou has quit IRC
[01:55:10] <penrod> higuita:\username, "|/usr/bin/vacation username"
[01:57:07] <higuita> try to see what command it uses, it doing something strange
[01:57:46] <higuita> strace -efile /usr/bin/vacation username < somefile
[01:58:56] <penrod> higuita: it looks like /usr/bin/vacation   starts the process of setting up the vacation program
[02:00:17] <penrod> higuita: oh yeah, there is lots of output there, I will take a few minutes and see if I can decipher it.
[02:01:13] *** war9407 has quit IRC
[02:07:58] *** growltiger has quit IRC
[02:11:57] <penrod> higuita: the output from strace  http://pastebin.ca/1084721
[02:12:22] *** xid has joined #postfix
[02:12:45] <xid> what is the best utility to sync two IMAP accounts?  (one is new/empty, another is full of stuff)
[02:14:12] *** dusty has joined #postfix
[02:14:30] *** dusty has quit IRC
[02:15:56] <sahil> xid: that is not a postfix question.
[02:16:01] <higuita> it seens that you run the vacation as a wrong user :)
[02:16:07] <higuita> vacation: .vacation.db: Permission denied
[02:16:08] <xid> well yeah
[02:16:11] <xid> but do you know?
[02:16:20] <sahil> xid: look into rsync.
[02:16:30] <higuita> xid: there is a imap syn tool
[02:16:41] <higuita> but i cant recall the name...
[02:16:46] <penrod> higuita: I thought that as well, I chmod 777 .vacation.db and still the same error
[02:17:03] <higuita> penrod: wrong user probably
[02:17:04] <sahil> xid: rsync if you have server-side access to both IMAP folders.  otherwise, you might need to google for something.
[02:17:15] <xid> cool guys thanks
[02:17:31] <sahil> np.  don't go getting hax0red, now.
[02:17:40] <xid> heh heh
[02:18:52] <penrod> higuita: I have now run the strace from root, and now it hangs some part way through the process.
[02:19:19] <higuita> xid: have you tried even a simple google search?!  first link of "imap sync" is http://freshmeat.net/projects/imapsync/
[02:19:34] <higuita> penrod: :)
[02:19:34] <xid> looks good
[02:19:50] <xid> never know what's out there that google doesn't link on the 1st page
[02:20:01] <xid> word-of-mouth is the best form of promotion
[02:21:05] <higuita> lets try another thing... have you removed the original sendmail? maybe that error is you tring o use the wrong sendmail command
[02:21:05] <sahil> bollocks.
[02:22:17] <penrod> higuita: I did remove sendmail ages ago. and have symlinks in there , ie. lrwxrwxrwx  1 root root          17 2008-07-20 18:47 sendmail -> /usr/sbin/postfix
[02:23:13] <higuita> xid: being lazy is bad and a waste of time fir others...  5s search gave the result, asking here took more time and waste time for more people... i dont mind to help, but poeple should help thenself first
[02:23:26] <xid> quite the antogaist aren't we?
[02:23:31] <xid> *antagonist
[02:23:52] <higuita> penrod: ?!
[02:23:58] <xid> I /wc
[02:23:59] *** xid has left #postfix
[02:24:31] <higuita> you did install postfix, rigth?
[02:24:46] <penrod> higuita: when a program is looking for sendmail, I have a link to postfix, I was of the unserstanding, that this was correct.
[02:25:10] <penrod> higuita: postfix has been running fine for ages, excepting the vacation program.
[02:25:28] <higuita> a correct install for postfix will put a special sendmail program in /usr/sbin
[02:25:39] <higuita> you would not need to do anything
[02:26:04] <higuita> even worst synlink sendmail to postfix, they are 2 different programs
[02:26:24] <penrod> higuita: ah, perhaps that is the issue. What if I were to upgrade the current version? or should I remove the current version and do a complete install ?
[02:26:29] <higuita> by doing the synlink, you broke the sendmail compatibility :)
[02:26:52] *** growltiger has joined #postfix
[02:27:22] <higuita> if you can, a backup of the /etc/postfix and a resinstall its better, i dont know what else you "broke" :)
[02:27:43] <penrod> higuita: thank you that is what I will do.
[02:27:50] *** cilly has joined #postfix
[02:44:16] *** growltiger_ has quit IRC
[02:45:58] *** ming_zym has joined #postfix
[02:48:46] *** coniptor has joined #postfix
[02:50:00] <coniptor> Hi, My upstream mail provider adds "[Suspected Junk Mail] " to the front of my incoming email which they think *might* be SPAM which half of the time is not.
[02:50:33] *** krusi has joined #postfix
[02:50:37] <coniptor> I was told by someone here to look into header_checks. Problem is anything I find while googling seems to deal with adding to the subject header and not **removing** text from the subject header.
[02:50:48] <krusi> Which mailinglist do you recommend with postfix?
[02:50:59] <coniptor> Is it possible to do a find and replace with header checks and if so can anyone point me to an example?
[02:55:04] *** toytoy has quit IRC
[03:01:32] <krusi> Hello? Noone here?
[03:01:38] *** gonewestcoast has joined #postfix
[03:04:43] *** cyr- has joined #postfix
[03:05:52] <gonewestcoast> Idle question: Given the choice, which authentication backend (MySQL, PostgresSQL, LDAP, flatfile, unix passwd file) tends to be the most efficient?
[03:10:24] *** Tachy_ has joined #postfix
[03:11:06] *** cilly has quit IRC
[03:22:15] *** Tachy has quit IRC
[03:24:39] <rob0> Efficient? That's a strange question. Obviously those which don't require another daemon or process would be more "efficient", no?
[03:25:55] *** growltiger_ has joined #postfix
[03:25:59] <gonewestcoast> Ideally, but there are also efficiencies of scale to consider.
[03:26:15] <gonewestcoast> I don't REALLY want to have the passwd file NFS mounted, for instance. :)
[03:27:05] <gonewestcoast> But let me be a bit more clear.
[03:27:32] <gonewestcoast> Between PostGres, MySQL, and LDAP, are there any obvious "This is the best solution for most circumstances" winners?
[03:28:15] <rob0> If you're already using LDAP for other purposes, LDAP.
[03:29:15] <rob0> I'm somewhat partial to pgsql, but mysql has a much larger group of fanboys.
[03:30:09] <gonewestcoast> I've always mostly stayed away from databases; used MySQL a bit for various projects, but don't know much about pgsql.
[03:31:20] <sahil> fanboys, interesting.
[03:32:14] <sahil> i use flat files.  wietse does too.  use what works for you. :)
[03:32:34] <rob0> I use local (passwd) mostly.
[03:33:01] <rob0> but I'm not running anything on the scale of gmail
[03:33:12] * sahil neither
[03:34:42] <gonewestcoast> The problem is, my environment spans four mx gateways and two endpoint servers, so flatfiles unfortunately aren't scaling as well as I'd hoped they would...
[03:34:43] <sahil> if we're doing the idle question thing, do either of you offer webmail services?  i'm allergic to horde.  was thinking squirrelmail.  there are enough users asking for it so i can't just keep saying 'use a real mail client' anymore.
[03:35:01] <gonewestcoast> Squirrelmail is great, but it's also hideous.
[03:35:02] 
[03:35:08] <gonewestcoast> Roundcube works for some.
[03:35:20] <gonewestcoast> mailman spanks majordomo most days.
[03:35:20] <sahil> gonewestcoast: roundcube seems overkill for what we need.
[03:35:32] <gonewestcoast> sahil: I thought it was relatively lightweight?
[03:35:45] <sahil> gonewestcoast: jay?
[03:35:51] <gonewestcoast> Squirrelmail is what I run personally, but everyone I've showed it to says it looks like 1994 came back to visit. :)
[03:36:04] <gonewestcoast> Yeah, that's me.
[03:36:07] <sahil> gonewestcoast: that's what i like about it; simple UI.
[03:36:14] <coniptor> Is it possible to do a sed like replace on the subject header line of incoming mail with header_checks and if so can anyone point me to a working example?
[03:36:18] <sahil> gonewestcoast: thought so; i recognized the domain name from the mailing list.
[03:36:33] <gonewestcoast>  Yeah, I lurk most days. :)
[03:36:37] <rob0> Squirrelmail, mailman here
[03:36:54] <gonewestcoast> What I like about squirrelmail is the number of plugins for it.
[03:36:57] <gonewestcoast> You can do most things there.
[03:37:02] <sahil> and in terms of scaling, hash dbs are quite fast. :)
[03:37:28] <sahil> as long as squirrelmail can support virtual mailboxes, we'll be fine.  will look into it.
[03:38:02] <gonewestcoast> Right, but it's hard to get hash dbs to scale across multiple nodes.
[03:38:07] <gonewestcoast> It does.
[03:38:15] <gonewestcoast> I run it for sequestered.net, which is a virtual system.
[03:38:21] <gonewestcoast> All it is is an IMAP client written in PHP. :)
[03:38:22] <rob0> Squirrelmail has nothing to do with the mailboxes directly. It is an IMAP client.
[03:38:34] <sahil> even better.
[03:38:40] *** Draecos has left #postfix
[03:38:46] <gonewestcoast> I don't think I'd trust a webmail client that didn't work that way.
[03:39:08] <sahil> gonewestcoast: *nod*
[03:42:55] *** growltiger has quit IRC
[03:45:19] *** krusi has quit IRC
[04:00:37] *** maxquerry has quit IRC
[04:11:58] *** toytoy has joined #postfix
[04:35:39] *** seekwill has quit IRC
[04:39:35] *** AcTiVaTe has quit IRC
[04:46:51] *** pickcoder has joined #postfix
[05:01:57] *** AcTiVaTe has joined #postfix
[05:12:54] *** internat1 has quit IRC
[05:13:19] *** internat has joined #postfix
[05:35:11] *** tombar_ has joined #postfix
[05:48:53] *** c0m has quit IRC
[05:52:29] *** tombar_ has quit IRC
[05:54:20] *** c0m has joined #postfix
[05:56:43] *** tombar has quit IRC
[06:02:43] *** seekwill has joined #postfix
[06:11:44] *** saurabhb has joined #postfix
[06:20:54] *** cyr- has quit IRC
[06:42:52] *** AllenJB_ has joined #postfix
[06:43:19] *** AllenJB has quit IRC
[06:55:48] *** pickcoder has quit IRC
[06:59:06] *** f3ew has joined #postfix
[07:05:54] *** xpoint has quit IRC
[07:14:58] <roe_> can I put "email at localdomain dot com email at foreigndomain dot com" in my transport file?
[07:17:04] <f3ew> no
[07:17:19] <f3ew> that's virtual_alias_maps or alias_maps
[07:22:27] <roe_> can I put two email addresses on the right side of that hash?
[07:22:32] <roe_> comma seperated?
[07:24:01] <roe_> wow, I love postfix
[07:24:07] <f3ew> yes
[07:24:09] <f3ew> !virtual
[07:24:12] <knoba> f3ew: "virtual" : a way to configure additional domains and user accounts (that do not need to exist in your /etc/passwd). See: http://www.postfix.org/VIRTUAL_README.html
[07:24:16] <f3ew> Also see man 5 virtual
[07:46:29] *** roe___ has joined #postfix
[07:46:40] *** roe___ has quit IRC
[07:46:52] *** seekwill has quit IRC
[07:53:53] *** romero has joined #postfix
[07:57:27] *** ki__ has joined #postfix
[07:58:30] <ki__> phpmailer is configured with send
[07:58:55] <ki__> err "from"
[07:59:08] <ki__> but the mail logs say it is from daemon at localhost dot localhost
[08:08:58] *** bisoc has quit IRC
[08:12:07] *** Fallenou has joined #postfix
[08:18:44] <stony> morn
[08:21:54] *** hever has joined #postfix
[08:26:44] *** Fallenou has quit IRC
[08:30:14] *** Filbert has joined #postfix
[08:31:27] *** sophokles has joined #postfix
[08:32:49] *** martiancode has joined #postfix
[08:33:04] *** phnord has joined #postfix
[08:37:38] *** Fallenou has joined #postfix
[08:38:14] *** Swat2 has joined #postfix
[08:38:49] <Swat2> Is there any way to Forward Emails for DomainA.com through to another server _AFTER_ it has been processed by SpamAssassin&ClamAV?
[08:39:31] <Swat2> im currently using Transports:  domaina.com smtp:mail.domainasmarthost.com
[08:39:47] <Swat2> but it doesnt look like spamassassin or whatever is processing the email
[08:57:12] <Swat2> Anyone?
[09:01:53] <sysmonk> Swat2: your spamassassin/clamav is run by what?
[09:01:59] <sysmonk> a pipe?
[09:02:05] <sysmonk> i.e. a .forward file ?
[09:04:20] <Swat2> umm
[09:05:34] <Swat2> i think it uses procmail aswell
[09:05:38] <Swat2> im using Virtualmin
[09:05:51] <Swat2> mailbox_command = /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME
[09:06:02] <sysmonk> then it won't be scanned
[09:06:12] <Swat2> ok
[09:06:14] <sysmonk> as it is scanned on delivery to the mailbox ( i.e. when using local(8) or virtual(8))
[09:06:21] <Swat2> ahh
[09:06:30] <sysmonk> and your mail doesn't get that far
[09:06:37] <Swat2> so, is there another way to do it?
[09:06:45] <sysmonk> before_queue / after_queue scaning
[09:07:09] <Swat2> after
[09:07:40] <Swat2> perhaps i need to make all the emails go into one account and pop3 it out?
[09:08:04] *** coniptor has left #postfix
[09:09:26] <sysmonk> Swat2: that was an answer
[09:09:55] <sysmonk> !FILTER_README
[09:09:56] <knoba> sysmonk: "FILTER_README" : http://www.postfix.org/FILTER_README.html :: After-queue content filtering
[09:10:00] <sysmonk> Swat2: ^^
[09:10:07] *** kRocKodile has joined #postfix
[09:11:38] <Swat2> sysmonk:sorry, bit slow today :(
[09:11:51] <Swat2> *reading doco*
[09:12:42] <Swat2> Can you scan the Email twice?
[09:15:47] *** loadkast has joined #postfix
[09:18:17] *** master_o1_master has joined #postfix
[09:22:34] <loadkast> need help setting up postfix
[09:23:00] *** cedric3 has quit IRC
[09:29:47] *** master_of_master has quit IRC
[09:30:17] *** simsimon has joined #postfix
[09:30:18] *** bboschman has joined #postfix
[09:30:26] <simsimon> hello !
[09:30:26] <bboschman> Hi
[09:32:33] <bboschman> I got some messages stuck in mailq. I had some misconfiguration on friday and fixed this by adding a mail-route in tranport file (which is working), but I still got 12 old mails in mailq that want to go the old (not working) way. I already tried to requeue all but that didn't helped. Any idea how to get these mails on their way?
[09:37:34] *** hever has quit IRC
[09:38:32] <simsimon> I'm using Virtual Accounts with postfix and I have those two lines for messages storage : http://pastebin.com/d14ed79bd but with this configuration, all the mails are in a single file. How to store them in maildir format ?
[09:38:38] *** hever has joined #postfix
[09:43:52] *** martiancode has quit IRC
[09:47:05] *** AllenJB_ is now known as AllenJB
[09:49:31] <gonewestcoast> bboschman: how did you requeue them?
[09:49:54] <gonewestcoast> loadkast: What're you trying to do specifically?
[09:51:06] <gonewestcoast> simsimon: I'm running something similar.  What's virtual_transport set to in postconf -n?
[09:51:17] *** keffff has joined #postfix
[09:51:48] <simsimon> gonewestcoast: virtual_transport = virtual
[09:52:36] <gonewestcoast> Yeah, I hand off to maildrop at that stage.
[09:52:40] <gonewestcoast> (built into Courier)
[09:52:46] <gonewestcoast> It handles delivery to various boxes.
[09:52:53] <gonewestcoast> What's your imap server?
[09:54:12] <bboschman> gonewestcoast, yes I did a postsuper -r ALL
[09:54:14] <simsimon> gonewestcoast: I've no imap server yet which one do you suggest for me ?
[09:54:47] *** jes-o-mat has left #postfix
[09:55:13] <gonewestcoast> simsimon: Depends on what you're doing.  I'm a fan of both Courier and Dovecot.
[09:55:29] <gonewestcoast> But it helps to have that configured before tackling the maildir storage problem. :)
[09:55:46] <gonewestcoast> bboschman: That SHOULD work.  Are you seeing messages getting new message IDs when you do it?
[09:56:01] <gonewestcoast> I'd have interest in seeing the mail log of what happens after the requeue; it sounds like your problem still manifests for those messages.
[09:56:02] <bboschman> just a sec
[09:56:03] *** war9407 has joined #postfix
[09:56:20] <simsimon> ok so I configure my imap server fist ;)
[09:56:28] *** keffer has quit IRC
[09:58:08] <gonewestcoast> mailbox_transport = virtual <-- This is what you'll want to set as well, simsimon
[09:58:44] <bboschman> gonewestcoast, they get new msg-IDs
[09:58:53] <bboschman> http://prod01.nfon.net/~nfon/
[09:59:11] <simsimon> gonewestcoast: ok thanks :)
[10:00:12] <bboschman> I also copied 'postconf -n' and transport file there
[10:00:39] <gonewestcoast> bboschman: While you're tossing things up there, why not include the last few dozen lines of the maillog? :)
[10:00:44] <simsimon>  what are the differences between the imap servers ? (Courier, Dovecot, Cyrus...)
[10:00:47] <gonewestcoast> Postfix is great about telling you what broke.
[10:01:11] <gonewestcoast> Courier's more robust, speaks MailDir.  Dovecot speaks both Maildir and mbox format, great for transition.  It's newer, though.
[10:01:23] <gonewestcoast> Cyrus is a pain all around, if you want my (admittedly prejudiced) opinion. :D
[10:02:26] <gonewestcoast> http://pastebin.com/m29da36f4 <-- has my config, realize I'm running a MySQL backend as well as courier / maildrop.
[10:04:44] <simsimon> thanks gonewestcoast ! Could you also pastebin your mysql_virtual_mailbox_maps.cf please ?
[10:04:47] <gonewestcoast> Was a funky couple of days getting it all set up.
[10:04:47] <gonewestcoast> Sure.
[10:05:28] <gonewestcoast> Other than the username and password stuff that say which DB it's in?
[10:05:29] <gonewestcoast> select_field = maildir
[10:05:29] <gonewestcoast> where_field = username
[10:05:30] <gonewestcoast> That's it.
[10:06:13] <simsimon> thanks :D
[10:06:59] <simsimon> I think the problem is I've no maildir attribute in my LDAP db...
[10:07:25] <simsimon> that's why I'm using the UID
[10:10:17] <bboschman> gonewestcoast, http://prod01.nfon.net/~nfon/mail.log.txt
[10:11:00] *** bisoc has joined #postfix
[10:11:14] <gonewestcoast> Yeah, without that, simsimon you're going to be just dumping to a file. :)
[10:11:18] <gonewestcoast> So you may wish to define one.
[10:11:39] <bboschman> Jul 28 09:57:09 prod01 postfix/postsuper[14367]: Requeued: 13 messages
[10:11:39] <bboschman>  <- this is where it starts
[10:12:12] <bboschman> but postfix still tries to deliver those mails to mail.nfon.net[88.217.254.5]
[10:12:28] <gonewestcoast> Yeah, which doesn't accept the connection.
[10:12:35] <bboschman> hm
[10:12:39] <bboschman> it different :)
[10:12:42] <bboschman> its
[10:13:07] <gonewestcoast> Oh!
[10:13:15] <bboschman> from the location where postfix is running, mail.nfon.net is only reachable through 192.168.3.5
[10:13:27] <bboschman> therefor I added the tranport rule
[10:13:34] <gonewestcoast> Not sure if this matters (I'm not THAT into syntax), but USUALLY for transport mappings you put the IP in square brackets.
[10:13:39] <bboschman> as well as an /etc/hosts entry
[10:13:58] <gonewestcoast> And just to be fun...
[10:14:04] <gonewestcoast> you DID postmap /etc/postfix/transport, right? :)
[10:14:32] <bboschman> postmap -p ...
[10:15:06] <bboschman> I can try using the ip address, but since I changed the tranport map all mails are delivered correctly
[10:15:13] <gonewestcoast> No no...
[10:15:17] <bboschman> only these old messeages
[10:15:22] <bboschman> are stuck :/
[10:15:28] <gonewestcoast> postmap /etc/postfix/transport <-- the entire command
[10:15:32] <bboschman> no no?
[10:15:38] <gonewestcoast> And just to verify that it's working...
[10:15:50] <gonewestcoast> postmap -q user at nfon dot net /etc/postfix/transport
[10:15:52] <gonewestcoast> That SHOULD return the IP.
[10:15:58] <bboschman> any postfix reload/restart?
[10:16:09] <gonewestcoast> Not unless you haven't since you added the transport map.
[10:16:49] <bboschman> postmap -q test at nfon dot net /etc/postfix/transport <- empty result
[10:16:53] <bboschman> oO
[10:17:38] <bboschman> very weird
[10:18:10] <gonewestcoast> Okay, put it in brackets, postmap the file again, then run the query.
[10:18:13] <gonewestcoast> Any change?
[10:19:16] <bboschman> the whole [smtp:ip] or just [ip]?
[10:19:25] <gonewestcoast> The latter.
[10:21:12] <bboschman> http://prod01.nfon.net/~nfon/transport.txt
[10:21:51] <bboschman> after postmap /etc/postfix/transport; postmap -q user at nfon dot net /etc/postfix/transport still does not print any output
[10:22:49] <gonewestcoast> Oh, butter my butt and call me a biscuit.
[10:22:53] <gonewestcoast> drop the user@ from the query. :)
[10:23:09] <bboschman> hrhr
[10:23:18] <bboschman> results in: :[192.168.3.5]
[10:23:59] <gonewestcoast> No smtp first?
[10:26:46] <bboschman> from the manpage it should not be mandatory: When no transport is specified, Postfix uses the transport
[10:26:47] <bboschman>  that matches the address  domain  class.
[10:26:55] <gonewestcoast> Yeah...
[10:27:11] <gonewestcoast> Yet if you send a NEW message to * at nfon dot net, it routes correctly?
[10:27:55] <bboschman> wait - I'll do
[10:28:52] *** nmh_2Grajw has joined #postfix
[10:29:57] *** cilly has joined #postfix
[10:30:13] <bboschman> http://paste.debian.net/13205/
[10:30:55] <bboschman> this is what drives me mad =)
[10:31:44] <gonewestcoast> That's odd as anything.
[10:31:55] <gonewestcoast> I've done this exact change before; queued mail rerouted accordingly upon requeue.
[10:37:40] <bboschman> hm
[10:37:46] <bboschman> I'll kick them :/
[10:38:51] <gonewestcoast> Stop and restart the Postfix service, then requeue?
[10:39:54] <bboschman> already done several times
[10:40:03] <gonewestcoast> You sure the service is stopping?
[10:40:06] <gonewestcoast> confirmed with ps -ef?
[10:41:30] *** madrescher has joined #postfix
[10:42:17] *** mark-use has joined #postfix
[10:43:40] <bboschman> gonewestcoast, 2 late :)
[10:43:48] <bboschman> thanks anyhow for your great support
[10:44:13] <gonewestcoast> Sorry, wish I coulda done more.
[10:47:29] *** lysander_ has joined #postfix
[10:53:25] <kRocKodile> when i send ONE mail with 3 recipients to @mycompany.gr they arrive 3 mails (identical. with 3 recipients each) in the mailserver on my ISP. is that normal?
[10:55:47] <sysmonk> kRocKodile: yes
[10:56:17] <kRocKodile> this is a problem then...
[10:56:47] <kRocKodile> because fetchmail gets 3 mails instead of 1
[10:56:57] <kRocKodile> and deliveres 9 mails total!
[10:57:33] <kRocKodile> (fetchmail uses a catchall account)
[11:02:29] *** lysander has quit IRC
[11:04:17] <kRocKodile> sysmonk: its still normal when these 3 recipients is from the same domain?
[11:04:40] <sysmonk> kRocKodile: yes
[11:04:49] <sysmonk> kRocKodile: smtp doesn't have a per-recipient DSN's
[11:05:03] <sysmonk> so it sends 3 different mails
[11:05:46] <kRocKodile> you know any 'keyword' to search for this fetchmail problem? (it seems its fetchmail problem then..)
[11:07:14] <sysmonk> i didn't read your problem description, sorry, don't have time now
[11:07:34] <kRocKodile> k
[11:09:02] *** Trengo has joined #postfix
[11:16:19] *** lysander_ is now known as lysander
[11:16:47] *** lysander has joined #postfix
[11:22:39] *** qiyong has left #postfix
[11:27:09] *** cilly has quit IRC
[11:28:04] *** bboschman has quit IRC
[11:28:23] *** diveli has quit IRC
[11:33:50] *** diveli has joined #postfix
[11:54:58] *** diveli has quit IRC
[11:55:21] *** diveli has joined #postfix
[11:56:35] *** JoAkKiNeN has quit IRC
[11:56:36] *** nmh_2Grajw has quit IRC
[12:01:01] *** xeer has joined #postfix
[12:02:25] <xeer> using imap and maildir, i am not getting any new files (mail entries) when an email is sent. where can i find out where my mail is going??
[12:05:08] <xeer> is there a log i can view of what is going on with postfix?
[12:05:29] *** nmh_2Grajw has joined #postfix
[12:06:22] *** Zeit|awy has joined #postfix
[12:07:15] *** diveli_ has joined #postfix
[12:09:02] <rikkyc> Hi - I'm really confusing myself here. I'm trying to use postfix and mysql; my mysql is not using tcp, just sockets. (unix:/pth/to/mysql.sql) When I run with postmap -q to test, everything works fine. When I'm leaving the server alone, it can't connect to mysql "can't connect .. (2)"
[12:09:12] <rikkyc> Any help is appreciated!
[12:09:21] <jelly> xeer: somewhere near /var/log/mail.log or maillog
[12:10:16] *** diveli has quit IRC
[12:12:07] *** Zeit|awy_ has quit IRC
[12:12:15] *** diveli has joined #postfix
[12:14:57] <f3ew> !chroot
[12:14:57] <knoba> f3ew: "chroot" : The fifth column in master.cf, if not n , means that the Postfix process described on that line runs in a chroot, see !debug and !queue_directory .
[12:15:00] <f3ew> @ rikkyc
[12:17:15] <rikkyc> Chroot is off; I've tried with and withour chroot. I have also tried moving the socket path "just in case", and checked permissions. Even gave the postfix user a shell /bin/sh instead of /bin/false and tried to connect to mysql as that user - all seems to work fine
[12:17:20] <rikkyc> !queue_directory
[12:17:22] <knoba> rikkyc: "queue_directory" : a configuration parameter in the main.cf: The location of the Postfix top-level queue directory. This is the root directory of Postfix daemon processes that run chrooted.
[12:17:38] <rikkyc> yup, tried that :(
[12:18:20] <rikkyc> f3ew: thank you though :)
[12:19:09] <f3ew> what's in the logs?
[12:19:17] <f3ew> Also show postconf -n and master.cf
[12:21:15] <rikkyc> postconf -n :  http://pastie.org/242381
[12:21:56] <rikkyc> master.cf : http://pastie.org/242382
[12:22:24] *** ming_zym has quit IRC
[12:22:28] <f3ew> logs?
[12:23:01] <rikkyc> just working on that one, the most relevent line is
[12:23:01] <rikkyc> warning: connect to mysql server localhost: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)
[12:23:35] <rikkyc> which is the wrong mysql socket (I was in the middle of testing another config) but that's the same error I get for the correct socket path
[12:26:44] *** diveli_ has quit IRC
[12:31:27] <rikkyc> f3ew: Anything glaringly obvious that's wrong here?
[12:34:38] * f3ew wuold turn off all chroots
[12:34:41] <f3ew> would
[12:34:49] <rikkyc> ok
[12:39:14] <rikkyc> f3ew: well, i think that worked
[12:40:11] <rikkyc> f3ew: seems to have worked, you're a *; thank you!
[12:43:41] *** Fallen[oqp] has joined #postfix
[12:45:41] *** tuxick has joined #postfix
[12:45:45] <f3ew> rikkyc FWIW, it would hav helped if you had posted the full line
[12:46:11] <tuxick> postmap -q foo at bar dot com ldap:ldap-users.cf giving me:
[12:46:14] <tuxick> postmap: warning: dict_ldap_lookup: ldap-users.cf: Search base '' not found: 32: No such object
[12:46:25] <f3ew> tuxick /path/to/ldap-users.cf
[12:46:29] <f3ew> use tghe full path
[12:46:29] <tuxick> why is it making that up? i defined search_base
[12:46:40] <tuxick> f3ew: that's a silly message then :)
[12:47:11] <tuxick> but you're right, thanks
[12:47:21] <sysmonk> nice, today i've met the first person from freenode :P
[12:47:34] * tuxick hates poor exception handling
[12:48:11] <sysmonk> f3ew: we can make a rule - post the full config, or die ;)
[12:48:20] <sysmonk> or die trying! ;)
[12:51:31] <f3ew> tuxick, that's actually legacy support
[12:51:35] *** ttf has joined #postfix
[12:51:45] <f3ew> Originally you would have to define a ldap config within main.cf
[12:52:01] <f3ew> so you would have ldap-users_search_base in main.cf
[12:52:08] <ttf> hi.. what does the "X-Flags" header stand for?
[12:52:11] <f3ew> sysmonk who?
[12:52:44] <sysmonk> f3ew: some guy from #poland
[12:52:47] <sysmonk> nick p4
[12:52:53] <ttf> or - which is the appropriate rfc I can read up on that?
[12:53:41] <f3ew> ttf X- is a custom header
[12:54:51] <ttf> f3ew: hmm - right. Any idea which piece of software sets: "X-Flags: 0000" ? I'm using kmail+dovecot+postfix and have been sending mail through a gmx mailserver to a gmx address
[12:55:08] <f3ew> Probably something at gmx then
[12:55:54] <ttf> ah - fdm and maildrop have also been involved into that mail
[12:57:59] <ttf> f3ew: yup - from googling it looks like it's a gmx added header
[12:58:46] <tuxick> how wise is it to try using milters for clamav and spamassassin instead of amavis?
[12:59:13] <tuxick> since clamav-milter even refuses to start if it doesn't find sendmail.cf configured for it
[12:59:32] <tuxick> hardcoded check :)
[13:04:29] *** mark-use has quit IRC
[13:10:14] *** Fallen[oqp] has quit IRC
[13:13:06] *** mark-use has joined #postfix
[13:13:55] *** cpm has joined #postfix
[13:21:22] *** swemark has joined #postfix
[13:23:28] <swemark> Hi, im looking for a guide to the commonly required headers in email-markup. Anyone have any idea?
[13:23:40] *** cilly has joined #postfix
[13:29:52] *** Zblakany has joined #postfix
[13:33:34] <xeer> i seem to be receiving mail fine as mbox but maildir "doesn't do anything". how can i debug this?
[13:36:32] <f3ew> swemark RFC 2822
[13:41:56] <xeer> my maildir fire structure isn't being created as well, the new/cur/tmp folders and etc..
[13:42:01] <xeer> fire = file
[13:42:33] <xeer> i'm not getting an error message, postfix seems to be accepting the email, but not putting it anywhere!?
[13:42:51] *** McJerry has quit IRC
[13:43:26] <f3ew> logs?
[13:44:49] <higuita> xeer: what the logs say about the local delivery?
[13:45:13] <higuita> what is the value of home_mailbox (postconf home_mailbox)
[13:46:00] <xeer> I wish I had logs! I see no debug logs..
[13:46:09] <xeer> there is no /var/log/maillog
[13:46:13] <higuita> does that dir (from home_mailbox) exists? are the permissions right? (user owned)
[13:46:31] <higuita> xeer: configure your syslog to log email data to that file
[13:46:42] *** simsimon has quit IRC
[13:46:43] <xeer> I'm using a virtual mailbox map
[13:46:52] <xeer> my syslog is configured to log mail to that file..
[13:47:32] *** dusty has joined #postfix
[13:47:55] <xeer> I can get mail fine in /var/spool
[13:48:15] <xeer> but one file for multiple domains/users is retarded.
[13:48:37] <xeer> im using imap as well
[13:48:46] <xeer> so i need folders
[13:50:17] *** dusty has quit IRC
[13:50:43] <higuita> xeer: if you dont have logs from postfix, your syslog is wrong or you forgot to create the log file first, or to restart syslog
[13:51:27] <higuita> postfix logs to syslog when it starts, receive email, etc
[13:51:34] <tuxick> trying spamass-milter i see it "spamd: clean message (2.5/5.0) for root:58 in 1.9 seconds"
[13:51:45] <tuxick> but the mail isn't for root, any doc explaining?
[13:52:27] <higuita> probably the root:58 is the process 58, owned by root
[13:52:50] <higuita> but i dont know spamass-milter
[13:53:38] <cpm> spamd running as root? hrmmm.
[13:53:48] <tuxick> i'm not even sure how reliable postfix milter support is at all
[13:54:06] <tuxick> cpm: /usr/local/bin/spamd -q -x -u spamd -d -r /var/run/spamd/spamd.pid
[13:54:21] <higuita> i dont used it yet, but AFAIK, its preaty reliable and stable
[13:57:30] *** hark has quit IRC
[13:57:36] *** hark has joined #postfix
[13:57:49] <cpm> how reliable postfix milter support is at all? I can't parse that.
[13:57:59] <cpm> what does it mean?
[14:02:30] <tuxick> most milters are written for sendmail
[14:03:08] <tuxick> anyway, spamd -q is supposed to query mysql for userprefs
[14:03:15] <tuxick> i see nothing like that
[14:04:42] <cpm> http://www.postfix.org/MILTER_README.html
[14:14:12] <tuxick> i know, that's a doc!
[14:14:44] <cpm> explains how postfix makes use of milters, and exceptions.
[14:15:05] <cpm> One of the reasons I moved from sendmail to postfix, was the joy of walking away from milters.
[14:15:06] <cpm> :)
[14:16:52] <jelly> otoh, being able to do complex filtering at smtp time cuts backscatter right away
[14:17:11] <tuxick> cpm: then how do you do the filtering?
[14:17:25] <cpm> tuxick, I use content filter
[14:18:01] <cpm> and amavis
[14:18:03] *** bondoer has quit IRC
[14:18:06] * robtone_ uses filters at SMTP stage, ad headers stage and finaly afterqueue content filters
[14:18:17] <robtone_> s/ad/at/
[14:18:29] * cpm ads robtone_
[14:18:51] * robtone_ buys the extra pack blue pills
[14:19:02] <cpm> good move
[14:20:13] <robtone_> oh, some body checks to hit the backscatter, yes
[14:23:08] <tuxick> looks like spamass-milter just doesn't know about per-user config
[14:23:50] <tuxick> i could still use it for crude prefiltering
[14:25:22] <pingouin> higuita: thanks for your help last night (it was night for me)
[14:25:41] <pingouin> i strace my mail but, the same error is said by strace
[14:26:12] <pingouin> postdrop: warning: uid=1000: Illegal seek     sendmail: fatal: lamatriste(1000): queue file write error
[14:27:44] <xeer> I just reinstalled postfix for the second time, this program is not user friendly for me.
[14:27:57] <xeer> with default settings, what exactly is needed for virtual support?
[14:28:09] *** cilly has quit IRC
[14:28:12] <xeer> I don't want a system account for each user, and I host multiple domains.
[14:28:31] <xeer> I don't want the mail to be stored in the spool, but as Maildir folders.
[14:28:44] <xeer> I'm sure this is all possible..??
[14:31:53] <lunaphyte_> xeer: certainly.
[14:32:02] <lunaphyte_> !virtual
[14:32:03] <knoba> lunaphyte_: "virtual" : a way to configure additional domains and user accounts (that do not need to exist in your /etc/passwd). See: http://www.postfix.org/VIRTUAL_README.html
[14:32:45] <xeer> yes, I've been sitting at this page for hours. reading it over and over again.
[14:33:36] <sysmonk> lunaphyte_: are you  a native english speaker?
[14:33:40] <xeer> my mail doesn't get returned to me, though I can never find it in the places I specified in my main.cf
[14:34:57] <lunaphyte_> sysmonk: as far as i know.
[14:35:19] <sysmonk> lunaphyte_: just wanted to hear an opinion from native english speaker abuot the new search engined name - cuil
[14:35:21] <higuita> xeer: have you already the logs working?
[14:35:29] *** dusty has joined #postfix
[14:35:37] <sysmonk> do you like it? is it easy to remember? is it 'easy' to pronounce?
[14:35:39] <lunaphyte_> sysmonk: lame.
[14:35:48] <lunaphyte_> but i'm a cycnic.
[14:36:24] <xeer> I've tried restarting syslog, but it still does not log anything
[14:36:25] <higuita> postfix is alot more friendly than sendmail, you have all you need in the postfix docs and many howto in the net, like everything, you should read a little before trying to configure something
[14:36:46] <lunaphyte_> sysmonk: but, to answer your questions, in order - no, yes, yes.
[14:37:03] *** xpoint has joined #postfix
[14:37:04] <lunaphyte_> !logs
[14:37:05] <knoba> lunaphyte_: "logs" : by default, postfix logs to the mail facility of syslog. Something like grep -i `postconf -h syslog_facility` /etc/syslog.conf or grep -rl `postconf -h syslog_name` /var/log/* should tell you where logs are going.
[14:37:16] <xeer> that advice would be helpful if I had only spent a little time configuring this program. I've sent literally 6+ hours reading and trying to configure this beast.
[14:37:21] <higuita> fix the logs first, postfix check must log this when starting
[14:37:22] <higuita> Jul 28 13:37:04 Couracado postfix/master[4784]: daemon started -- version 2.5.1, configuration /etc/postfix
[14:37:29] <sysmonk> lunaphyte_: but it's not as attractive as yahoo or google (as in names)
[14:37:46] <sysmonk> attractive - easy to remember, easy to pronounce and etc
[14:37:48] <higuita> if that doesnt work, you will lose half of the tools to do any debug
[14:38:20] <lunaphyte_> sysmonk: it's an obvious attempt at being "clever".
[14:38:30] *** fort_sev has joined #postfix
[14:39:01] <nmh_2Grajw> sysmonk: to butt in on this, I do not think it is "easy" to pronounce - I had to think a minute to (I think) get it.  Of course, when I first saw google and yahoo, I thought they were bad names.
[14:39:04] *** cilly has joined #postfix
[14:39:12] <lunaphyte_> sysmonk: i associate wordplay with companies/organizations that are a flash in the pan.
[14:39:27] <higuita> pingouin: do a strace -fefile /usr/sbin/sendmail our@email < /etc/hosts
[14:39:39] <higuita> pastbin the result
[14:40:09] <sysmonk> nmh_2Grajw: dunno, i liked the names of yahoo and google
[14:40:13] <sysmonk> maybe because of the oo's :)
[14:40:21] <xeer> I've even directed the mail.* logs to the console, still nothing..
[14:40:27] <lunaphyte_> sysmonk: it's like the internet very of "wacky morning zoo" radio.
[14:40:43] <lunaphyte_> err, *the internet equivalent of ...
[14:40:57] <sysmonk> heh
[14:41:42] <lunaphyte_> now that i may have sufficiently offended you - what's your stake in cuil?
[14:42:05] <sysmonk> none
[14:42:10] <lunaphyte_> :)
[14:42:26] <sysmonk> we just had a discussion about the name with some guys on our local (non-english) network
[14:42:43] <sysmonk> and some said that the name is fine, but it's hard to tell that when we're not native english-speakers
[14:43:30] <xeer> what permissions should my maillog be?
[14:43:56] <xeer> it's not mentioned in the readme
[14:48:28] *** jellis-real has joined #postfix
[14:52:46] *** dragonheart has joined #postfix
[14:54:58] *** the_fafa has quit IRC
[14:58:56] *** rapha has joined #postfix
[14:59:02] <rapha> Hi!
[14:59:06] <rapha> Does Postfix depend on "libc-client"?
[15:00:11] <thumbs> rapha: perhaps your distro made it that way
[15:00:30] <rapha> thumbs: but not normally?
[15:00:49] <thumbs> most sane distros don't have dependencies
[15:01:44] <rapha> I'm not worried about the distro, I compiled from source anyway.
[15:01:59] <thumbs> rapha: did it compile fine?
[15:02:10] <rapha> It's been a while...
[15:02:13] <rapha> But yes it did
[15:02:27] <rapha> Distro update wants to install a "libc-client without support for Maildirs" now
[15:02:42] <rapha> So I wanted to know if postfix could have made itself dependant on that lib
[15:02:56] <thumbs> well can you revert back to the old packages if your distro installs that?
[15:03:05] <thumbs> seems like a distro-made name
[15:03:26] <rapha> Yeah, prolly can do
[15:03:35] <rapha> But I installed it now - everything seems to work fine still :-)
[15:03:36] <rapha> so thanks
[15:05:51] <rikkyc> I know this is dumb, but I've been up way too long for someone to safely configure a server. My debian box isn't allowing connections to postfix, no firewall rules up that I can see. There is no entry for smtp in init.conf, but not sure if that's relevant
[15:06:55] <rikkyc> the port is open, but it's taking a long time to get a response externally. Which implies to me somethings amis with the DNs
[15:07:32] *** nmh_2Grajw has quit IRC
[15:08:32] <shasta> libc-client is imap-related thing
[15:13:59] *** Odd_Bloke has quit IRC
[15:17:04] *** Zblakany has quit IRC
[15:18:30] *** Aurica has joined #postfix
[15:23:35] *** rapha_ has joined #postfix
[15:24:47] *** Silowyi has quit IRC
[15:26:19] *** Silowyi has joined #postfix
[15:40:07] *** mark-use has quit IRC
[15:41:22] *** rapha has quit IRC
[15:41:52] *** Odd_Bloke has joined #postfix
[15:42:17] *** Aurica has quit IRC
[15:45:36] <lunaphyte_> rikkyc: what is init.conf?
[15:46:32] *** Silowyi has quit IRC
[15:46:33] <lunaphyte_> rikkyc: does it eventually connect?
[15:47:26] <rikkyc> lunaphyte_: I meant inetd.conf. My actual problem at the moment is getting the server to reject unknown recipients using the MySQL module. For some reason, whatever at example dot com goes through, and dovecot even creates the maildir, even though the user doesn't exist int he MySQL database
[15:47:46] <rikkyc> lunaphyte_: no, I cannot connect. I'm just testing at the moment to make sure my ISP doesn't block the port.
[15:48:19] <lunaphyte_> !reject_unlisted_recipient
[15:48:20] <knoba> lunaphyte_: Error: "reject_unlisted_recipient" is not a valid command.
[15:48:25] <lunaphyte_> bah
[15:49:08] <lunaphyte_> well, use reject_unlisted_recipient in your restrictions
[15:49:47] <rikkyc> lunaphyte_: Okay, my ISP was blocking port 25; which is a relief. I do have rej.. hmm. no, I have reject_unknown. Let's try this...
[15:50:28] <lunaphyte_> that seems backwards.  you're relieved to find out your isp is blocking port 25?
[15:51:12] *** Silowyi has joined #postfix
[15:52:12] <rikkyc> lunaphyte_: Only as far as it means my server is still getting mail :) I added the reject_unlisted, but it still allows the recipient
[15:52:58] <rikkyc> lunaphyte_: postfix reload. I promise, I'm going to bed after this one.
[15:53:05] <lunaphyte_> your isp is blocking port 25, but somehow your server is getting mail?
[15:55:41] <rikkyc> lunaphyte_: sorry, no - my isp is blocking port 25, which is why I wasn't able to test connecting to my server. My server is a virtual server located elsewhere.
[15:56:05] <lunaphyte_> oh, i see.
[15:56:58] <xpoint> !warn_if_reject
[15:56:59] <knoba> xpoint: "warn_if_reject" : a parameter that means: Change the meaning of the next restriction, so that it logs a warning instead of rejecting a request (look for logfile records that contain "reject_warning"). This is useful for testing new restrictions in a "live" environment without risking unnecessary loss of mail.
[15:57:40] <rikkyc> very good option to know!
[15:58:25] *** jordan_cbr has joined #postfix
[15:58:27] <jordan_cbr> hi
[15:58:38] *** swemark has quit IRC
[15:58:59] <jordan_cbr> can someone help me configuring postfix with gosa and smba?
[15:59:02] <jordan_cbr> *samba
[16:01:42] <cpm> ?
[16:01:56] <cpm> what has samba to do with postfix?
[16:01:58] * cpm boggles
[16:02:25] *** Draecos has joined #postfix
[16:03:30] <sysmonk> cpm: | netsend ...
[16:03:31] <sysmonk> ;)))
[16:04:30] <cpm> hmm, intriguing.
[16:05:42] *** kRocKodile has quit IRC
[16:11:53] *** jordan_cbr has quit IRC
[16:14:35] *** alys has joined #postfix
[16:15:01] *** seekwill has joined #postfix
[16:16:07] *** alys has quit IRC
[16:23:59] *** romero has left #postfix
[16:24:14] <xeer> what is this all about?? postfix/smtpd[4893]: fatal: open lock file pid/inet.smtp: cannot create file exclusively: No such file or directory
[16:26:17] <f3ew> cpm, users defined in AD?
[16:26:24] <f3ew> xeer, permissions?
[16:26:46] <xeer> it doesn't even tell me where it's trying to create the file
[16:26:58] <xeer> like i'm suppose to know where ./pid is
[16:28:25] <cpm> f3ew, ah, ldap-isms. K, I get it.
[16:28:46] *** c0m has quit IRC
[16:32:33] <tuxick> how to specify in which order milters are called?
[16:33:58] *** toytoy has quit IRC
[16:42:49] *** c0m has joined #postfix
[16:49:53] *** saurabhb has quit IRC
[16:51:05] *** saurabhb has joined #postfix
[16:54:03] <rapha_> When I get "status=deferred (mail system configuration error)", is there anything I can do to get more information as to what or where that error might be?
[17:09:28] <f3ew> xeer queue_directory
[17:09:38] <f3ew> rapha_ a few lines before that?
[17:11:33] <rapha_> f3ew: got the problem fixed already - Postfix suggested that "pipe flag `D' requires dovecot_destination_recipient_limit = 1" and that helped.
[17:13:11] *** loadkast has quit IRC
[17:19:39] *** AcTiVaTe has quit IRC
[17:24:36] *** diveli has quit IRC
[17:25:05] *** diveli has joined #postfix
[17:29:42] *** F6F has joined #postfix
[17:29:57] *** sbathe_ has joined #postfix
[17:30:26] *** diveli has quit IRC
[17:30:54] *** diveli has joined #postfix
[17:37:55] *** gonewestcoast has quit IRC
[17:38:16] *** gonewestcoast has joined #postfix
[17:42:03] *** rapha_ has left #postfix
[17:43:16] *** saurabhb has quit IRC
[17:50:17] *** dusty has quit IRC
[17:50:52] *** dusty_ has joined #postfix
[17:50:53] *** growltiger has joined #postfix
[17:53:51] *** JasonF has joined #postfix
[17:54:21] <JasonF> Hey. I'm wondering if it's possible (supported in postfix) to forward authentication credentials, i.e. in a "proxy" type of situation
[17:55:03] <JasonF> user  -(using user:pass)-> postfix -(using user:pass)-> destination
[17:55:58] <sysmonk> nope, atleast not that i know of
[17:56:25] <seekwill> Why does the destination need a u/p?
[17:56:49] *** phnord has quit IRC
[17:56:53] <JasonF> seekwill: it doesn't /have/ to have one, but giving it one makes nice things happen that we'd like to happen
[17:57:13] <seekwill> Like?
[17:57:18] <sysmonk> yeah, how nice
[17:57:23] <sysmonk> and what benefit will i have from that?
[17:57:24] <sysmonk> ;)
[17:57:39] <JasonF> like in Exchange, it makes the user appear as an internal user (GAL) rather than as some random internet guy
[17:58:05] <sysmonk> ah, so no benefits...
[17:58:11] <f3ew> JasonF no
[17:58:19] <JasonF> hey f3ew
[17:58:39] <JasonF> you've gotta be in more IRC channels than me. you're in at least 75% of all ones I join
[18:00:20] <sysmonk> JasonF: f3ew is a bot
[18:00:31] *** Severed_Head_Of_ has joined #postfix
[18:00:49] <f3ew> JasonF, only two I think
[18:01:03] <f3ew> JasonF, you should be able to trivially auth to AD though
[18:01:31] <JasonF> f3ew: it's not /quite/ that easy :)
[18:02:04] <f3ew> JasonF, after you have configured Cyrus SASL, it _is_ easy
[18:03:06] <JasonF> f3ew: I know the auth itself is that easy
[18:03:52] <JasonF> but not but the environment is too complex to allow for it being done that way
[18:03:53] <f3ew> layer 8 problems?
[18:03:56] <f3ew> Ah
[18:04:27] * f3ew wonders if you could just pass certificates to your users?
[18:04:42] <f3ew> SSL certificates
[18:04:58] <JasonF> that's exceptionally unpossible
[18:05:19] * f3ew gives JasonF a cluestick
[18:05:22] <JasonF> If I could get all my users to modify configs, I'd just have them all auth to the destination directly rather than going through my proxy.
[18:05:42] <f3ew> what's #pihlopase ?
[18:05:48] <f3ew> heh
[18:06:01] <f3ew> Is this some sort of ISP setup?
[18:06:11] <JasonF> f3ew: a wreched hive of scum and villany (not really, just some friends of mine)
[18:06:16] <JasonF> f3ew: yeah, sorta.
[18:06:34] <f3ew> hmmm
[18:06:48] *** growltiger_ has quit IRC
[18:07:04] <f3ew> I wonder if you could trivially reimplement the SASLauthd protocol?
[18:07:33] <f3ew> http://cpan.uwinnipeg.ca/htdocs/POE-Component-SASLAuthd/README.html
[18:07:38] <f3ew> MUHAHAHAHAHAHAHAHAHAHAHA
[18:07:38] <jelly> saslauthd is rather simple
[18:07:45] <f3ew> What I was thinking about :P
[18:08:12] *** growltiger_ has joined #postfix
[18:11:47] *** Cypher has joined #postfix
[18:13:09] *** hark has quit IRC
[18:13:09] *** growltiger has quit IRC
[18:13:16] *** pfak has joined #postfix
[18:13:22] *** pfak has left #postfix
[18:17:13] *** F6F has quit IRC
[18:19:05] *** bisoc has quit IRC
[18:19:16] *** bisoc has joined #postfix
[18:19:38] *** Zelest has quit IRC
[18:20:20] *** sbathe_ has quit IRC
[18:21:32] *** hark has joined #postfix
[18:26:31] *** Severed_Head_Of_ has quit IRC
[18:30:50] <f3ew> http://search.cpan.org/dist/POE-Component-SASLAuthd/lib/POE/Component/SASLAuthd.pm
[18:30:51] *** aronofsky has joined #postfix
[18:30:54] * f3ew heart Perl
[18:31:29] <aronofsky> hi, can anyone tell me if the msg "Lost connection while receiving the initial server greeting" can come from a blacklisting on the relay host ?
[18:32:18] *** tombar has joined #postfix
[18:32:27] <gonewestcoast> aronofsky: Not usually.  Not within Postfix, anyway.
[18:32:38] <gonewestcoast> There are a LOT of bustified MTAs out there.
[18:33:54] <aronofsky> gonewestcoast: actually the thing is strange because I can telnet thx relay MX from anoter IP and I cant telnet other MXs from the computer on which I got that error
[18:34:02] <aronofsky> dont know if im clear here ...
[18:34:18] <gonewestcoast> Not exactly.
[18:34:26] <gonewestcoast> Although they COULD be router blackholing you. :)
[18:34:50] *** internat1 has joined #postfix
[18:34:52] *** internat has quit IRC
[18:35:51] <aronofsky> ok I got 2 computers A and B and 2 remote MX relay ... A->MX1 gives me the error , A->MX2 OK, B->MX1, B->MX2
[18:36:07] <aronofsky> thats why i was thinking to some kind of blacklisting
[18:36:24] <gonewestcoast> So you think MX1 is blocking the connection from A only?
[18:36:29] <aronofsky> A and B having differents public adresses$
[18:36:46] <aronofsky> yes , thats the only explanation I saw ...
[18:36:52] <aronofsky> maybe you see another ?
[18:37:08] <gonewestcoast> Can you pastebin the telnet session to that host?
[18:37:32] <gonewestcoast> You also might make sure that your DNS on host A is set correctly, backwards and forwards.
[18:38:05] <aronofsky> gonewestcoast: what specific DNS problem are you thinking of ?
[18:38:19] <gonewestcoast> Some places are anal about FcRDNS.
[18:38:37] <gonewestcoast> You'd also probably want to make sure your HELO matches.
[18:38:42] <gonewestcoast> Just a shot in the dark...
[18:39:05] <aronofsky> gonewestcoast: what do you mean by HELO matches ?
[18:39:31] <gonewestcoast> Okay, my dns A record matches my PTR here, of alcatraz.sequestered.net
[18:39:53] <aronofsky> yes youre right , i will check that
[18:40:00] <gonewestcoast> so my mx here HELOs as "alcatraz.sequestered.net" instead of "heywood.jablowme.com" :)
[18:40:07] *** j_s has joined #postfix
[18:40:20] <gonewestcoast> If the HELO doesn't match, you might see issues from poorly configured mailservers.  Can you get me that pastebin?
[18:40:23] <gonewestcoast> Feel free to PM it to me.
[18:41:21] <aronofsky> i will recheck everything and i come back to you if im still stucked
[18:41:28] <aronofsky> thx for your answers
[18:41:40] <gonewestcoast> No worries.
[18:41:45] *** aronofsky has quit IRC
[19:06:49] *** danbeck has joined #postfix
[19:06:57] *** growltiger has joined #postfix
[19:08:20] *** Aw0L has left #postfix
[19:11:54] *** madrescher has quit IRC
[19:13:21] *** Fallenou has quit IRC
[19:14:58] *** krawek has joined #postfix
[19:16:31] *** neoeinstein has joined #postfix
[19:24:12] *** growltiger_ has quit IRC
[19:29:20] *** diqpib has joined #postfix
[19:34:32] *** madrescher has joined #postfix
[19:35:45] *** Fallenou has joined #postfix
[19:37:54] *** PcPixel has joined #postfix
[19:38:22] *** internat1 has quit IRC
[19:38:29] *** internat has joined #postfix
[19:39:57] <PcPixel> something odd has just happened to my server. a sender connected & got a reject on the EHLO. however, they were still able to send mail & it made it through. pastebin: http://pastebin.com/d73308084 . Ideas?
[19:41:03] <seekwill> Uh
[19:41:35] <PcPixel> you can see where the IP was rejected in the EHLO, but somehow it was able to send mail through.
[19:41:51] <seekwill> How?
[19:41:58] <seekwill> Which line?
[19:42:18] <PcPixel> first line: connect from unknown[78.96.92.153]
[19:42:26] <PcPixel> reject: NOQUEUE: reject: EHLO from unknown[78.96.92.153]: 501 5.5.2 <{bot_hostname}>: Helo command rejected: Invalid name; proto=SMTP helo=<{bot_hostname}>
[19:42:36] <PcPixel> mail queueud: 2DE871130007: client=unknown[78.96.92.153]
[19:42:53] *** m1n3s6 has joined #postfix
[19:44:00] <PcPixel> so what i dont get is, if the greeting was rejected, how is it possible the mail got queued and delivered?
[19:44:03] <seekwill> Ah, misread the paste. Sorry
[19:45:08] <PcPixel> seekwill: no problem :)
[19:45:19] <PcPixel> yeah there is another mail in between them
[19:45:30] <PcPixel> i dunno if it matters, but the mail content looks Russian
[19:45:46] <PcPixel> but again, ehlo rejected & yet mail was queued
[19:46:56] <PcPixel> i can post up my main.cf if that'll help
[19:47:39] <PcPixel> also, heres the mail header from the accepted email: http://pastebin.com/d58f68c91
[19:49:25] *** Cypher has quit IRC
[19:49:58] <Nockian> PcPixel: they're two different messages. the one *was* rejected, the other one got through (from <jacqmaria at free dot fr> )
[19:50:20] <PcPixel> nockian: if that was true, then why dont i see another connection from that IP?
[19:50:56] <Nockian> because it's earlier in your log, maybe a minute or two
[19:51:00] <Nockian> grep for 4738
[19:51:13] <PcPixel> [4738]? ok. one sec
[19:51:16] <Nockian> 4813 is the one that got rejected properly
[19:51:50] <Nockian> 4738 is the one that got through
[19:52:26] <PcPixel> aaah. good eye there
[19:52:30] <PcPixel> ok give me one second
[19:53:40] <PcPixel> no.. dont think im seeing it
[19:54:34] <PcPixel> shouldnt i be able to find a connect from that IP earlier as well?
[19:56:45] <Nockian> http://pastebin.com/m5241f2f5
[19:57:26] <PcPixel> ok but thats still my point
[19:57:28] <PcPixel> it connected
[19:57:31] <PcPixel> it got rejected
[19:57:33] <PcPixel> but it queued
[19:58:03] <PcPixel> if i search my log for that IP i only have 4 instances of it. all of which are in that pastebin
[19:58:04] <Nockian> well, the offending message queue is 2DE871130007 - so maybe grep for that
[19:58:13] <PcPixel> i think i did that, but ill try again
[19:58:43] <PcPixel> nope
[19:58:49] <PcPixel> those 5 you see are the only 5 occurances
[19:58:50] <Nockian> you could also grep for '001601c8f0d3$01bb1874$f9a24eb8@srpfyqsk' - since that's the message ID in your spam message header
[19:59:30] <PcPixel> just trwo instances
[19:59:33] <PcPixel> again, in paste :/
[20:00:58] <Nockian> maybe someone else has some more insight...
[20:01:01] <PcPixel> there is no other way to trace a mail after connection other than what ive/youve suggested?
[20:01:11] <PcPixel> yeah thats what im hoping. i mean dont get me wrong. i really do appreciate the help :)
[20:01:28] <PcPixel> but either oim overlooking something really sinmple, or something is really wrong
[20:06:20] <PcPixel> would the {} in the greeting mess up the check?
[20:10:25] <PcPixel> and how would you greet a system as unknown in a tenlet session
[20:13:57] <krawek> hi guys
[20:14:20] <PcPixel> hi
[20:14:27] *** SuperSpy has joined #postfix
[20:16:27] <krawek> do you know how to accept a mail from any user (the user is not present in the passwd file) and process it using the mailbox_command ?
[20:17:47] <SuperSpy> I have set up postfix as in this (http://workaround.org/articles/ispmail-etch/) howto, but postfix always bounces the emails and reports "unknown user", if i check manually with postmap -q it finds the address though, but not if i actually try to send a mail there.  Anyone know where I should start looking to fix it?
[20:19:11] <Signum> SuperSpy: what does your mail.log say about the bouncing?
[20:19:19] *** madrescher has quit IRC
[20:20:24] <SuperSpy> Jul 28 13:09:34 cvimail postfix/local[3233]: 7C530D98057: to=<admin at cvimail dot hopto.org>, relay=local, delay=21, delays=21/0.01/0/0.01, dsn=5.1.1, status=bounced (unknown user: "admin")
[20:21:20] <Signum> SuperSpy: you defined that domain as a local (!) domain. did you want that?
[20:22:19] <SuperSpy> well, im sort of a noob regarding email, so i dont know for sure
[20:24:00] <SuperSpy> the 'relayhost = ' setting in main.cf is blank, would that be the cause of that?
[20:24:07] *** Joe_Wulf has joined #postfix
[20:26:28] <Signum> SuperSpy: the tutorial explains in deep detail the difference of local versus virtual domains.
[20:26:49] <SuperSpy> yes, im (re-) reading that right now
[20:27:04] <PcPixel> Signum: any thoughts on my problem? :)
[20:27:41] <SuperSpy> ill see if i just skipped something silly then, thanks =)
[20:30:13] *** m1n3s6 has left #postfix
[20:31:41] <PcPixel> Is there any way possible for mail to get queued after the EHLO was rejected? pastebin: http://pastebin.com/d73308084
[20:32:13] <cpm> No
[20:32:21] <cpm> those are 2 different mails
[20:33:19] *** JoeWulf has quit IRC
[20:34:31] <PcPixel> cpm: but then why when i look at the whole maillog can i not find the other connection
[20:34:39] <rob0> Why would you mung the "bot_hostname"? Who are you protecting?
[20:34:50] <PcPixel> im not munging it
[20:34:53] <PcPixel> thats whatsd in my maillog
[20:35:06] <cpm> whoa!
[20:35:16] <rob0> oh haha someone forgot to set up their ratware
[20:35:21] <PcPixel> basically
[20:35:33] <cpm> indeed
[20:35:37] <PcPixel> but whata irqing me is if i search for that IP address, there are only the instances of it you see in the pastebin
[20:35:40] <PcPixel> nowhere else does that show up
[20:35:57] <rob0> this is an unedited, uninterrupted sequence from maillog?
[20:36:18] <PcPixel> correct
[20:36:24] <PcPixel> all i did was locate the IP in question
[20:36:31] <PcPixel> select all around it to what i belived to be related
[20:36:32] <PcPixel> and pasted
[20:36:33] <rob0> but it only covers 2 seconds
[20:37:16] <PcPixel> i can paste more if you like
[20:37:31] <PcPixel> but like i said, if i search my entire log for that IP then the section you see is the only spot it appears
[20:37:49] <rob0> well, looks like maybe the bot did a RSET
[20:38:13] <PcPixel> TCP or email command?
[20:38:19] <rob0> maybe you don't have smtpd_require_helo
[20:38:24] <PcPixel> no i think i do
[20:38:37] <PcPixel> yeah cause if i telnet & issue commands it says it needs the helo first
[20:39:16] *** steve_z has joined #postfix
[20:39:24] <steve_z> Hi everybody
[20:39:51] <PcPixel> no its there: 503 5.5.1 Error: send HELO/EHLO first
[20:39:58] <PcPixel> so i am requiring it
[20:40:13] <rob0> ask the administrator, don't guess
[20:40:20] <PcPixel> i am the admin
[20:40:26] <rob0> so WHY GUESS?
[20:40:34] <PcPixel> im just multitasking so my brain is a little scattered. we're down one poerson
[20:40:52] 
[20:41:18] <PcPixel> plus my pc was unresponsive doing a backup so switching windows was a PITA
[20:41:22] <rob0> then take a break and clear your mind and get ready. If you're spewing spam, "postfix stop" before you go.
[20:41:24] <PcPixel> ah, now its responsive again
[20:42:13] <PcPixel> ok
[20:42:19] <PcPixel> now i can focus back on this
[20:42:29] <PcPixel> i dont belive im sending out any spam
[20:42:38] *** Emilio_Eiji has joined #postfix
[20:42:40] <PcPixel> ive been keeping an eye on ly logs & the connections all seem normal
[20:42:44] <PcPixel> my
[20:42:45] <Emilio_Eiji> boa tarde
[20:42:49] <Emilio_Eiji> algum brasileiro???
[20:42:52] * sysmonk see spammers!
[20:43:11] <PcPixel> sysmonk: huh?
[20:43:54] <sysmonk> PcPixel: just joking around
[20:44:04] <PcPixel> sysmonk: dont DO that lol ;)
[20:45:00] <steve_z> sorry, anybody can help me please? I want to redirect all received mail from my user (username@localhost , p.ej) to a gmail account (external at gmail dot com). How can I do that?
[20:46:19] <PcPixel> the other thing i have noticed, i am accepting more mail lately. i used to be at ~10% accepted. today ive reached 20%
[20:47:53] <rob0> maybe a DNSBL you're using went belly-up?
[20:48:02] <Emilio_Eiji> algum brasileiro aew???
[20:48:06] <PcPixel> only two im using are spanhaus & spamcop
[20:48:22] <PcPixel> im also using rfc-ignoratn
[20:48:24] <PcPixel> ignorant
[20:49:03] <PcPixel> no, three actually
[20:49:24] <PcPixel> zen.spamhaus.org, cbl.abuseat.org, and bl.spamcop.net
[20:49:43] <PcPixel> and before those, im using: dsn.rfc-ignorant.org
[20:50:05] <rob0> CBL is part of zen
[20:50:31] <PcPixel> rob0 so those two lines are redundant?
[20:52:29] *** Azrael_- has joined #postfix
[20:52:30] <Azrael_-> hi
[20:52:42] *** LordDicranius has joined #postfix
[20:54:13] <PcPixel> if you like, i can post a larger portion of my maillog
[20:55:27] <steve_z> sorry, anybody can help me please? I want to redirect all received mail from my user (username@localhost , p.ej) to a gmail account (external at gmail dot com). How can I do that? May be using ~/.forward file or are there a better way?
[20:56:01] <Azrael_-> http://de.pastebin.ca/1085491 <-- i have several mails that already stay in the deferred-queue and i want it to automatically inform the original senders after e.g. 5 retries. But till now nothing happens
[20:58:44] <Azrael_-> is this even possible?
[20:59:00] <Azrael_-> because till now i haven't found anything related to such a task in the docs
[20:59:54] <steve_z> nobody can help me?
[20:59:59] <steve_z> hello?
[21:00:04] <PcPixel> yup. no other connections from that IP outside of the log
[21:00:16] <Dominian> steve_z: are you using sendmail?
[21:00:23] <steve_z> no, i'm using postfix
[21:00:23] <Dominian> er.. wait
[21:00:25] <Dominian> stupid question
[21:00:32] <Dominian> .forward fle should work
[21:00:36] <PcPixel> but entertaining none-the-less
[21:00:37] <PcPixel> :)
[21:00:39] <steve_z> hi Dominian, thanks for answering
[21:01:06] <steve_z> Dominian, yeah it works but... I don't know if there are a better way for do that
[21:01:37] <Dominian> not that I know of.
[21:02:15] <steve_z> i receive a lot of spam (~400 mail/day) and I can't control it. And i want to redirect to gmail for that reason
[21:02:15] <sysmonk> Azrael_-: not 5 retries, but after _some_ time - yes
[21:02:25] <sysmonk> !delay_warning_time
[21:02:25] <knoba> sysmonk: "delay_warning_time" : a configuration parameter in the main.cf: The time after which the sender receives the message headers of mail that is still queued.
[21:02:32] *** danbeck has quit IRC
[21:02:33] <sysmonk> Azrael_-: ^^
[21:02:36] <PcPixel> ateve_z: are you doing any content/restrictions in your main.cf
[21:02:49] *** McJerry has joined #postfix
[21:02:56] <steve_z> hi PcPixel, nop
[21:03:05] <PcPixel> stve_z: you might want to do that
[21:03:15] <PcPixel> try strengthening your main.cf first
[21:03:27] <Azrael_-> sysmonk: thx, is there a way to kick them automatically after "some" time out of the queue, how long is it normally, where can i specify it?
[21:03:52] <steve_z> $ cat /etc/postfix/main.cf|grep restr
[21:03:52] <steve_z> smtpd_recipient_restrictions =
[21:03:52] <steve_z> smtpd_data_restrictions =
[21:03:59] 
[21:04:01] <sysmonk> Azrael_-: yes, default 5 days
[21:04:08] <PcPixel> steve_z: yes. but you ahve to actually have stuff in there
[21:04:22] <sysmonk> Azrael_-: look for *_queue_lifetime
[21:04:52] *** master_o1_master is now known as master_of_master
[21:05:47] <Azrael_-> thx
[21:06:54] <Azrael_-> strange: searching with google in postfix.org for "queue lifetime" returns no result
[21:07:33] <PcPixel> if i telnet'ed to my mail server, how would i identify myself as "unknown"?
[21:08:09] <sysmonk> PcPixel: telnet from a host which doesn't have a ptr entry
[21:09:14] <PcPixel> and greet as what
[21:09:17] <PcPixel> anything?
[21:09:23] <sysmonk> anything, valid.
[21:10:04] <PcPixel> ok then that shouldnt be it, cause im doing a check for that as well
[21:10:08] <PcPixel> ugh this is making me crazy
[21:10:09] <steve_z> using .forward file doesn't keep any copy in server, doesn't it?
[21:10:24] <sysmonk> PcPixel: 99.99$ and your problem will be fixed ;P
[21:10:52] <sysmonk> steve_z: depends on your .forward file source
[21:11:12] <PcPixel> sysmonk: at this point if someone could tell me the exat cause id almost be willing to pay it
[21:11:31] <sysmonk> PcPixel: almost, or not almost? :P
[21:11:49] <PcPixel> sysmonk: almost lol
[21:11:55] <PcPixel> nobody can seem to identify how this sucker got through
[21:12:01] <PcPixel> its just not adding up
[21:12:40] <sysmonk> PcPixel: pastebin everything ( or give access )
[21:12:44] <sysmonk> and ready your 99.99$
[21:12:53] <PcPixel> sysmonk: i can pastebin the entire maillog
[21:12:54] <sysmonk> :)
[21:13:08] <PcPixel> sysmonk: ok, ets see how good you are
[21:13:08] <sysmonk> PcPixel: -10 lines +10 lines
[21:13:18] <PcPixel> sysmonk: ok, illw iden by 20
[21:13:19] <steve_z> ok, thanks, bye
[21:13:19] <sysmonk> full log might be to much for me
[21:13:23] *** steve_z has left #postfix
[21:13:30] <sysmonk> (my log goes up to few gigs per day)
[21:13:34] <sysmonk> maillog that is
[21:14:49] <sysmonk> PcPixel: and, in another pastebin, pastebin your postconf -n + master.cf
[21:15:03] <PcPixel> ok thats never happened before
[21:15:08] <PcPixel> i tripped pastebins spam filter
[21:15:17] <sysmonk> haha :) happened to me once
[21:15:22] <PcPixel> any way around it?
[21:15:38] <sysmonk> PcPixel: echo >> FILE && post the file ;)
[21:15:46] <sysmonk> that is, post the link to the file
[21:15:55] <PcPixel> i dont have anywhere i can link to the file
[21:16:07] <sysmonk> ouch, no web server?!
[21:16:16] <cpm> try #flood
[21:16:17] <PcPixel> sysmonk: dont get me started on the IT infrastructure here
[21:16:24] <sysmonk> heh, k
[21:16:37] <sysmonk> PcPixel: dunno, mail me, dcc me, anything
[21:16:39] <sysmonk> doesn't matter
[21:16:54] <cpm> try #flood, , oh wait, I already said that
[21:17:13] <sysmonk> yea
[21:17:17] <sysmonk> PcPixel: try another pastebin
[21:17:19] <sysmonk> if not - then #flood
[21:17:27] <PcPixel> ok one sec
[21:19:01] <PcPixel> ok, got 2 out of 3
[21:19:18] <PcPixel> master.cf - http://pastebin.com/d5cadc096
[21:19:34] <PcPixel> postconf -n - http://pastebin.com/d73b783bd
[21:19:59] *** growltiger_ has joined #postfix
[21:20:16] <PcPixel> ok yeah it wont let me paste bin the log llike you asked. how do i use #flood?
[21:21:11] *** carl- has joined #postfix
[21:21:12] <sysmonk> dunno :)
[21:21:16] <sysmonk> just flood maybe?
[21:21:17] <PcPixel> DOH
[21:21:25] <PcPixel> ok let me stry splitting it in two pieces
[21:21:33] <sysmonk> PcPixel: you can flood into my PM
[21:21:38] <sysmonk> this time... ONLY :)
[21:21:56] *** Severed_Head_Of_ has joined #postfix
[21:22:11] <gonewestcoast> PcPixel: What exactly is your question?
[21:22:14] <gonewestcoast> I musta missed something.
[21:22:16] <PcPixel> ok that worked
[21:22:32] <PcPixel> maillog part a: http://pastebin.com/d2d841753
[21:22:33] <sysmonk> everybody wants 99.99? :P
[21:22:42] <PcPixel> maillog part b: http://pastebin.com/d6a0adbc
[21:22:42] *** carl- has quit IRC
[21:22:57] <sysmonk> and where is the problem?
[21:23:17] <PcPixel> sys,onk: if you look at part A
[21:23:25] <PcPixel> there is a system that connected: 78.96.92.153
[21:23:36] <sysmonk> ye
[21:23:45] <PcPixel> if you follow the log, it connects, gets rejected in the EHLO but then somehow queues a message & sends it through
[21:23:58] <PcPixel> but if i search the entire maillog, the ONLY connection it ever made was the one thats rejected
[21:24:22] <gonewestcoast> Does Postfix permit a fallback to HELO when EHLO fails?
[21:24:31] <PcPixel> but you can see it queued a message: 2DE871130007
[21:24:45] <PcPixel> and im no thiding anything either. everyrthing you see in the pastebins is genuine
[21:24:57] <gonewestcoast> smtp_discard_ehlo_keyword_address_maps = cidr:/etc/postfix/checks/pipeline_disable <-- What is this set to do?
[21:25:31] <PcPixel> one moment
[21:26:02] <PcPixel> currently only one line:
[21:26:03] <PcPixel> 203.150.230.74/32       pipelining
[21:26:03] <gonewestcoast> And <{bot_hostname} has to be generated somewhere that you're doing some kind of list comparison or regex for.
[21:26:20] <PcPixel> gonewest: not to my knowledge
[21:26:35] <PcPixel> i literally stopped postfix, took the log out, pastebined it
[21:26:37] <PcPixel> thats it
[21:26:52] <PcPixel> i FTPd the log out of the system
[21:26:56] <PcPixel> in its entireity
[21:27:33] <gonewestcoast> pcre:/etc/postfix/checks/helo_checks That's probably where it is.
[21:28:09] <gonewestcoast> I'd ask the mailing list.  If Wietse doesn't know, nobody knows. :D
[21:28:13] <PcPixel> helo_checks: http://pastebin.com/d32fd30c4
[21:28:43] <PcPixel> gonewestcoat: i think thats my next step. Travel to Oz & ask the Great Wizard. ;)
[21:29:07] <sysmonk> PcPixel: ghm
[21:29:10] <sysmonk> PcPixel: what's your ip?
[21:29:14] <sysmonk> pm it to me
[21:29:16] *** McJerry has quit IRC
[21:29:36] *** McJerry has joined #postfix
[21:31:36] <gonewestcoast> It's in one of your many, many checks in smtpd_recipient_restrictions
[21:31:37] <gonewestcoast> :)
[21:33:04] <gonewestcoast> Theoretically anyway.
[21:33:13] *** Filbert has quit IRC
[21:33:17] <gonewestcoast> It ALMOST looks like a warn_if_reject is set.
[21:35:26] *** AnotherDay` has joined #postfix
[21:36:22] *** vegbx has joined #postfix
[21:36:52] *** growltiger has quit IRC
[21:37:30] *** growltiger has joined #postfix
[21:39:35] *** growltiger_ has quit IRC
[21:40:42] <PcPixel> no, we got it
[21:40:52] <gonewestcoast> What was it?
[21:40:55] <PcPixel> i feel a touch stupid
[21:41:05] <PcPixel> its only logging the errors
[21:41:07] <PcPixel> and my hard limit is 10
[21:41:16] <PcPixel> so first try: helo im.totally-invalid#!@$
[21:41:21] <PcPixel> log error, but STILL connected
[21:41:26] *** Severed_Head_Of_ has quit IRC
[21:41:27] <PcPixel> second try: helo now.im.ok
[21:41:31] <gonewestcoast> Doof. :)
[21:41:32] <PcPixel> no log, keep going
[21:41:37] <PcPixel> my sentiments exactly
[21:42:04] <gonewestcoast> Where was that set?
[21:42:10] <gonewestcoast> syslog?
[21:42:23] <PcPixel> thats what im asking him now :)
[21:42:51] <gonewestcoast> It'd almost have to be in syslog, I show nothing in your postconf.
[21:43:21] <PcPixel> yeah
[21:43:25] <PcPixel> i think its at the log level
[21:43:29] <PcPixel> br, bathroom break
[21:44:03] <gonewestcoast> Enjoy piddling.
[21:44:10] *** vertigo- has quit IRC
[21:44:24] <sysmonk> ... not in syslog
[21:44:39] <sysmonk> you don't log ehlo/helo on successiful delivery
[21:44:46] <sysmonk> but it does log ehlo/helo on a REJECT
[21:45:01] <sysmonk> so if my firs try was a REJECT it DOES log the ehlo/helo
[21:45:21] <sysmonk> and then i supply a good ehlo/helo and do deliver the email - so it goes fine without logging ehlo/helo
[21:45:52] <PcPixel> back
[21:46:18] *** cpm has quit IRC
[21:51:33] *** growltiger_ has joined #postfix
[21:51:42] *** vertigo- has joined #postfix
[21:56:17] <vegbx> which is better for mx filtering, emailmx or mxlogic?
[21:57:00] <sysmonk> let me think... i'd take the third choice!
[21:57:25] <vegbx> whts the third choice?
[21:57:25] <vegbx> lol
[21:57:42] <sysmonk> don't know, but atleast it's none of the first two
[22:00:28] <PcPixel> THAT was funny
[22:00:34] <vegbx> i see
[22:01:13] <sysmonk> vegbx: telling the truth, i don't know what those 2 words mean
[22:01:22] <sysmonk> and if i don't, then mostly it means they're not worth it :P
[22:01:23] <vegbx> ahhh
[22:01:38] <vegbx> Just trying to find a good way to manage spam
[22:01:39] <adaptr> vegbx: at the veay least, they have absolutely nothing to do with postfix
[22:01:48] <sysmonk> yup
[22:01:50] <adaptr> so install amavisd with SA
[22:02:01] <adaptr> you get clamav thrown in for free
[22:02:08] <vegbx> well i want a company to manage the lists and so forth
[22:02:49] <vegbx> I currently use postfix/amavisd/clamav/mailscanner
[22:02:50] <sysmonk> then... maybe it's a wrong channel vegbx
[22:03:02] <vegbx> but i got a client that is not satisfied with those results
[22:03:18] <adaptr> then sell him outhouse and charge him double
[22:03:26] <PcPixel> the Great One said not to use MAilScanner
[22:03:38] <vegbx> i mean mailwatch
[22:03:38] <adaptr> a LONG TIME AGO
[22:03:39] <sysmonk> vegbx: hire 2000 chineese, they'll manage your spam for less than those 2 services take :P
[22:03:50] <pingouin> :D
[22:03:56] <vegbx> umm so i heard mx filtering is pretty good
[22:03:58] <PcPixel> sysmonk: i like you :)
[22:04:11] <sysmonk> PcPixel: should i be afraid already? :P
[22:04:11] <adaptr> vegbx: I have no idea what you think that means
[22:04:23] <vegbx> okay okay
[22:04:28] <PcPixel> sysmonk: no, its not like that. :P
[22:04:31] <sysmonk> vegbx: ask those 2 companies for a test-drive
[22:04:39] <sysmonk> and see if they match your needs
[22:04:49] <vegbx> I will probably do that
[22:04:56] <sysmonk> we can't tell what you think is right
[22:05:02] <sysmonk> and what kind of mails do your clients get
[22:05:21] <sysmonk> if you're clients are porn industries, then yeah, there would be lots of false positives ;P
[22:05:31] <sysmonk> or rolex re-sellers :P
[22:05:46] <sysmonk> or maybe medical industry selling viagra or something :P
[22:08:08] <krawek> is it possible to accept and save mail for non-existent users?
[22:08:21] *** growltiger has quit IRC
[22:09:22] <sysmonk> !luser_relay
[22:09:23] <knoba> sysmonk: "luser_relay" : a configuration parameter in the main.cf: Optional catch-all destination for unknown local(8) recipients. By default, mail for unknown recipients in domains that match $mydestination, $inet_interfaces or $proxy_interfaces is returned as undeliverable.
[22:09:27] <sysmonk> + catchalls
[22:09:28] <sysmonk> krawek: ^^
[22:10:47] <krawek> sysmonk: does that call to the mailbox_command ?
[22:11:08] <krawek> I have something like: luser_relay: $user
[22:11:34] <adaptr> nonono
[22:11:40] <adaptr> *destination*
[22:11:41] <krawek> and I am using mailbox_command to save the rfc822 data
[22:11:44] <adaptr> read what it says
[22:14:41] *** PcPixel has quit IRC
[22:14:50] *** LordDicranius has quit IRC
[22:15:00] <sysmonk> and there went my 99.99$ :(
[22:15:25] <krawek> thanks guys, I think I got it
[22:20:48] *** Severed_Head_Of_ has joined #postfix
[22:25:58] *** tombar_ has joined #postfix
[22:25:59] <seekwill> Oh my...
[22:26:06] <seekwill> Need some DNS help... :)
[22:26:37] <seekwill> dig lists.thekelleys.org.uk mx  has two MX records and a CNAME. What's the CNAME mean in that context?
[22:27:00] <adaptr> who knows ? what is the CNAME ?
[22:27:10] <seekwill> The cname is thekelleys.org.uk
[22:27:20] *** AnotherDay` has quit IRC
[22:27:21] <adaptr> and neitehr MX points to that ?
[22:27:31] <seekwill> mx5 points to that
[22:27:48] <adaptr> so lists is not an actual A record, but points to the mail server, which is, again, itself
[22:27:53] <adaptr> simple
[22:28:00] <adaptr> nothing wrong with it
[22:28:41] <seekwill> oh
[22:29:08] <seekwill> lists. is a cname with mx records...
[22:29:27] <seekwill> I guess this is to organize the listserver...
[22:29:36] <seekwill> Futureproofing...
[22:32:11] <adaptr> no, it's just economizing
[22:32:31] <rob0> lists.thekelleys.org.uk. 23288  IN      CNAME   thekelleys.org.uk.
[22:32:36] <adaptr> it's a way to create MXen for a subdomain that has no physical existence, as it points to the parent server
[22:33:02] <adaptr> do a dig on the MX for the parent domain, if it's different from the lists one, you know why they did it
[22:33:21] <adaptr> (to assign a different MX to the lists subdomain)
[22:33:55] <rob0> lists.thekelleys.org.uk. being a CNAME cannot have its own MX. It uses the MX of the name it points to.
[22:34:31] *** Emilio_Eiji has left #postfix
[22:34:42] <adaptr> and yet it has one set
[22:34:55] <adaptr> two, even
[22:35:08] <rob0> thekelleys.org.uk. (parent, and target of the CNAME) has the MX
[22:35:50] <adaptr> seekwill: you silly sod, why did you lie lie lie ?
[22:35:59] <adaptr> lists does not have any MX records
[22:36:05] <rob0> pants on fire
[22:36:27] <adaptr> oh wait, org.uk does, but co.uk doesn't
[22:36:32] <adaptr> rob0: which one did you check ?
[22:36:39] <seekwill> oh
[22:37:04] <adaptr> and are you implying that it would not *look* at those MX records ? because there';s nothing in the RFC that says that a CNAME may not have MXen
[22:37:08] *** jellis-real has quit IRC
[22:37:23] <adaptr> au contraire, the RFC specifically states that an MX may not *BE* a CNAME
[22:37:31] <adaptr> it must be an A record
[22:37:36] <adaptr> and that's not even enforced
[22:37:46] <rob0> A CNAME cannot coexist on the same RR name with *any* other RR type. A CNAME must be the only record at any given name.
[22:37:46] <adaptr> as I cheet all der time
[22:37:59] <adaptr> well, he dun it anyway
[22:38:13] <rob0> BIND 9 will not allow "CNAME and other data", and no, Simon didn't do that.
[22:38:19] <adaptr> oh, he didna
[22:38:37] <rob0> The CNAME points to a name which has MX.
[22:38:41] <adaptr> damn SQL users - learn DNS already!
[22:38:41] <seekwill> What's RR?
[22:38:47] * seekwill doesn't think railroad
[22:38:48] *** growltiger_ has quit IRC
[22:38:53] <adaptr> he's not Simon, he is Will
[22:38:54] <rob0> damn SQL users - learn DNS already!
[22:39:00] <seekwill> :(
[22:39:09] <adaptr> he's good at SQL, granted
[22:39:09] <rob0> Fire at will!
[22:39:16] <seekwill> adaptr: Who says?
[22:39:41] <adaptr> but he sucks worse at DNS than I do at .. at.. DNS!
[22:39:48] <adaptr> seekwill: I observe and judge ;)
[22:39:58] *** McJerry has quit IRC
[22:41:25] <seekwill> Hmm.. I think I get it
[22:41:40] <sysmonk> seekwill: dns has rail roads
[22:41:44] <rob0> file:///usr/doc/bind-9.4.2/arm/Bv9ARM.ch06.html#types_of_resource_records_and_when_to_use_them
[22:41:45] <sysmonk> dns is all about rail roads
[22:41:51] <sysmonk> or resource records
[22:41:54] <sysmonk> but i like rail roads more!
[22:41:58] <seekwill> :)
[22:42:23] <rob0> change the path to suit, but that should help
[22:42:56] *** McJerry has joined #postfix
[22:43:05] <adaptr>  /usr/share/doc ftw!
[22:44:22] *** tombar has quit IRC
[22:44:49] *** Mez has joined #postfix
[22:45:04] <seekwill> Awesome, thanks
[22:45:15] <seekwill> So much to read
[22:45:32] <sysmonk> oh no! you have to read now :(
[22:45:49] * seekwill makes adaptr do his reading
[22:45:51] <sysmonk> f3ew: bastard, now he has to spend his time for that :(
[22:46:04] <sysmonk> woops, f3ew , that should be rob0 :)
[22:46:08] <seekwill> I could be doing something more productive, like Halo
[22:46:33] *** F6F has joined #postfix
[22:46:54] *** F6F has quit IRC
[22:47:37] <Mez> whats the easiest way to delete all messages from a queue.
[22:47:46] <Mez> Got a couple that have been being deferred for like - a week now
[22:48:06] <rob0> mkfs(8)
[22:48:52] <Mez> kill -9 rob0
[22:48:58] <Mez> found it anywya
[22:49:30] <sysmonk> dd if=/dev/zero ...
[22:50:40] <adaptr> Mez: postsuper -d ALL
[22:50:46] <adaptr> just make sure you mean it
[22:51:28] <Mez> adaptr, yeah, it was trying to deliver to a non-addressable thing
[22:51:31] <sysmonk>  adaptr 'found it anyway'
[22:51:38] <Mez> I found it anyways :)
[22:51:56] <sysmonk> adaptr: your slow :P
[22:52:36] <adaptr> killall monks
[22:53:02] <sysmonk> monks are nice, monks kick useless adaptrs
[22:53:31] <xpoint> rob0, nice file:/// link
[22:53:45] <adaptr> heh, but what will you do if your monk's cellphone is dead and he needs a Tibet -> EU adaptr ?
[22:54:05] <sysmonk> xpoint: same one you should have printed out years ago and read
[22:54:07] *** SuperSpy has quit IRC
[22:54:13] <sysmonk> bv*arm's are nice
[23:00:00] *** unixSnob has joined #postfix
[23:08:26] <xpoint> yes localhost i nice 2
[23:09:14] <xpoint> http://opengrok.creo.hu/openbsd/xref/src/usr.sbin/bind/doc/arm/Bv9ARM-book.xml my firefox do not parse it, so i see it in raw :/
[23:09:51] <sysmonk> xpoint: oO isc.org/bind/
[23:09:58] <sysmonk> html + pdf available afair
[23:10:24] <adaptr> I think it's in PDF and plain text
[23:10:43] *** j_s has quit IRC
[23:10:43] <sysmonk> adaptr: we're both lazy to go and see ;P
[23:11:18] <adaptr> memory is good enough when I'm not getting paid
[23:24:51] *** brancaleone has joined #postfix
[23:37:35] *** neuonyx has quit IRC
[23:40:04] *** tombar_ has quit IRC
[23:40:47] *** remalit has joined #postfix
[23:51:25] *** tomocha6 has quit IRC
[23:53:10] *** tomocha6 has joined #postfix
[23:53:53] <xeer> "Held 2 messages in /var/spool/mail/root". What variables need modification to have it store it in Maildir format?
[23:55:18] <adaptr> !home_mailbox
[23:55:19] <knoba> adaptr: "home_mailbox" : a configuration parameter in the main.cf: Optional pathname of a mailbox file relative to a local(8) user's home directory.
[23:55:31] <adaptr> append it with a slash (/) to specify maildir delivery
[23:55:38] <adaptr> postfix will create the maildir if not present
[23:55:55] <adaptr> but postfix *WILL NOT* deliver to root
[23:55:56] <adaptr> evah
[23:56:04] <xeer> Right, relative to the user's home directory. But I don't want eighty bajillion user (uid's) accounts for all my address. I
[23:56:14] <xeer> I'm trying to go for virtual users
[23:56:16] <adaptr> it only works on local users
[23:56:26] <adaptr> and root is most definitely local
[23:56:38] <adaptr> so.. what did you break ?
[23:57:05] <adaptr> root. must. have. alias....root. must. have. alias...
[23:57:13] <xeer> I see..
[23:57:29] <xeer> I'll have postmaster be an alias of root?
[23:58:15] *** Fallenou has quit IRC
[23:58:30] <xeer> If I specifiy the home_mailbox, I still wont get messages?
[23:58:41] <adaptr> no, root must go TO an actual user
[23:58:55] <adaptr> postmaster is an alias OF root
[23:59:27] <xeer> and I need virtual_alias_maps for that?
[23:59:56] <adaptr> for *what* ?

top