May 10, 2007  
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

[00:01:01] <chessir> I am running xunbuntu 6.1 trying to do virtualization but having priblems
[00:01:22] <chessir> have you done virtual?
[00:06:51] <rob0> !virtual
[00:06:52] <knoba> rob0: '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
[00:07:02] <rob0> I have done some, not much.
[00:07:09] *** karvad has joined #postfix
[00:14:02] *** Roobarb has quit IRC
[00:14:22] *** Roobarb has joined #postfix
[00:22:24] <chessir> something seems to be wrong with my vmplayer but I cannot seem to uninstall or remove so I can reinstall
[00:23:02] <chessir> no clear instructions on what to do if this happens
[00:24:50] *** Taube is now known as taube
[00:29:41] <okta> is thee a way to make postfix to add a header line in the process of checking so i can take this into account later when doing the mime checks (eg. adding it while checking smtpd_recipient_restrictions)?
[00:30:02] *** pirho has quit IRC
[00:30:44] *** karvad has quit IRC
[00:32:22] *** pirho has joined #postfix
[00:32:33] *** hemry has quit IRC
[00:33:19] <rob0> LOL: "Catch suspicious messages before you open them - with Windows Live Hotmail." That assumes that hotmail didn't just throw it away!
[00:35:26] *** flami has quit IRC
[00:53:41] *** okta has quit IRC
[00:54:32] *** Mr_Sako has joined #postfix
[00:56:33] *** dj-fu has quit IRC
[00:59:26] *** Mr_Sako has quit IRC
[01:00:54] <xpoint> rob0, html mail is ewill
[01:02:29] <rob0> Sure, but the thing with hotmail, we always have people here asking, "Where did my message to $FOO@hotmail go? The log says 'Accepted for delivery.'"
[01:03:19] <rob0> One of the most F'ly of all Q's A'ed in this channel.
[01:03:32] <rob0> (and probably other MTA channels too.)
[01:03:56] *** imcsk8 has quit IRC
[01:09:43] <jduggan> i wonder which domain handles the most inboxes like, ever
[01:09:48] <jduggan> hotmail?
[01:09:53] <jduggan> aol?
[01:10:55] <jduggan> are there any sorta stats on this?
[01:11:13] <xpoint> jduggan, senderbase
[01:12:08] <xpoint> jduggan, dont know if it is this stats you need, but something along the way it is
[01:12:47] <rob0> I think f3ew said it was AOL.
[01:13:01] <xpoint> rob0, hotmail have accourding to there spf record, 1 million mta that can handle spam scanning before queue
[01:13:23] <jduggan> wow, ran by ironport... i recently installed two ironports
[01:13:27] <jduggan> what a coincidence
[01:14:02] *** Roobarb has quit IRC
[01:14:09] <jduggan> omg i never realised hotmail were doing SPF checks
[01:14:22] *** Roobarb has joined #postfix
[01:14:49] <xpoint> jduggan, maybe not doing, but hotmail.com has spf records
[01:15:19] <xpoint> jduggan, i reject alot of spam from it :-)
[01:16:01] <jduggan> aah, sorry yea, they provide spf records
[01:16:02] <jduggan> hmm
[01:17:17] <xpoint> jduggan, send a mail to a hotmail.com email if you have one, and use abuse at junc dot org as sender :-)
[01:18:00] <xpoint> did you recieve it ?
[01:18:05] <jduggan> i dont have one
[01:18:06] <jduggan> :)
[01:18:18] <xpoint> good :)
[01:18:31] <jduggan> no i mean, i dont have a hotmail address so i cant check
[01:18:32] <higuita> hotmail isnt using SPF, its using sender id...
[01:18:32] <jduggan> :P
[01:18:45] <jduggan> what is sender id?
[01:18:49] <xpoint> higuita, so ?
[01:19:12] <higuita> they are very close, but spf is for the envelope sender and sender id is for the mail sender
[01:19:29] <xpoint> higuita, why do thay provide something that is not used ?
[01:20:02] <higuita> most of the time they are the same, but not always, and that sometimes cause problems, as SPF is setup for one field and might not be totally true for the other
[01:20:03] <xpoint> nice and realy needed :-)
[01:20:31] <higuita> sender id is from then, they want people to use it (they have a patent about it)
[01:20:50] <xpoint> higuita, is there any known sender id software for our os then ?
[01:21:02] <higuita> but luckly nobody uses it and they had to steal the needed info for sender id to work from the SPF fields
[01:21:33] <higuita> AFAIK, only exchange 2007 have support for sender id
[01:21:50] <jduggan> sorry, but in dumb mans terms, sender id is what 'generic' field found in a typical mail client?
[01:21:53] *** havoc has joined #postfix
[01:21:56] <havoc> evening
[01:21:57] <xpoint> ah and nobody cares about exchange :-)
[01:21:59] *** alecs has joined #postfix
[01:22:02] <alecs> hi there
[01:22:16] <higuita> and almost no domain have sender id fields in the DNS, but many already have SPF fields
[01:22:23] <alecs> post fix is for smtp .... but what to use for POP ?
[01:22:26] *** dj-fu has joined #postfix
[01:22:27] <alecs> *** POP3
[01:22:38] <higuita> popa3d if you only want pop3
[01:22:59] <alecs> is it courier good ?
[01:23:02] <jduggan> envelope sender is the FROM: address in outlook etc?
[01:23:06] <higuita> pop3 and imap, i use dovecot
[01:23:15] <xpoint> courier-mta have it all
[01:23:31] <xpoint> jduggan, yes
[01:24:01] <jduggan> xpoint: So which generic  field is sender id?
[01:24:08] <xpoint> jduggan, but does outlook sends it as enveloppe is another thing
[01:25:08] <xpoint> jduggan, in postfix its Return-Path: with is allways last added header, eg top line in every mailheader
[01:25:45] <jduggan> so its teh 'REPLY TO' field in in ur mail client?
[01:25:55] <xpoint> jduggan, no
[01:26:16] <xpoint> this is completely another header
[01:26:19] <higuita> jduggan: IIRC, the sender id is the from that is shown in the email client (the from: in the message headers), the SPF is the one that is set in the "mail from:" command in the smtp protocol (sender: and return-path in the headers)
[01:32:12] <xpoint> higuita, http://why.junc.info/cgi-bin/pypd.cgi?ip=83.13.139.246
[01:32:36] <xpoint> i check spf :-)
[01:33:25] *** binwiederhier has quit IRC
[01:33:59] *** war has quit IRC
[01:39:25] *** tiwari has joined #postfix
[01:39:31] * xpoint will have tea now, others can sleep now :-)
[01:39:38] *** havoc has left #postfix
[01:43:06] *** pickcoder has quit IRC
[01:43:48] *** technoid_ has joined #postfix
[01:53:34] *** tiwari has quit IRC
[01:54:26] *** hparker has quit IRC
[01:57:36] *** Mazon is now known as mazon
[02:03:17] *** hparker has joined #postfix
[02:05:11] *** dj-fu is now known as fujin
[02:05:22] *** Ryushin has quit IRC
[02:07:18] *** pirho has quit IRC
[02:07:31] *** chessir has quit IRC
[02:08:46] *** pirho has joined #postfix
[02:14:03] *** Roobarb has quit IRC
[02:14:22] *** Roobarb has joined #postfix
[02:23:35] *** alecs has quit IRC
[02:32:54] *** jpon has joined #postfix
[02:33:55] *** fujin has quit IRC
[02:34:07] *** scoates has joined #postfix
[02:34:10] <scoates> hi
[02:34:17] <Supaplex> oi
[02:34:53] <scoates> postfix broke. All mail is getting rejected: May  9 20:32:59 mail2 postfix/smtpd[17590]: NOQUEUE: reject: RCPT from localhost[127.0.0.1]: 450 <localhost>: Helo command rejected: Host not found; from=<sean at iconoclast dot caedmon.net> to=<sean at mail2 dot tabini.ca> proto=ESMTP helo=<localhost>    any idea where I should look?
[02:35:52] <rob0> DNS / resolver issues, maybe chroot too.
[02:36:24] *** pirho has quit IRC
[02:36:38] <scoates> not chroot
[02:36:41] <scoates> and the box resolves fine
[02:36:55] <scoates> the only thing I changed was my netmask from 255.255.255.0 to 255.255.0.0
[02:37:25] <scoates> which "Host not found" is it talking about?
[02:37:44] *** sc00p has joined #postfix
[02:37:49] <shasta> localhost :)
[02:37:58] <rob0> 450 <localhost>: Helo command rejected
[02:38:03] <scoates> well, ping localhost is fine
[02:38:18] <scoates> there's an entry in /etc/hosts for it
[02:38:39] <scoates> oddly, eth0 shows up before lo in ifconfig
[02:38:41] <scoates> .. ?
[02:39:37] <shasta> their order doesn't matter
[02:40:36] <scoates> yeah, it just seems odd
[02:40:52] <scoates> oh.. how do I find out why postfix can't resolve...... itself?
[02:42:51] <scoates> this also changed recently: mynetworks = 127.0.0.0/24, 10.8.2.0/24
[02:42:58] <scoates> is that wrong?
[02:43:09] *** rmayorga has quit IRC
[02:48:20] <scoates> well, I cheated by commenting: #       -o smtp_helo_timeout=5 -o smtp_connect_timeout=5
[02:48:23] <scoates> #    -o smtpd_helo_restrictions=reject_unknown_hostname
[02:49:54] *** owh has joined #postfix
[02:50:40] *** MezzoTiNt has joined #postfix
[02:51:02] <owh> I am trying to setup postfix to use gmail as the smtp smart host. Using this recipe: http://prantran.blogspot.com/2007/01/getting-postfix-to-work-on-ubuntu-with.html -- my mail comes in via fetchmail. Email arrives at my registered domain, is forwarded to gmail, then retrieved via fetchmail across a satellite link. If I follow the instructions, my email is attempting to be relayed to gmail, rather than delivered to my local mailbox. Where do I start looking to f
[02:51:29] *** MezzoTiNt has left #postfix
[02:51:44] <owh> My current setup uses a different smtp smarthost and works as expected.
[02:53:16] <owh> There are no bits that use either generic or transport. The only difference that I want to implement is secure authentication to gmail. But at present all I'm achieving is grey hair.
[02:54:04] * owh wonders out loud if I can just comment out the new references to generic and transport.
[02:55:18] <rob0> scoates: $ egrep '^[a-z]' /etc/postfix/master.cf | while read -a SERVICE ; do if [ "${SERVICE[4]}" = "n" ] ; then CHROOT="NOT" ; else unset CHROOT ; fi ; echo $SERVICE is $CHROOT chrooted ; done
[02:56:47] *** ziro has quit IRC
[02:56:57] <scoates> rob0: smtp-amavis is chrooted
[02:57:01] <scoates> is that it?
[02:57:33] <rob0> could be
[02:57:39] <scoates> hrm
[02:59:09] *** alecs has joined #postfix
[02:59:21] <alecs> what is the command for \postfix to run spamassassin filter?
[02:59:41] <scoates> rob0: how do I find out the chroot?
[03:00:38] *** Zand3r_ has joined #postfix
[03:00:45] <scoates> found it
[03:00:56] <scoates> but /var/spool/postfix/etc/hosts has localhost
[03:01:12] <scoates> anyway.. mail is coming through. I'll probably see an increase in spam, but that's better than no mail
[03:01:17] <scoates> (at least in theory)
[03:01:33] <rob0> and /var/spool/postfix/etc/resolv.conf is wrong
[03:01:52] <rob0> but if it's just the reinjection port it won't matter much
[03:01:59] <Zand3r_> Hi... I am using Postfix with MySQL and have relaised that my Virtual Aliases do not support catch-all aliases. My current configuration is query = SELECT goto FROM aliases WHERE local_part='%u' AND domain='%d'  - I am wondering if someone can suggest how I might expand this to support catch all addresses as I;pve tried a couple of variations that jsut do not work.
[03:02:02] <alecs> what is the command for postfix to run spamassassin filter?
[03:03:36] <scoates>  diff -u /var/spool/postfix/etc/resolv.conf /etc/resolv.conf   (no diffs)
[03:05:02] <owh> Whoot, for those watching, I've now managed to receive email by commenting out the transport reference.
[03:07:42] <scoates> anyway thanks
[03:09:28] <xpoint> Zand3r_, you should stop using catchall anyway
[03:09:47] *** Kalavera has joined #postfix
[03:09:51] <Kalavera> hi guys
[03:10:16] <Kalavera> i have a problem with a postfix server with mysql and SMTP with auth
[03:10:33] *** Tachy_ has joined #postfix
[03:10:43] <Kalavera> and an Outlook Express client over MAC OS 9
[03:11:12] <Kalavera> the pop3 client seems to be right validation over POP3
[03:11:43] <Kalavera> but when i try to send email using sames credencials, i receive an error message
[03:11:49] <Kalavera> Relay Access Denied
[03:11:59] <xpoint> good
[03:12:22] *** fujin has joined #postfix
[03:12:25] <xpoint> then smtp auth does not work, postfix does work, sure ?
[03:12:35] <Kalavera> my point is, i have "my smtp server requires authentication" checked
[03:12:36] <rob0> In your logs you'll see that the Outhouse Distress client did not attempt to AUTH.
[03:12:39] <xpoint> Kalavera, we need logs
[03:12:55] * rob0 has been there, done that
[03:12:56] <Kalavera> and the first option, "Use ..."
[03:12:58] <Zand3r_> xpoint: Thanks but on this occassion that doesn;t really help. I don;t actually use catch all myself but I have people on an old server that do that I am moving over.
[03:13:01] <Kalavera> xpoint, ok
[03:13:26] <Kalavera> take care the client is an outlook express over MAC OS 9 and not windows
[03:13:43] <rob0> It's still poorly written and unsupported.
[03:13:58] <xpoint> Kalavera, we take the basic problem, not mac problems :-)
[03:14:03] *** Roobarb has quit IRC
[03:14:13] *** knoba has quit IRC
[03:14:18] <rob0> Test AUTH with a real MUA, then try to make Outhouse work.
[03:14:23] *** Roobarb has joined #postfix
[03:14:24] *** knoba has joined #postfix
[03:15:03] <Kalavera> xpoint, http://www.pastebin.ca/480034
[03:16:04] <Kalavera> rob0, i have more less 40 domains with windows and MAC OS/X using mail, microsoft entourage, MS Outlook and outlook express working fine
[03:16:12] <Kalavera> but these is a special case
[03:16:13] <xpoint> Kalavera, and postconf -n to be usefull help
[03:16:19] <Kalavera> http://www.pastebin.ca/480034
[03:16:25] <Zand3r_> Is there a way for me to see what Postfix passes to MySQL if I am using %s and %d - My catch-alls aren;t working
[03:16:39] <xpoint> Zand3r_, mysql logs
[03:16:59] <rob0> Zand3r_: and postmap(1) with -q option
[03:17:38] <Kalavera> xpoint, http://www.pastebin.ca/480037
[03:18:51] <Kalavera> xpoint, i want to know if these problem is specific for these version of MAC OS
[03:19:13] <Kalavera> and has relation with the MUA version
[03:19:20] *** fred87_ has joined #postfix
[03:20:54] <xpoint> Kalavera, i dont have a mac to test it with, but you can test that "telnet localhost 25" says plain or login, this should be it if tls is working between postfix and clients
[03:21:30] <xpoint> Kalavera, yes tls is needed in you postfix
[03:22:53] <xpoint> Kalavera, and the client will hopefully send clear passwords
[03:23:05] *** Tachy has quit IRC
[03:24:20] <xpoint> Kalavera, to be sure postfix is not reject smtp auth in fail, you can add permit_sasl_auth.. to all smtpd_*_stages
[03:28:59] <Zand3r_> rob0: Sorry to be a pain but the output from postmap with -q summarises whats in my configuaration. Is there a way to get examples of what will actually get passed to MySQL? I have SQL working fine but the SQL I am using does not seem to allow Catch All aliases
[03:31:45] <Kalavera> xpoint, i don't have problems with others clients
[03:31:52] <Kalavera> only with these one
[03:32:00] <alecs> if i want to have an automaticaly CC from all users (not catch all) for example i have 2 mail addresses info and sales .... all what sales receives then info also receive same mails ... how can i do that ?
[03:32:22] <alecs> *** and sales tro be able reading his own mail
[03:33:25] *** Zand3r__ has joined #postfix
[03:33:26] <Kalavera> xpoint, server works with others MUAs including MAC OS/X clients
[03:33:45] <Kalavera> i don't understand what is wrong with these one
[03:34:19] *** Zand3r_ has quit IRC
[03:34:55] *** jpon has quit IRC
[03:35:08] <alecs> also.... how do i move the SMTP (sendmail) out from 25
[03:35:09] <alecs> ?
[03:35:47] <xpoint> Kalavera, enable -v on smtpd listner in master.cf and stop and start postfix again, then try again from mac os 9
[03:36:15] <xpoint> Kalavera, this should get you more verbose logs of the problem
[03:36:34] <Kalavera> xpoint, ok thanks for your advice
[03:36:45] *** magyar has joined #postfix
[03:37:22] <xpoint> Kalavera, is ssl / tls working from the mac os 9 ?
[03:37:52] <xpoint> Kalavera, maybe os9 does not support tls or sslv3 ?
[03:38:01] <Kalavera> no, and the same configuration is for the other clients without problems
[03:38:25] <xpoint> so it works for some os9 user ?
[03:40:47] *** alecs has quit IRC
[03:42:16] <Kalavera> no
[03:42:30] <Kalavera> xpoint, it works for all MAC OS/X users
[03:42:37] <Kalavera> OS/X is over 10
[03:42:40] <Kalavera> i think so
[03:43:12] <Kalavera> in OS/X i work with microsoft entourage and works fine
[03:43:23] <Kalavera> another client is mail and works fine too
[03:43:51] <Kalavera> the only problem is with these OS 9 and outlook express for MAC
[03:44:10] <Kalavera> i have active broken_clients
[03:51:54] <xpoint> why do mac users need to use microsoft is another good question, but this is not the solution, i will give up here since i dont have mac os here to test it with, but i belive problem is related to ssl/tls
[03:53:15] <xpoint> but accourding to you postconf -n you don't use ssl/tls where it works, i will not recomend this since passwords are exchange in clear text
[03:55:46] *** JasonF has left #postfix
[03:56:08] <xpoint> Kalavera, is outlook the latest for mac os ?
[03:56:30] <Kalavera> xpoint, nop. is an older version
[03:56:31] <xpoint> Kalavera, could be this problem is fixed
[03:56:56] <Kalavera> the question is how?
[03:57:01] <xpoint> Kalavera, imposibble to upgrade ?
[03:57:46] <Kalavera> i think nop, i suggest the client do an upgrade for OS and change your MUA for Microsoft entourage or mail
[03:58:12] <xpoint> Kalavera, yes if this is the only option
[03:58:16] <Kalavera> but at these moment these upgrade is more expensive for her
[03:58:41] <Kalavera> the other solution is using webmail
[03:58:49] <Kalavera> but she has another new problem
[03:58:59] <Kalavera> she can not send attachments properly
[03:59:00] <Kalavera> :s
[03:59:23] <xpoint> Kalavera, yes licenses is pain in the ass mostly for secure ones computer
[04:00:19] <xpoint> Kalavera, ftp should olso be used anyway for files over 8M
[04:02:06] <xpoint> but if less then 8M files give problems this is uncomming
[04:08:33] <Kalavera> is a lot less
[04:08:34] <Kalavera> :D
[04:08:41] <Kalavera> only 500 k
[04:10:54] <xpoint> hmm this is a irretating problem, even
[04:14:02] *** Roobarb has quit IRC
[04:14:21] *** Roobarb has joined #postfix
[04:14:27] <Kalavera> sure
[04:17:19] *** rmayorga has joined #postfix
[04:20:19] *** doomas has joined #postfix
[04:22:39] *** alanhaggai has quit IRC
[04:23:55] *** owh has left #postfix
[04:25:44] *** doomas_ has quit IRC
[04:26:45] *** JMartinC has joined #postfix
[04:29:50] <JMartinC> hi guys, I've migrate an old e-smith server to Gentoo/postfix, the sys admins usually add an user and add it to a group, somehow they were able to send e-mail to that group and it's redirected to each user on the group.
[04:30:32] <JMartinC> I think that's not possible anymore, so I make the use alias database but it's harder to config than just adding an user to a group, we are talking a lot of users, (1,000)
[04:31:09] <JMartinC> is there a way to send e-mail to a linux group and make postfix deliver to each member on the group? it sound logical, don't now why it doen's work.
[04:32:53] *** scoates has left #postfix
[04:33:02] <Kalavera> JMartinC, do you speak spanish?
[04:33:15] <JMartinC> yep
[04:33:25] <Kalavera> ok explicamelo en  privado
[04:33:28] <Kalavera> uso gentoo
[04:33:29] <Kalavera> uso postfix
[04:33:59] <JMartinC> tambien tengo postfix y gentoo.
[04:34:41] <Kalavera> ok pero lo decia para no cansar al resto con el espa?ol
[04:34:43] <Kalavera> XD
[04:35:11] <Kalavera> usas mysql para el backend o pam
[04:40:01] *** Zand3r__ has quit IRC
[04:57:29] *** Dominian has joined #postfix
[04:58:10] *** Nockian has quit IRC
[05:10:05] *** rmayorga has quit IRC
[05:14:02] *** Roobarb has quit IRC
[05:14:23] *** Roobarb has joined #postfix
[05:16:27] *** yam has quit IRC
[05:24:36] *** rmayorga has joined #postfix
[05:40:24] *** amrit|wrk is now known as amrit
[05:41:59] *** MarcoH has quit IRC
[05:50:53] *** fujin has quit IRC
[06:08:51] *** Kalavera has quit IRC
[06:09:15] *** cilly has quit IRC
[06:09:45] *** cilly has joined #postfix
[06:12:06] *** vn has joined #postfix
[06:14:02] *** Roobarb has quit IRC
[06:14:13] *** GMFlash has quit IRC
[06:14:21] *** GMFlash has joined #postfix
[06:14:22] *** Roobarb has joined #postfix
[06:20:23] *** vnm has quit IRC
[06:20:39] *** JMartinC has quit IRC
[06:34:13] <f3ew> !cheatsheet
[06:34:14] <knoba> f3ew: 'cheatsheet' : http://jimsun.linxnet.com/misc/postfix-anti-UCE.txt : A HOWTO for pre-DATA spam control.
[06:41:51] *** hparker has quit IRC
[06:46:13] *** magyar has quit IRC
[06:52:21] *** memetic has quit IRC
[06:52:28] *** memetic has joined #postfix
[06:54:27] *** fujin has joined #postfix
[06:55:39] <fujin> yo, I've got a box setup as sending through a smar host
[06:55:47] <fujin> and I want to bypass the smarthost for a particular outgoing mail domain
[06:55:52] <fujin> I assume that this can be done with a transport
[06:56:05] <fujin> but, if I do it with a transport and the endpoint mailserver's address changes, how can I get around this
[06:56:22] <fujin> in theory, I want to be able to transport to an mx record
[06:56:25] <fujin> is this possible
[06:59:48] <fujin> and if so, how? ;
[06:59:56] <fujin> I can't work out how to do it from the docs
[07:00:58] *** KhensU has quit IRC
[07:08:00] *** xpoint has quit IRC
[07:11:48] *** ivx has joined #postfix
[07:12:38] <fujin> nevermind,foudn it
[07:13:38] <ivx> hello, I set up a server based on a tutorial and it works great. I can send mail just fine and get it fine except to one person. When i send mail to that person i get an message back that says: host mail.bluetie.com[206.65.164.155] said: 450 4.7.1   Client host rejected: cannot find your reverse hostname, [64.27.13.21] (in reply to RCPT TO command). I get mail from her just fine. does anyone know what that means
[07:14:02] *** Roobarb has quit IRC
[07:14:28] *** Roobarb has joined #postfix
[07:15:43] *** fujin has quit IRC
[07:21:38] *** KhensU has joined #postfix
[07:24:18] <rworkman> ivx: it means exactly what it says.  The other person's mail server is rejecting mail from you because it can't find a hostname for your ip.
[07:24:58] <ivx> rworkman, so how do i fix this?
[07:25:24] <rworkman> You need a reverse lookup record - DNS.
[07:26:55] <rob0> Call your ISP.
[07:27:11] <rob0> ... or ...
[07:27:16] <rob0> !relayhost
[07:27:16] <knoba> rob0: 'relayhost' : a configuration parameter in the main.cf: The default host to send non-local mail to when no entry is matched in the optional transport(5) table. When no relayhost is given, mail is routed directly to the destination.
[07:27:24] <rob0> relay through the ISP's server
[07:31:56] <ivx> okay thanks ya'll
[07:33:34] *** rmayorga has quit IRC
[07:36:50] *** memetic has quit IRC
[07:40:34] *** [tor] has joined #postfix
[07:43:41] *** Motoko-chan has joined #postfix
[07:48:23] *** TheOutlander has joined #postfix
[08:01:43] *** memetic has joined #postfix
[08:05:06] *** cilly has quit IRC
[08:09:59] *** neko_ has quit IRC
[08:10:05] *** neko has joined #postfix
[08:10:53] *** cilly has joined #postfix
[08:14:02] *** Roobarb has quit IRC
[08:14:22] *** Roobarb has joined #postfix
[08:30:38] *** foo has joined #postfix
[08:30:52] <foo> postfix/smtp[4633]: warning: no MX host for YOURSERVER.com has a valid address record - if the server only sends out mail and doesn't receive, it doesn't need a valid mx record, does it?
[08:31:16] <Motoko-chan> Correct.
[08:31:34] <foo> Ok, so I can ignore that error, I'm assuminmg
[08:31:37] <Motoko-chan> But there should be an MX for the domain you are sending under, even if it isn't that server itself.
[08:33:18] <foo> Hm, but that domain only sends mail out ... doesn't receive... still needs mx?
[08:33:24] *** af_ has joined #postfix
[08:34:41] <Motoko-chan> It really should since you need to get bounce notifications and such.
[08:34:49] <foo> ohh, I see
[08:34:52] <foo> True, ok
[08:34:53] <foo> postfix/smtp[4635]: 415372D3C335: to=<www-data at www dot domain.com>, relay=none, delay=0.36, delays=0.01/0/0.35/0, dsn=5.4.6, status=bounced (mail for www.domain.com loops back to myself)
[08:34:58] <foo> hmm. I've seen that before
[08:35:01] * foo checks some past notes
[08:38:58] <foo> hm, that was sendmail
[08:42:18] *** theSefirah has joined #postfix
[08:42:54] <theSefirah> hi, is postfixadmin suppose to create maildirs on the harddrive ?
[08:43:04] <theSefirah> meaning: there are no actual dirs created
[08:43:14] <f3ew> !loop
[08:43:15] <knoba> f3ew: Error: "loop" is not a valid command.
[08:43:17] <f3ew> !loopback
[08:43:18] <knoba> f3ew: 'loopback' : 'Mail loops back to myself' means that your Postfix wanted to send out the mail to the internet but then discovered that the DNS says your mail server should be responsible. Most likely you forgot to list your domain in mydestination or virtual_(alias|mailbox)_domains
[08:43:34] <f3ew> theSefirah, send the account an email to create the maildir
[08:43:47] *** sn00p- has joined #postfix
[08:44:12] <theSefirah> just a mom
[08:46:51] <theSefirah> heh, it did something, though not exactly as I expected, thanks :)
[08:46:53] <foo> f3ew: thanks
[08:51:08] <theSefirah> hmm, from config.inc in postfix admin am I to understand that my mailboxes will be in the form /vmail/domain.tld/userjohn ?
[08:57:56] *** atilaX has quit IRC
[08:59:54] *** ivx has quit IRC
[09:01:58] *** rootsvr has joined #postfix
[09:04:01] *** Turt|e has joined #postfix
[09:08:32] *** Lap_64 has joined #postfix
[09:09:31] *** Motoko-chan has quit IRC
[09:13:15] *** [miles] has joined #postfix
[09:14:02] *** Roobarb has quit IRC
[09:14:10] *** pmjdebruijn has joined #postfix
[09:14:22] *** Roobarb has joined #postfix
[09:23:03] * sid3windr wonders why Motoko-chan said there should be an MX record, it's not even required to receive mail at all
[09:23:41] *** cilly has quit IRC
[09:27:35] *** theSefirah has quit IRC
[09:45:00] *** aukjan|gone is now known as aukjan
[09:47:37] *** aukjan is now known as aukjan|gone
[09:50:14] *** Mez has joined #postfix
[09:53:39] *** war has joined #postfix
[09:54:17] *** frennkie has joined #postfix
[09:58:47] *** Mez is now known as Mez|OnAir
[10:02:21] *** cutmasta has joined #postfix
[10:06:04] *** Lap_64 has quit IRC
[10:06:41] *** cilly has joined #postfix
[10:08:27] *** Lap_64 has joined #postfix
[10:14:02] *** Roobarb has quit IRC
[10:14:21] *** Roobarb has joined #postfix
[10:16:24] *** amrit is now known as amrit|zzz
[10:16:36] *** mazon is now known as Mazon
[10:18:17] *** taube is now known as Taube
[10:20:07] *** grte has joined #postfix
[10:20:40] <grte> I don't suppose anyone knows where I could find Thawte_Premium_Server_CA.pem?
[10:20:47] <grte> Sadly, my distro doesn't seem to have it by default.
[10:25:20] *** frennkie has quit IRC
[10:28:50] *** grte has left #postfix
[10:30:47] *** skor has joined #postfix
[10:31:22] *** skor is now known as stellina
[10:44:36] *** mastachand has joined #postfix
[10:53:20] *** untitled has joined #postfix
[10:54:10] <untitled> hi, i am configuring postfix with smtp and can't see anything when typing #telnet localhost 25 and then ehlo localhost
[10:54:37] <untitled> what can be the problem?
[11:04:38] *** af_ has quit IRC
[11:06:14] *** raqamy has quit IRC
[11:07:32] <R1ck> untitled: look in your /var/log/maillog
[11:07:48] *** Sp4rKy has left #postfix
[11:07:54] <untitled> oh yes, i did)
[11:08:00] <untitled> i found it already)
[11:08:39] *** Zeit|awy has joined #postfix
[11:08:41] * untitled hates typo
[11:08:51] <untitled> *misstyping)
[11:10:31] *** cilly has quit IRC
[11:10:40] <untitled> how can i check if imap works correctly?
[11:14:02] *** Roobarb has quit IRC
[11:14:22] *** Roobarb has joined #postfix
[11:15:42] *** Zeit|idle has quit IRC
[11:16:26] *** Roobarb has quit IRC
[11:16:46] *** Roobarb has joined #postfix
[11:18:10] *** alecs has joined #postfix
[11:18:14] <alecs> hi there
[11:19:00] <alecs> i'm using postfix to send mails, but i don't know what to use to receive ...
[11:19:47] <Roobarb> Courier-imap/pop, dovecot, squirrelmail....
[11:19:52] <Roobarb> pick one
[11:19:55] <untitled> how can i check if imap works correctly?
[11:20:05] <alecs> any hints? also how to configure for multi domains ?
[11:20:19] <Roobarb> untitled: with an IMAP client, but postfix is not an IMAP server
[11:20:40] <alecs> untitled: use exim or courier
[11:20:40] <untitled> Roobarb: name one imap client
[11:20:48] <untitled> i am using courier
[11:20:54] <alecs> *** sorry those are servers
[11:20:57] *** rootsvr has quit IRC
[11:21:05] <untitled> just don't know how to ensure that it works
[11:21:08] <Roobarb> untitled: Mozilla Thunderbird, Microsoft Outlook, Kmail
[11:21:09] <alecs> an imap client is mozilla thunderbird
[11:22:11] <untitled> ok
[11:25:48] *** keanne has quit IRC
[11:28:00] <alecs> any ideeas where i can find a good blacklist database? how to use it?
[11:31:09] *** jduggan__ has joined #postfix
[11:33:32] <R1ck> alecs: www.spamhaus.org
[11:33:45] <jduggan__> hey all, ive received an abuse email, im having difficulties understanding the headers, can someone read them and see if they can make sense for me?
[11:33:55] <jduggan__> http://shell.noffle.net/~jduggan/abuse.txt
[11:34:33] <jduggan__> its 83.100.191.243 which is my clients IP, but im having difficulties figuring out if this is the actual source
[11:45:32] <untitled> can't i check imap working without clients? :(
[11:45:57] <R1ck> untitled: you can telnet to the server on port 143
[11:46:24] <untitled> ok
[11:46:33] <untitled> it said ok)
[11:46:46] <untitled> do i have to ehlo or something?
[11:47:52] *** ikaro^ has joined #postfix
[11:48:40] <R1ck> read the IMAP rfc
[11:49:18] *** ikaro has quit IRC
[11:52:51] <f3ew> . login username password
[11:52:53] <f3ew> . logout
[11:53:12] <alecs> any free blacklists?
[11:53:30] <f3ew> !cheatsheet
[11:53:31] <knoba> f3ew: 'cheatsheet' : http://jimsun.linxnet.com/misc/postfix-anti-UCE.txt : A HOWTO for pre-DATA spam control.
[11:54:01] * f3ew reads jduggan's message'
[11:55:20] <f3ew> jduggan who sent you that message?
[11:58:37] *** cilly has joined #postfix
[12:00:10] *** rootsvr has joined #postfix
[12:03:11] *** rootsvr has quit IRC
[12:08:02] *** nescius has joined #postfix
[12:10:56] *** havoc has joined #postfix
[12:10:59] <havoc> morning
[12:11:10] <havoc> what starts "postdrop"?
[12:11:37] <havoc> I keep getting errors like:  postfix/postdrop[15772]: warning: mail_queue_enter: create file maildrop/815243.15772: Permission denied
[12:12:04] <havoc> apparently /usr/sbin/postdrop -r is running as www-data, WTF?
[12:13:07] *** alecs has left #postfix
[12:14:02] *** Roobarb has quit IRC
[12:14:22] *** Roobarb has joined #postfix
[12:16:14] * havoc waits
[12:18:40] <f3ew> sendmail(8)
[12:18:45] <f3ew> errr, sendmail(1)
[12:18:47] <f3ew> not 8
[12:18:56] *** TheOutlander has quit IRC
[12:20:27] * havoc mans
[12:20:58] <havoc> f3ew: the question is "why is it running as www-data?"
[12:21:09] <havoc> f3ew: and can I change the user it's running as?
[12:21:37] <f3ew> Apache?
[12:21:45] <havoc> yes, it's running as apache
[12:21:46] <f3ew> A form of some sort sending mail
[12:22:06] <havoc> ah, that would make sense, so it's group perms that are messed up
[12:22:27] <f3ew> probably
[12:22:34] <f3ew> or authorized_submit_users
[12:24:42] <havoc> no, that's still static:anyone
[12:26:28] <havoc> drwx-wx--T  2 postfix postdrop 4096 2007-05-10 04:35 /var/spool/postfix/maildrop/
[12:26:41] <havoc> it *looks* right to me, but apparently not :(
[12:27:42] *** bostik has joined #postfix
[12:28:43] <havoc> f3ew: any ideas?
[12:28:53] <havoc> and/or suggestions?
[12:29:29] <f3ew> postfix check says what?
[12:29:32] <havoc> all perms/uids/gids should be the same as the old system I moved everything from yesterday
[12:29:53] <havoc> $postfix check
[12:29:53] <havoc> postfix/postfix-script: warning: not owned by group postdrop: /usr/sbin/postqueue
[12:29:57] <havoc> postfix/postfix-script: warning: not owned by group postdrop: /usr/sbin/postdrop
[12:30:00] <havoc> :)
[12:30:01] * havoc fixes that
[12:30:10] <f3ew> Heh
[12:30:30] <havoc> hmm, but the perms are the same as the old system
[12:30:50] <havoc> ah, but new version of postfix on this system
[12:30:57] <f3ew> _ownership_ <> permissions
[12:31:08] <havoc> yes, that is the same too
[12:31:20] <havoc> I'm using the term "perms" very loosely
[12:31:36] <havoc> 'perms' == owner/group/attributes
[12:33:03] <f3ew> postfix set-permissions
[12:34:28] <havoc> $chgrp /usr/sbin/postdrop /usr/sbin/postqueue
[12:34:28] <havoc> chgrp: invalid group `/usr/sbin/postdrop'
[12:34:30] <havoc> ack
[12:34:40] <havoc> yet it's in /etc/group, WTF?
[12:36:07] *** noetik has joined #postfix
[12:36:38] * havoc smacks self
[12:39:34] * havoc will [try to] always remember to run postfix check from now on ;)
[12:40:06] <havoc> f3ew: thanks :)
[12:41:39] <shasta> you should also read chgrp manpage ;)
[12:42:17] <havoc> shasta: no, I should read what I typed ;)
[12:42:45] <havoc> it's been a *long* 36+ hours or so :(
[12:46:38] *** cilly has quit IRC
[12:46:48] *** kjalil has left #postfix
[12:53:16] *** havoc has left #postfix
[13:01:58] *** UQlev has joined #postfix
[13:13:21] *** UQlev has quit IRC
[13:14:03] *** Roobarb has quit IRC
[13:14:23] *** Roobarb has joined #postfix
[13:43:41] *** mastachand has quit IRC
[13:55:40] *** master_of_master has joined #postfix
[13:58:12] *** mastachand has joined #postfix
[13:59:49] *** memic has quit IRC
[14:00:29] *** memic has joined #postfix
[14:06:26] *** raqamy has joined #postfix
[14:06:59] *** master_o1_master has quit IRC
[14:08:09] *** ikaro has joined #postfix
[14:10:47] *** ikaro^ has quit IRC
[14:11:30] *** noetik has quit IRC
[14:12:33] *** sepski has joined #postfix
[14:14:02] *** Roobarb has quit IRC
[14:14:22] *** Roobarb has joined #postfix
[14:18:27] *** inflex has joined #postfix
[14:18:59] <inflex> hi there everyone - I've got a strange problem where when trying to deliver data via a high port (ie, 10025) Postfix will seem to get upset and I get a "connection reset by peer"
[14:19:28] <inflex> but I can't seem to nail down exactly -why- it's getting upset
[14:20:19] <f3ew> inflex, tcpdump is your friend
[14:20:25] <f3ew> or turn verbose logging n
[14:20:26] <inflex> joy o joy
[14:20:27] <f3ew> on*
[14:20:59] <dman_> warning: SASL authentication failure: cannot connect to saslauthd server: Permission denied
[14:21:10] <dman_> what could be causing that error
[14:21:18] <dman_> my sasl server works
[14:21:23] <inflex> f3ew: *sigh* this could be fun
[14:21:52] <f3ew> dman_ permissions on the parent directory
[14:21:55] <dman_> i 'n'd all the chroot options in the postfix config
[14:22:05] <dman_> parent directory of what?
[14:22:26] <dman_> i tested sasl out with testsaslauthd or whatever too, and it returned OK
[14:22:34] <f3ew> the saslauthd socket
[14:22:40] <f3ew> test as a nonn root user
[14:22:43] <f3ew> non
[14:23:50] *** reber has joined #postfix
[14:23:52] <reber> hi all
[14:24:03] <dman_> yeah, failing as non root
[14:24:06] <reber> anyone using nginx proxy here ?
[14:24:34] <Euforia_> Hi. My amavisd does recognize spam but does not add the ***SPAM*** to the subject. Anybody any thoughts?
[14:25:12] <dman_> f3ew: so how do i make it so postfix user can access sasl?
[14:26:32] <dman_> ah, theres a sasl group
[14:26:55] *** frennkie has joined #postfix
[14:27:51] <untitled> is quota built in postfix by default?
[14:28:33] <R1ck> no
[14:30:18] *** frennkie has quit IRC
[14:32:08] *** Nockian has joined #postfix
[14:38:34] *** f3ew has quit IRC
[14:39:01] *** f3ew has joined #postfix
[14:39:13] *** noetik has joined #postfix
[14:40:56] *** hparker has joined #postfix
[14:48:50] *** reber has left #postfix
[15:01:20] *** untitled has quit IRC
[15:12:58] *** dman_ has left #postfix
[15:13:05] *** Ryushin has joined #postfix
[15:14:02] *** Roobarb has quit IRC
[15:14:24] *** Roobarb has joined #postfix
[15:17:24] *** Mez|OnAir is now known as Mez
[15:23:54] <Rattail> Goodmorning All, I'm sorry I'm back asking this same question
[15:24:09] <Rattail> I'm using amavis to scan email on a postfix server with clamav, it is slowing my email down "very very slow" emails are taking over a hour just on the server it's self
[15:24:26] <Rattail> anybody got any clues as to what I can do ????   Thanks
[15:24:35] <hparker> Is it swapping?
[15:32:01] *** inflex has left #postfix
[15:32:33] <rob0> You might consider taking that to the amavisd-new (or clamav) mailing list (depending where the bottleneck appears to be.)
[15:34:29] <Rattail>  ClamAV-clamd av-scanner FAILED: Too many retries to talk to /var/run/clamav/clamd.sock (Can't connect to UNIX socket /var/run/clamav/clamd.sock: No such file or directory)
[15:34:44] <f3ew> there yuo go
[15:34:48] <f3ew> you*
[15:35:13] <hparker> That might slow it down a bit, reverts to clamscan then
[15:36:27] <Rattail> okay so I touched the file clamd.sock and changed the own to clamav.amavis
[15:36:36] <Nockian> Rattail: i told you this yesterday. check to make sure that you have permissions set properly for amavisd and clamav.
[15:36:37] <rob0> no, start clamd
[15:37:07] <rob0> and yes, amavisd needs permission to the socket.
[15:37:15] <Rattail> Nockian, I uninstalled and reinstalled clamav, clamd and -db
[15:37:24] <Rattail> now I'm going over everything again
[15:37:43] <Nockian> Rattail: yes, and a lot of distro's don't work out of the box with their default permissions. you have to manually set the permissions so that amavisd has permission to clamav and such.
[15:37:55] <rob0> Simplest way to handle permissions is to have amavisd and clamd running as the same user.
[15:38:49] * rob0 doesn't bother with clamav: nothing gets through to it. Shoot the fish in the barrel before you cast a line.
[15:38:59] <Rattail> Permission denied, should I just open the permissions up on clamd.sock
[15:40:14] <Rattail> rob0, I'm not sure what you mean
[15:41:19] <Rattail> Nockian, I have made clamav a member of amavis and amavis a memeber of clamav group
[15:43:47] *** Taube is now known as taube
[15:44:34] *** taube is now known as Taube
[15:45:40] <Rattail> rob0, are you telling me to turn clamav off ?
[15:46:19] <Rattail> I have given all the permissions I know to both clamav and amavisd on the file clamd.sock
[15:52:58] <rob0> !cheatsheet
[15:52:59] <knoba> rob0: 'cheatsheet' : http://jimsun.linxnet.com/misc/postfix-anti-UCE.txt : A HOWTO for pre-DATA spam control.
[15:53:17] <rob0> Viral spew is usually blocked by the HELO checks ^^ therein.
[15:54:30] *** stellina has quit IRC
[15:55:06] <rob0> "clamd.sock" is not (should not be) a regular file. It is (should be) a Unix socket. It is created when clamd starts. Permissions are set in the clamd config, but also in the directories leading to it.
[16:03:24] *** [tor] has quit IRC
[16:05:12] <rob0> Um, should you turn off clamav? If you have it running already, it takes care of itself, so no, leave it on. But if you're having a lot of trouble with it, !cheatsheet is going to do a better job of spam/virus protection anyway.
[16:07:35] <rob0> I had 2 sites with clamav running (lost one of them). Over a period of several months, between the two, it caught maybe 2-3 phish. SpamAssassin was/is running behind clamd in amavisd-new, and it probably would have caught those.
[16:09:35] *** fzzzt has joined #postfix
[16:09:56] <Rattail> rob0, I did use the cheatsheet to setup my system years back, things got crazy and the company hired a MS guy who has pushed exchange into the office, With get this!!! the same name as our postfix email server
[16:10:15] <Rattail> He is a stupid MS guy
[16:11:23] <Rattail> So I can't bounce the domain name if it comes from other systems thank to MS guy
[16:11:27] <Rattail> Stupid
[16:12:23] <Rattail> anyway I have been using clamav until this last upgrade with no problems "now for some unknown reason" ClamAV-clamd: Can't connect to UNIX socket /var/clamav/clamd.sock
[16:13:03] <sysmonk> Rattail: but you can change the name of your system, can't you ? :)
[16:14:03] *** Roobarb has quit IRC
[16:14:22] *** Roobarb has joined #postfix
[16:14:51] <rob0> I don't understand the "can't bounce the domain name if it comes from other systems" part. You mean in HELO? Sure you can. Do a check_client_access lookup to permit the HELO from MSexChange. Or better and simpler, have MSexChange in $mynetworks and use permit_mynetworks before the HELO checks.
[16:16:05] *** nescius has quit IRC
[16:17:05] <rob0> But at some point you might want to take a stand and tell the company to choose one path or the other. I just got dismissed from an IT task for my employer; they went with stupid MS junk. So I step out of the way and don't offer to do IT work now.
[16:18:19] *** Mazon is now known as mazon
[16:18:21] <rob0> IT service is like a religion to the end user who knows nothing about it. And MS knows exactly how to target those people and exploit their ignorance.
[16:18:43] <Dominian> amen brother
[16:18:59] <rob0> They lost faith in me ... fine. They will deal with the consequences.
[16:19:15] <Rattail> I am having the same problems
[16:19:40] <rob0> I will definitely quit if I am asked to fix their MS mess, unless they let me do it the right way (clear out the MS junk.)
[16:19:50] <Dominian> heh
[16:19:52] *** rmayorga has joined #postfix
[16:20:06] <rob0> in the meantime, I'm doing non-IT-related work.
[16:21:45] <Rattail> they have moved down to troubleshooting endusers XP boxes (unless the postfix server starts having problems that is)
[16:22:15] *** nescius has joined #postfix
[16:22:47] <Rattail> and the only trouble I have is end users getting virus (so I update clamav)
[16:25:14] <Rattail> I don't understand why I had to touch the clamd.sock file anyway ? shouldn't that have been created by amavis or clamav?
[16:26:18] *** d0uglas has quit IRC
[16:29:54] *** cutmasta is now known as oxzibit
[16:30:06] *** oxzibit is now known as oXzibit
[16:30:09] <Dominian> Rattail: clamd should've created it
[16:30:30] <Rattail> that is what I thought
[16:30:46] <Rattail> is clamd being lazy
[16:30:49] <Rattail> ???
[16:31:16] <Dominian> hehe
[16:31:45] <Dominian> was that rhetorical?
[16:32:36] <Rattail> just did a find on clamd.sock and it only shows the one file I created
[16:32:57] <Dominian> There should only be *one* file anyway
[16:33:13] <rob0> Rattail: if you create it, it is *not* a socket.
[16:33:14] <Rattail> but it shouldn't matter if I created it and pointed clamd.conf and amavisd.conf at it I should be gold right Jerry
[16:33:21] <Rattail> OH
[16:33:24] <Dominian> no
[16:33:27] <Rattail> rob0
[16:33:30] <Rattail> OH
[16:33:33] <Dominian> that's a special "device" of sorts.
[16:33:40] <Rattail> OH
[16:33:41] <Dominian> doing touch clamd.sock will make clamd go "wtf?!?"
[16:34:00] <Rattail> I'm going wtf right now
[16:34:05] <rob0> and clamd is probably logging all of this WTFage.
[16:34:06] <Rattail> didn't know that
[16:34:18] <Rattail> sir yes sir
[16:34:23] <Dominian> clamd is not that hard...
[16:34:33] <Rattail> but I'm hard headed
[16:34:37] <Dominian> although I do need to right an rc.clamd file though
[16:34:45] <Dominian> s/write/right
[16:35:00] <Dominian> er.. s/right/write
[16:35:09] <rob0> Again, the simple thing to do with amavisd-new is to run both as the same user. Adjust permissions and ownerships accordingly.
[16:38:05] <Rattail> rob0, I run both as amavis and I change permissions to clamav.amavis on files
[16:40:15] <rob0> Can the amavis user write in /var/clamav ? If not, guess what you get for a socket?
[16:40:16] <Rattail> the problem is fixed
[16:41:36] <Rattail> I went back to clamd.conf and looked it over; I had made the change to LocalSocket /var/clamav/clamd.sock
[16:41:57] <Rattail> but when I installed clamav it was set to LocalSocket LocalSocket /tmp/clamd.sock
[16:42:17] <Rattail> so I just set it back restarted both clamav and amavis
[16:42:23] *** nescius has quit IRC
[16:42:27] <Dominian> rob0: I may just do the switch now.. we'll see..
[16:42:30] <Rattail> went and looked in tmp and guess what
[16:42:38] <Dominian> rob0: I'll have to rework my tutorial lol
[16:42:45] <Rattail> clamd.sock was there
[16:42:57] <Rattail> HIPP HIPP HURRRAAAAA
[16:43:29] <rob0> Dominian: When I set up amavisd-new, I named the reinjection service "reinjection" and made that a symlink to smtp(8) in /usr/libexec/postfix/ .
[16:43:46] <rob0> that is undocumented and likely to break without notice BTW :)
[16:44:00] <rob0> BUT ... it really helps to make logs more readable.
[16:44:26] <Dominian> yeah
[16:44:29] <Dominian> that's a good idea
[16:44:32] <rob0> Reinjections are logged as "postfix/reinjection" and not "postfix/smtp".
[16:44:41] <Dominian> nice
[16:44:46] <Dominian> that would make things easier
[16:44:53] <rob0> (Maybe a name like "scanned" would be clearer.)
[16:45:14] <rob0> If you do that be VERY CAREFUL when upgrading Postfix.
[16:45:35] <Dominian> yeah, cause it would wipe out the symlink
[16:45:43] <Dominian> whichi s why I'll document everything :)
[16:45:50] <Dominian> and add it to my tutorial
[16:45:54] <rob0> s/would/might/ ... last time I don't think it did.
[16:46:09] * Dominian nods
[16:47:17] <rob0> I do the same symlink trickery with smtpd(8)'s listening on smtps and submission ports.
[16:47:25] *** Mez has quit IRC
[16:47:47] <rob0> When someone hits 465 or 587, the name in syslog tells me so.
[16:48:00] * Dominian nods
[16:48:45] <Dominian> Going through the install now
[16:49:16] <rob0> Um ... I might have confused a detail there; the reinject is a symlink to smtpd(8). I think multiple names for smtp(8) are already broken because of lmtp(8).
[16:49:42] <Dominian> k
[16:50:24] <rob0> (smtp and lmtp are the same binary, it acts differently depending how called.)
[16:52:32] *** xpoint has joined #postfix
[16:52:50] *** nescius has joined #postfix
[16:53:30] *** nescius has quit IRC
[16:58:09] *** ziro has joined #postfix
[17:04:25] *** noetik has quit IRC
[17:04:43] <Dominian> rob0: I'll emss with this more when I have time to concentrate on it
[17:04:55] <Dominian> looks like a lot of options I need to read up on that I"m not quite sure about
[17:05:45] <rob0> The hardest part is to wrap your head around how it works. :)
[17:06:02] <Dominian> yes..
[17:06:10] <Dominian> I know how mailscanner works.. but amavisd-new is a bit different
[17:07:05] <rob0> spew+mail from Internet -> amavisd (loses/redirects spew) -> mail back into Postfix -> delivery agent
[17:09:44] * Dominian nods
[17:09:47] <Dominian> about the same way then
[17:09:56] <Dominian> er.. no
[17:10:12] <Dominian> postfix with mailscanner grabs the email.. puts it into a queue that mailscanner grabs it.. scans it.. then reinjects it
[17:10:21] <Dominian> so with amavisd-new.. it intercepts the email right away?
[17:13:11] *** pmjdebruijn has quit IRC
[17:14:01] *** Roobarb has quit IRC
[17:14:23] *** Roobarb has joined #postfix
[17:14:55] *** Roobarb has quit IRC
[17:15:16] <rob0> content_filter says that everything has to go through that filter as a relay of sorts.
[17:15:26] *** Roobarb has joined #postfix
[17:18:55] *** Lap_64 has quit IRC
[17:21:28] *** Blackvel has joined #postfix
[17:22:43] *** mazon is now known as Mazon
[17:22:58] <Blackvel> Hi all. Can I combine reject_rbl_client in smtpd_client_restrictions with any caching so subsequent rbl checks won't hit the list server?
[17:24:54] <Dominian> rob0: which is fine
[17:25:04] <Dominian> rob0: doesn't look all that hard.. I just need time I can dedicate to it.
[17:25:20] <rob0> yes, you might not get it all in one day
[17:25:55] <rob0> also consider using access(5) rules to FILTER target, if you don't want the filter to do all mail.
[17:26:21] <rob0> the problem with that is that receive_override_options go in the wrong place.
[17:26:44] <rob0> but that's a minor issue. That's how I did it at one of my 2 amavisd-new sites.
[17:30:19] *** oXzibit has quit IRC
[17:34:41] *** mordaunt has joined #postfix
[17:35:23] *** frieder has joined #postfix
[17:37:28] *** birmaan has joined #postfix
[17:40:48] *** frennkie has joined #postfix
[17:43:18] <hparker> !cheatsheet
[17:43:19] <knoba> hparker: 'cheatsheet' : http://jimsun.linxnet.com/misc/postfix-anti-UCE.txt : A HOWTO for pre-DATA spam control.
[17:54:52] <Blackvel> do reject_rbl_clinent
[17:56:02] <rob0> Blackvel: the best thing to do there is to run a local caching nameserver.
[18:03:38] *** nescius has joined #postfix
[18:04:06] *** bostik has quit IRC
[18:05:29] *** Rattail has quit IRC
[18:08:25] *** eracc has joined #postfix
[18:14:02] *** Roobarb has quit IRC
[18:14:10] *** GMFlash has quit IRC
[18:14:16] *** GMFlash has joined #postfix
[18:14:23] *** Roobarb has joined #postfix
[18:15:44] *** [miles] has quit IRC
[18:16:04] *** jhutchins has joined #postfix
[18:17:17] <jhutchins> We seem to be logging IP addresses when we meant to log hostnames.  Is there a known issue with this?  Have we just been stupid in our conf files?
[18:18:18] <eracc> Same here. Lots of "connect from unknown" ...
[18:18:37] <eracc> Kinda messes up postgrey.
[18:20:41] <Supaplex> isn't postgrey running as a policy daemon? It should get the ip address...
[18:23:02] <rob0> Sounds like another infamous chroot issue.
[18:23:37] <eracc> rob0, I think you are correct.
[18:23:45] * eracc is checking that now
[18:24:54] <rob0> Of course a lot of zombies ARE spewing from hosts without rDNS. Check a few of those.
[18:26:42] <eracc> Hmm, a diff of /etc/resolv.conf and /var/spool/postfix/etc/resolv.conf shows those are the same. So it is not the resolv.conf file.
[18:27:38] <rob0> is /var/spool/postfix/etc/resolv.conf world readable?
[18:31:56] <Dominian> rob0: Going to take anohter stab at amavisd
[18:37:41] <Supaplex> use a BIG knife :)
[18:38:31] <eracc> At least this has happened to others in the past (wow, back in 2005): http://archives.neohapsis.com/archives/postfix/2005-06/2326.html
[18:38:45] <eracc> Unfortunately I saw no answer to that post.
[18:39:26] <eracc> Here is output from our 'postconf -n': http://rafb.net/p/06AjyF71.nln.html
[18:40:45] *** JMartinC has joined #postfix
[18:41:18] *** darkphader has joined #postfix
[18:42:25] *** mastachand has quit IRC
[18:53:43] <jhutchins> Hm.  Weather channel appears to be down.
[18:54:56] <rob0> USA? I use weather.gov for USA wx.
[18:55:21] <Dominian> wunderground.com for me
[18:58:28] *** foo has left #postfix
[18:59:29] *** frieder has quit IRC
[18:59:30] *** ScottMac has joined #postfix
[18:59:46] <ScottMac> I've got a mail that won't send, postfix just hangs
[18:59:57] <ScottMac> its the content of the email but I can't see what
[19:00:02] <ScottMac> is there an easy way to check?
[19:14:02] *** Roobarb has quit IRC
[19:14:22] *** Roobarb has joined #postfix
[19:19:13] *** nescius has quit IRC
[19:21:33] <lunaphyte_> ScottMac: what does your log say?
[19:23:08] <ScottMac> nothing, it shows all the SASL authentication and then shows the standard 250 2.1.5 Ok, data etc etc
[19:23:15] <ScottMac> then times out
[19:23:25] <ScottMac> I'm setting up a copy of tshark but I can't remember how to do capture filters
[19:23:51] <Dominian> just use tcpdump
[19:23:52] <Dominian> ;)
[19:23:58] <Dominian> tcpdump -ni eth0 port 25
[19:24:09] <rob0> Um ... it shows nothing?
[19:24:12] <lunaphyte_> !debug
[19:24:13] <knoba> lunaphyte_: 'debug' : http://www.postfix.org/DEBUG_README.html : a good starting point for how to deal with problems and to report information to those who might help. Post your information in a pastebin such as http://pastebin.com/ or http://rafb.net/paste/ .
[19:24:47] <lunaphyte_> i once had someone tell me that they were ready to go, they just had to get dressed.
[19:24:54] <rob0> How do you know Postfix is "hanging"? What's the symptom?
[19:25:03] <rob0> indeed
[19:26:36] <ScottMac> rob0, i don't know what the mail is yet so I can't work it out  i'll ask more details in abit
[19:28:07] *** devdas has joined #postfix
[19:30:22] *** Ryushin has quit IRC
[19:31:07] *** ALKH has joined #postfix
[19:37:36] <eracc> Sent myself some mail on the server and it appears DNS lookups are not being done in the headers either: Received: from mailout17.yourhostingaccount.com (unknown [65.254.253.138]) ...
[19:40:56] <rob0> $ dig mailout17.yourhostingaccount.com. any # that IS an unknown host.
[19:42:26] <Dominian> and dns shouldn't be done in the headers anyway should they?
[19:42:48] <eracc> Reverse DNS on 65.254.253.138 should pull up that name. It does here.
[19:43:18] <rob0> eracc: ^^
[19:43:20] <Dominian> RDNS working doesn't mean that DNS will work.
[19:43:34] <Dominian> whoa
[19:43:41] <eracc> DNS works here too.
[19:43:42] <Dominian> mailout17.yourhostingaccount.com has like.. 8 IPs
[19:44:04] <eracc> It is a big hosting company. I have a site there.
[19:44:08] <Dominian> So DNS is working, but maybe not working on your postfix box
[19:44:18] <rob0> oh nm ... I got confused
[19:44:27] <eracc> The Postfix box is also the DNS box.
[19:44:32] <ScottMac> ok my error is a bad checksum
[19:44:41] <rob0> .138 is one of the IP's.
[19:44:44] <ScottMac> nfc why its only happening with this one mail
[19:46:04] <eracc> The Postfix box is a web/e-mail/DNS server for a small business that prefers to self-host for more control.
[19:46:11] <fzzzt> anyone know what kinda modem a dsl modem actually uses? it cant be a 56k can it? dsl can be loads faster than that...
[19:46:32] <fzzzt> wow wrong channel, sorry :)
[19:46:38] <eracc> Heh
[19:46:45] <Dominian> hah
[19:46:50] <Dominian> fzzzt: and the answer is: no
[19:47:00] <Dominian> DSL has nothing to do with 56k dialup internet.. totally different protocol
[19:47:03] <rob0> :) Dominian is a DSL tech
[19:47:05] <fzzzt> yeah
[19:47:06] <fzzzt> oh :)
[19:47:24] <Dominian> rob0: DSL is a bitch really.
[19:47:25] <Dominian> :)
[19:47:31] <rob0> something about ATM ... I don't know much about it
[19:47:35] <Dominian> yeah
[19:47:38] <eracc> DSL "modems" are not MOdulation/DEModulation devices AFAIK.
[19:47:43] <fzzzt> ok
[19:47:44] <rob0> neither does my (*&^*&^$ ISP.
[19:47:45] <Dominian> they use ATM circuits to DSLAMs.. split those off to the homes.
[19:47:55] <Dominian> eracc: yes and no
[19:48:06] <Dominian> eracc: DSL still does a type of modulation...
[19:48:18] <Dominian> which is why you have to use a freakin' DSL filter on phones.
[19:48:40] <fzzzt> sooo i cant just buy a pcmcia card and use that eh
[19:48:41] <fzzzt> :P
[19:48:46] <Dominian> nope
[19:48:57] <Dominian> Who are you getting DSL through?
[19:49:13] <fzzzt> a local phone company
[19:49:31] <rob0> Sometimes asking in the wrong channel is the Right Thing to do. :)
[19:49:34] <fzzzt> but i was more looking into making my own dsl modem/wifi machine using fbsd
[19:49:38] <fzzzt> no kidding! :)
[19:49:58] *** Turt|e has quit IRC
[19:51:13] <Dominian> heh
[19:51:32] <Dominian> You can get an external DSL modem and hook it into your fbsd box which should be able to negotiate pppoe
[19:51:36] <Dominian> if that is what the phone company uses.
[19:51:46] <fzzzt> i used to have a little server that dialed up back in the ppp days, was looking into re-doing that again, but maybe its not feasible
[19:51:49] <rob0> There *are* internal devices, but hard to find.
[19:51:52] <Dominian> however, the modem you'd want is a bridge modem or one that can be bridged.
[19:51:59] <Dominian> rob0: and expensive
[19:52:13] <Dominian> and harder to replace hehe
[19:53:00] <rob0> What I don't understand is why SBC/AT&T has us (my company) doing PPPoE for a static IP?
[19:53:10] <Dominian> Easier administration
[19:53:16] <fzzzt> we did that too when we used dsl
[19:53:17] <Dominian> We do the same hre.
[19:53:25] <Dominian> authenticates to RADIUS and issues your IP out that way.
[19:53:30] <fzzzt> yeah
[19:53:56] <rob0> In other words, they're a bunch of lusers. :)
[19:54:34] <fzzzt> there's no reason one couldn't do dhcp and MAC addresses, is there? i dont know much about phone networks
[19:56:03] <rob0> I've heard of some DSL providers using DHCP. Never seen one hands-on, tho.
[19:56:25] <devdas> not worth the trouble, generally
[19:56:44] <devdas> Plus, PPPoE lets you configure bandwidth automagically
[19:56:47] <devdas> DHCP doesn't
[19:57:07] <fzzzt> configure bandwidth?
[19:57:47] <fzzzt> oh like setting radius attributes for max up/down rates
[19:58:00] <devdas> yes
[19:59:02] <fzzzt> ok, well thanks for the info guys :)
[20:00:58] <Dominian> well..
[20:01:13] <Dominian> Doing so with PPPoE is actually a waste of time considering the  LEC determine bandwidththe DSL signal
[20:02:17] *** cilly has joined #postfix
[20:02:37] <rob0> Hey, we would have paid whatever they wanted to get as much bandwidth as they could stuff down the pipe.
[20:02:39] <fzzzt> interesting... so "line noise" can affect your speed, like in the good old days? :P
[20:02:59] <rob0> We don't need no steekin' limits!
[20:05:33] <devdas> rob0: but then they can charge you more for that
[20:05:41] <devdas> fzzzt: yes
[20:05:47] <devdas> but on different frequencies
[20:05:55] <devdas> rob0: move to .jp or .hk
[20:06:13] <rob0> Money is no object for this company. But Asia is a bit out of the way. :)
[20:07:02] <Dominian> so..
[20:07:21] <Dominian> is it possible to setup an email alias that allows you to send email to everyone on your postfix server, but limit it to one email address that can send to it...
[20:08:11] <devdas> yes
[20:08:16] <devdas> !restriction_classes
[20:08:17] <knoba> devdas: Error: "restriction_classes" is not a valid command.
[20:08:24] <devdas> !restriction_class
[20:08:25] <knoba> devdas: 'restriction_class' : http://www.postfix.org/RESTRICTION_CLASS_README.html
[20:08:47] *** amrit|zzz is now known as amrit|wrk
[20:09:45] *** Ryushin has joined #postfix
[20:10:00] <rob0> Dominian: you'd want a special SQL query to generate the list of users on the fly.
[20:10:26] <Dominian> hmm
[20:10:28] <Dominian> that can be done.
[20:11:11] <rob0> Yup, there's your project in place of amavisd-new. :)
[20:12:49] <Dominian> hah
[20:12:54] <Dominian> mailscanner works for me
[20:12:55] <Dominian> :)
[20:14:03] *** Roobarb has quit IRC
[20:14:22] *** Roobarb has joined #postfix
[20:20:53] *** ziro has quit IRC
[20:24:24] *** funtable has joined #postfix
[20:27:04] *** funtable has left #postfix
[20:34:17] *** devdas has quit IRC
[20:35:15] *** rootsvr has joined #postfix
[20:51:32] <ALKH> Hi, I have created an email alias like team at toto dot com which redirect to many users mail address.I want to know if it is possible in postfix config to automatically fill the reply-to field to team at toto dot com when a mail is sent to this address?
[20:52:05] *** x-spec-t has joined #postfix
[20:56:31] <hparker> Why not use a mailing list?
[20:57:41] <xpoint> ALKH, emerge -av mlmmj
[20:57:44] <rob0> hparker takes all the challenge out of it
[20:58:04] <hparker> lol
[20:58:12] * hparker prefers mailman
[20:58:34] <xpoint> hparker, mailman is .... well you geess :-)
[20:58:59] <hparker> It Just Works ;)
[20:59:08] <xpoint> so does mlmmj
[20:59:41] <hparker> I have to admit, I've not looked at anything else in a couple/few years
[20:59:57] <xpoint> Mailling List Made Joid
[21:00:47] <ALKH> xpoint: ok thx, i take a look
[21:00:54] <ALKH> is it easy to use?
[21:01:05] *** cilly has quit IRC
[21:01:17] <ALKH> (mailman is too complex;))
[21:01:20] <xpoint> hparker, i once tryed to get mailman working with postfix, after it failed 20 times i gave mlmmj a try that worked first time
[21:02:10] <xpoint> ALKH, you ask to much :-)
[21:02:13] <hparker> It took me a bit to get it going using all virtuals... Once I figured out to make it a local, worked peachy
[21:02:38] <ALKH> ;)
[21:02:47] <hparker> And I only host like 3 lists
[21:03:03] <xpoint> hparker, overkill
[21:03:19] <hparker> Not it's not, they can admin them themselves ;)
[21:03:26] <rob0> (It's easy enough to use virtual(5) aliases to redirect lists to local(8) delivery.)
[21:03:40] <hparker> I just used lists. ;)
[21:03:51] <hparker> Dump it in mydestination
[21:04:10] <hparker> Now, if I was a big hosting provider, I'd look at something less manual
[21:04:16] <rob0> yes, that's easy too :)
[21:07:16] *** ScottMac has left #postfix
[21:08:44] <xpoint> hparker, is suggested to author of mlmmj to add sql config, he rejected it :(
[21:09:05] <hparker> :(
[21:09:39] <xpoint> imho silly when there is php / perl webcontrol in the contrib
[21:10:52] <xpoint> but i like what mlmmj does anyway, when maillist is op, it rooks
[21:14:02] *** Roobarb has quit IRC
[21:14:22] *** Roobarb has joined #postfix
[21:26:37] *** stellina has joined #postfix
[21:27:24] *** cpm has joined #postfix
[21:28:50] <Blackvel> what webinterfaces are you using with postfix? two weeks ago I saw someone from a company using outlook client web access with an exchange server. wouldn't it be possible to use that web client for postfix?
[21:29:27] * hparker likes the simplicity of squirrelmail
[21:29:57] <Blackvel> my other email provider uses it with imap
[21:29:58] <sepski> i use squirrelmail and roundcube
[21:29:59] <Dominian> Blackvel: outlook web access is for exchange only
[21:30:05] <Dominian> roundcube here
[21:30:19] <Blackvel> but hmm. looks there is no way even to search for subject/bodytext in the emails?
[21:30:24] * cpm likes the simplicity of hparker
[21:30:24] <Dominian> Using outlook web client for IMAP is ok.. but only gives you email access.. not calendar
[21:30:47] <Blackvel> oh it works really?
[21:30:54] <Blackvel> the pretty was really very nice
[21:30:54] * hparker likes being simple
[21:32:25] <Blackvel> bit better than squirrel :)
[21:32:25] <Blackvel> i run latest 2.3 postfix
[21:32:25] <Blackvel> for IMAP I would have to add another front server?
[21:33:10] <Blackvel> in mail.log I can find from time to time this:
[21:33:27] <Blackvel> in.qpopper[12623]: (null) at <remoteservername> -ERR POP EOF or I/O Error [popper.c:820]
[21:34:12] <Blackvel> is qpopper a pop3 server in front of postfix?
[21:34:34] <Dominian> don't think so
[21:34:39] <Dominian> qpopper is a pop server iirc
[21:36:40] *** JMartinC has quit IRC
[21:37:49] <Blackvel> ehm
[21:37:53] <Blackvel> what is postfix? :)
[21:38:03] * rob0 likes the volume knobs on hparker's wife's sound system!
[21:38:18] <Blackvel> whenever my email client connects to my email server I can see this in.qpopper messages
[21:38:22] * rob0 runs away fast
[21:38:28] <Dominian> Blackvel: postfix doesn't do pop
[21:38:34] <Dominian> Blackvel: its an MTA
[21:38:40] <Dominian> I use dovecot for POP
[21:38:41] <Blackvel> i mean all mails are stored by postfix
[21:38:46] * hparker trips rob0
[21:39:04] <Alan_Hicks> hparker: It's ok, man.  I have root on one of his boxes. :^)
[21:39:20] <hparker> Alan_Hicks: Nice, fix him up  please
[21:39:39] * rob0 *&^$(*#&$()*&)(*%
[21:39:46] <Alan_Hicks> rob0: You heard him rob0.  I can't be responsible for my actions....
[21:39:59] <hparker> *BEWM*
[21:40:29] *** rob0 has quit IRC
[21:40:48] * Alan_Hicks whistles innocently.
[21:42:14] <hparker> rfol
[21:42:15] <hparker> rofl
[21:42:28] <Dominian> roflcopter
[21:42:42] <Alan_Hicks> Wonder if he's noticed yet.
[21:42:43] <hparker> Alan_Hicks: He was referring to http://rafb.net/p/RKMmjw62.nln.html
[21:43:07] <Alan_Hicks> haha
[21:44:21] <Alan_Hicks> Maybe when I killed his irssi client it just continued to display on his screen and he thinks that no one has said anything?
[21:45:17] <hparker> lol
[21:45:48] <Dominian> wait
[21:45:52] <Dominian> You killed his irssi client?
[21:45:59] <Alan_Hicks> :^)
[21:46:00] <hparker> Hard core...
[21:46:02] <Dominian> It should've dropped him out right away
[21:46:06] <Dominian> unless he was using screen..
[21:46:10] <Alan_Hicks> He was.
[21:46:13] <Dominian> hhaha
[21:46:28] <hparker> Shoulda just iptabled him out ;)
[21:46:51] <Alan_Hicks> hparker: Nah.  He owns 1/3 of the box with me.  I can't do that. :-)
[21:47:00] <hparker> Ahh
[21:47:19] <Alan_Hicks> I can't believe he hasn't noticed and come back yet.  Must've got distracted.
[21:47:33] <Dominian> Either that or he's going to do something even worse to you.
[21:47:40] *** rob0 has joined #postfix
[21:47:40] <hparker> So, we normally get battle of the bots in irc, now we're going to watch battle of the roots ;)
[21:47:49] * Alan_Hicks whistels innocently.
[21:47:56] <hparker> wb rob0
[21:48:13] <hparker> We were just wondering where you got off to
[21:48:23] <rob0> Were we?
[21:48:38] <Alan_Hicks> I don't know what happened...  more connection problems?
[21:48:47] <rob0> I think we knew where I was.
[21:48:58] *** StormTide has joined #postfix
[21:49:07] <Alan_Hicks> Sorry man, couldn't resist.
[21:49:12] *** FMJaguar has joined #postfix
[21:49:25] <Alan_Hicks> Funny thing though.... I killed AthlonRob's irssi session first by mistake.
[21:49:37] <rob0> Alan_Hicks did me a favor, I wasted too much time in IRC today :)
[21:49:58] <hparker> Damnit, that's 2 keybards today!
[21:49:58] <StormTide> anyone familiar with the pgsql adapter and if its possible to configure a maximum number of connections it will try to use
[21:50:03] <rob0> AthlonRob'll never miss it.
[21:50:08] * Alan_Hicks thinks this ain't a good time to ask rob0 for more advice on integrating dnsmasq and bind to block websites.
[21:50:30] <rob0> Yeah, that's all in my LOST scrollback buffer.
[21:50:49] <hparker> Alan_Hicks: rbldnsd is easy
[21:51:17] <Alan_Hicks> hparker: Don't need RBLs, I need a whitelist, and only for a couple of nodes on the LAN, not everyone.
[21:51:49] <hparker> I use it for a local blacklist, set it up as a forward in bind
[21:52:41] <Alan_Hicks> I'd go into more detail, but it's off-topic here.
[21:53:05] <rob0> hey, #postfix is off topic today.
[21:53:08] <hparker> google for "rbldnsd mjt".. It has the docs on how to do it iirc
[21:53:41] <Blackvel> have a good evening
[21:53:47] <Blackvel> going to relax on my sofa :)
[21:54:12] <rob0> Good evening? No thanks, I have other plans.
[21:55:45] <rworkman> Damn.  I wish I'd caught on to what happened...  Alan_Hicks would have disappeared too ;)
[21:55:54] <Alan_Hicks> HAHAHA!
[21:56:03] <Blackvel> yes, little bit late here
[21:56:05] <Blackvel> cya
[21:56:08] *** Blackvel has quit IRC
[21:56:15] <Alan_Hicks> And naturally, I'd've blamed that on rob0.
[21:56:19] <rworkman> Yep :)
[21:57:00] <rob0> That's all so not nice.
[21:58:16] * hparker blames rob0
[21:58:37] * rob0 blames the Republicrats
[21:58:46] <rob0> and cpm
[21:59:00] <rob0> got to drag cpm into this
[21:59:04] * Alan_Hicks blames the Demlicans.
[21:59:28] * rob0 wields a spork
[22:01:18] *** x-spec-t is now known as Spec
[22:10:24] *** FMJaguar has quit IRC
[22:11:50] *** frennkie has quit IRC
[22:13:28] * cpm picks at his drag marks
[22:14:02] *** Roobarb has quit IRC
[22:14:22] *** Roobarb has joined #postfix
[22:14:57] * rob0 laughs at cpm in drag!
[22:15:08] * cpm pouts
[22:15:21] <rob0> You look ridiculous.
[22:15:48] <cpm> what? is it the push up? or the lipstick?
[22:16:12] <rob0> The lipstick is wrong for your hair color.
[22:16:19] <cpm> ah, thanks
[22:16:26] * cpm runs from the fashion police
[22:16:26] <rob0> Looks like you applied it with a spork!
[22:16:36] <cpm> well, maybe I did, what's it to ya?
[22:17:57] *** meandtheshell has quit IRC
[22:19:04] <hparker> Alan_Hicks: And you were worried about talking about bind...
[22:20:10] <rob0> cpm needs to rebind that pushup
[22:20:32] *** fzzzt has quit IRC
[22:20:33] <Alan_Hicks> I thought he needed to wear a masq.
[22:21:42] <Dominian> man.. I'm dumb-founded by the amount of legit businesses that use hotmail for correspondence..
[22:22:10] <Alan_Hicks> I'm dumbfounded the amount of use of hotmail for correspondence.
[22:22:21] * rob0 spits on hotmail!
[22:22:36] <rob0> I'm in just such a company.
[22:22:42] * Alan_Hicks watches rob0's spit sizzle on the hotmail.
[22:22:43] <Dominian> hehe
[22:22:52] <rob0> And hotmail users lose their mail.
[22:23:30] <Dominian> yep
[22:23:36] <rob0> Asbestos I can tell, they delete any BCC's.
[22:24:12] <rob0> not even "spma folder".
[22:24:15] <rob0> spam
[22:25:52] <cpm> any business, aside from Billy's plumbing, that uses hotmail is suspect.
[22:26:50] * rob0 suspects his employer is devoid of Clue
[22:27:45] <cpm> Billy's Plumbing?
[22:28:16] <Alan_Hicks> Parker Bradley?
[22:29:15] <Alan_Hicks> Excuse me, Parker Brothers?
[22:29:32] <rob0> whew now the joke makes sense
[22:29:43] <Alan_Hicks> Sorry. :-(
[22:30:57] <rob0> It was Colonel Mustard, in the Ballroom, with the Knife.
[22:31:23] *** lclimber has joined #postfix
[22:31:37] *** AJ__Z0 has joined #postfix
[22:31:43] <Dominian> Milton Bradley
[22:31:44] <Alan_Hicks> But it was self-defense!  Mr. Body cut the Colonel Mustard with that knife in the dining room.
[22:31:50] <Alan_Hicks> Dominian: Thanks.
[22:31:55] * Dominian nods
[22:32:04] <rob0> Milton Brothers!
[22:32:05] <Alan_Hicks> Though I think Parker Brothers made Clue.
[22:32:16] * hparker got rich off of that
[22:32:21] <lclimber> hello everyone, i need to build a cluster for mail. what debian package do you recommend?
[22:32:32] <rob0> hparker: and your brother?
[22:32:46] <rob0> Hey ... that's ME
[22:32:50] <hparker> lol
[22:32:51] <rob0> where's my share?
[22:33:00] <hparker> Oppppsss
[22:33:48] <rob0> lclimber: People in #postfix might tend to suggest using Postfix.
[22:34:32] <lclimber> sure, but is there a way to build a ha cluster for postfix??????
[22:34:45] <Alan_Hicks> lclimber: Yes.
[22:35:12] <lclimber> do you know where can i find documentation about it?????????
[22:35:26] <Alan_Hicks> Sure.  Look in the source code or on postfix.org.
[22:35:28] <Dominian> hmm if squirrelmail had a nicer interface I might use it.
[22:35:29] <rob0> www.postfix.org on the Postfix part.
[22:35:48] <rob0> The HA cluster is a different matter.
[22:35:51] <Alan_Hicks> Understand postfix as a stand-alone mail server first, then start working on the HA stuff.
[22:35:52] <lclimber> cool, thanx
[22:35:54] <lclimber> really
[22:36:12] <Alan_Hicks> You're welcome.
[22:36:13] <lclimber> what kind of clusterint is ussually set for postfix?
[22:36:31] <Alan_Hicks> Whatever kind you need.
[22:37:16] <lclimber> cool. am gonna check the postfix page then....you are all very kind
[22:37:23] <Alan_Hicks> Do you even know what you need?
[22:38:16] *** AJ_Z0 has quit IRC
[22:38:22] *** AJ__Z0 is now known as AJ_Z0
[22:38:46] <lclimber> yeah, well, am not sure of the exact name of the cluster i need, but i want one machine to do tje other? work when failing
[22:39:28] <Alan_Hicks> You've got a LOT of ground to cover.
[22:39:44] <Dominian> You mean backup mx?
[22:40:02] <cpm> backup mx=evil
[22:40:04] <rob0> http://www.linuxvirtualserver.org/ maybe?
[22:40:17] <cpm> maybe virtualrob0
[22:40:18] <Dominian> cpm: backup mx's can be a life saver.
[22:40:50] <sepski> Dominian, why ?
[22:40:51] <cpm> MX, do it once, do it right. backup MX is only good for radically increasing spam
[22:41:01] <lclimber> hmmm, i guess, the thing is that i need this server to be able to do the work of the other if one fails, e.g. if one server fails, i needo the other to assume all the work
[22:41:02] <sepski> it's not doing anything the sending MTA allready is dooing
[22:41:04] <rob0> I can't even get my MX to go forward, much less in reverse!
[22:41:13] <cpm> since none of us are on dialup anymore, it really serves no purpose
[22:41:25] <rob0> My transmission won't transmit
[22:41:33] * Alan_Hicks puts his MX in 4WD.
[22:41:41] <cpm> separate incoming/outgoing makes a lot more sense for high traffic
[22:41:54] <sepski> lclimber, that's not a backupmx. you'd want a failover like heartbeat. or a mailcluster setup
[22:42:20] <lclimber> yeah, do you have any sugestions?
[22:42:39] <rob0> A backup MX is like a sign taped on your back, saying "Spam me!"
[22:42:52] *** birmaan has quit IRC
[22:42:59] <sepski> lclimber, get a shared storage. connect X number of smtp servers to it. have roundrobin MX records all on the same priority.
[22:43:14] <sysmonk> rob0: not "Spam me!" but "PLZ PLZ PLZ PLZ I LIKE SPAM PLZ PLZ PLZ "
[22:43:15] <sysmonk> ;)
[22:43:44] <sepski> rob0, perhaps it could be a anti spam techniqe ? the backups mx delivers to /dev/null :)
[22:43:55] <sysmonk> sepski: there is one already
[22:44:11] <sysmonk> pointing a backup mx to a IP without SMTP runing on it
[22:44:32] <rob0> Oh I do that.
[22:44:32] <sysmonk> the spammers wich mail without looking if the mail delivers or not, won't try the normal mx'es
[22:44:35] <Dominian> sepski: I do backup mx for a few domains.. they go down.. my box holds their email until their main box comes back up
[22:44:43] <sysmonk> and normal SMTP servers should try to deliver to other MX'es too
[22:44:49] *** pickcoder has joined #postfix
[22:45:02] <sysmonk> so, you get all the email you need, and fight spam a bit
[22:45:10] <sepski> Dominian, yes i know the purpose. but it's pointless. since the sending MTA is allreay holding the mail until the box comes up again :)
[22:45:19] * Dominian shrugs
[22:45:26] <Dominian> With the sending MTA it will time out after a while
[22:45:43] <rob0> $ dig room101.us.eu.org. mx # note the "null" and "void" names
[22:45:59] <sepski> Dominian, ofcourse. but you realy can't sit on the mail for >4 days and still claim it's not "lost"
[22:46:23] <Dominian> and with the backup mx use.. the backup mx will hold it for as long as you tell it to
[22:46:34] <Dominian> So, its a matter of opinion if its useful or not.
[22:47:17] <rob0> It comes up on postfix-users regularly, and the "it's useful" opinions get shouted down. :)
[22:47:20] <sepski> i guess if you'r mailserver is down more then 4 days and your clients like to get the old mails that they have being calling eachother about a few times allready it's useable
[22:48:25] <sysmonk> we've got one mx backup wich is used only for clients which want to use it
[22:48:38] <sysmonk> but we don't use any for the servers we own
[22:48:45] * Dominian shrugs
[22:48:53] <Dominian> You can shoot down my opinion all you want.
[22:48:56] <Dominian> I really don't care :)
[22:49:12] * rob0 takes aim ... BLAM! :)
[22:49:16] <sysmonk> i think it'd be better to use that box as another mail server
[22:49:17] <rob0> got it
[22:49:21] <Dominian> If I had multiple email servers that I managed.. be a different story.
[22:49:28] <sysmonk> and set both boxes to store the mail in one place
[22:49:34] <Dominian> sysmonk: slackadelic.com has 1 email server. that's it.
[22:49:37] * rob0 cuts up and shares Dominian's opinion
[22:49:40] <Dominian> sysmonk: I'm not big enough to have anymore than that.
[22:49:55] <Dominian> So doing backup mx with other friends and doing the same for them is viable.
[22:49:58] <Dominian> and makes sense
[22:50:20] <sysmonk> Dominian: mxbackup IS good, but having it as another mail server with the same weight is better
[22:50:46] <sysmonk> if the choice is between mxbackup and nothing, i'd choose mxbackup
[22:50:59] <rob0> hparker can have a piece of the opinion when I get my share of Clue profits.
[22:51:15] * pickcoder prefers an mx that doesn't go down
[22:51:23] <sysmonk> pickcoder: :)
[22:51:24] <Dominian> sysmonk: I do see your point.
[22:51:34] <sn00p-> May 10 12:50:05 tweaked postfix/smtpd[39147]: warning: problem talking to service rewrite: Unknown error: 0 whats the problem anybody knows?
[22:51:35] <Dominian> sysmonk: but right now.. my setup works.
[22:51:40] <sysmonk> Dominian: sure it does!
[22:51:41] <sysmonk> ;)
[22:51:50] *** StormTide has quit IRC
[22:51:51] <Dominian> if I do ever get anohter email server.. that's what I will morethan likely do.. backup mx at the same weight as the main.
[22:52:05] <Dominian> sn00p-: What do your logs say?
[22:52:12] <sn00p-> thats from the logs
[22:52:13] <sysmonk> Dominian: that's called load balancing
[22:52:29] <rob0> If your backup MX does exactly the same checking as the primary does, and if the primary never EVER rejects anything from the backup, it works. But that's usually a lot of trouble to maintain.
[22:52:41] <Dominian> sysmonk: hang on.. I'm having like 5 conversations revolving around mail servers.. I'm getting them mixed up :)
[22:52:45] * hparker sends rob0 a clue-by-four
[22:52:59] <sysmonk> Dominian: heh, i understand you :)
[22:53:21] <sysmonk> i didn't read everything you guys were talking about, so i might be talking off the topic or something
[22:53:23] * rob0 uses a spork to launch a bloody piece of opinion at hparker
[22:53:24] <sysmonk> ;)
[22:53:41] <hparker> mmMmm... Tasty
[22:53:50] <rob0> Hmmm now that would look good with cpm's lipstick!
[22:53:55] <sysmonk> ;)))
[22:54:08] <hparker> lol
[22:57:08] <Dominian> sysmonk: for now.. if my mail server goes down.. I really don't care.. most MTA's will hold on to it and forward later.. my box is never down more than 30secs.. if that... I would, however, like to get another email server setup for delivery at some piont in time, but that is way off in the future.. for now.. having another mx box off site only to do backup mx operations and hold email works fine.
[22:57:52] <hparker> Just be sure they're setup the same as far as recepient validation, spam filtering, etc
[22:58:21] <Dominian> hparker: that's not an issue.. really don'tcare about that
[22:58:32] <Dominian> but yes.. they do have spam/virus filtering
[22:58:49] <Dominian> which is useless on the backupmx since I log everything to mailwatch to allow for archiving.
[22:59:19] <hparker> Uhm.. It is an issue.. If it accepts mail for non-existent addresses, then forwards it to your main MX, it'll bounce it to the forged From...
[22:59:41] <sysmonk> Dominian: i do understand you and i know what you mean, and i told that having a backup mx is better than not having any at all
[23:00:17] * hparker has been doing this for a few years for several ISPs, including his own and has never used a backup MX
[23:01:02] <Dominian> sysmonk: oh.. I missed that part lol
[23:01:03] * sepski dont have backup mx's either but we only have 1000 domains  / ~6000 accounts too. so we't not very big.
[23:01:31] <Dominian> hparker: if it forwards to non-existent addresses postfix drops the mail.
[23:01:35] <Dominian> end of story
[23:01:38] <rob0> Mexico is too big for me to backup. They can handle their own backups.
[23:01:43] <hparker> You're missing it Dominian
[23:01:46] <Dominian> possibly
[23:02:15] <Dominian> Either that or I'm getting really tired.
[23:02:25] <sepski> Dominian, i could never do that. people send orders on mail. if someone lost a order becouser the sender mistyped a email address, we'd get fried.
[23:02:38] <hparker> If you don't validate To: on backup MX, it then sends to your main MX with invalid To:.. Main MX then bounces it to forged From: ..
[23:02:39] <sepski> or atlest sued for damages
[23:02:45] <rob0> Mmmmmmmmmm, fried sepski :)
[23:03:17] <Dominian> hparker: odd.
[23:03:39] <hparker> Dominian: You lost all of the envelope information on the backup MX
[23:03:44] <Dominian> hparker: because.. I've had my mail server down.. for about 3 hours during a rebuild... came backup..backup MX forwarded the emails.. nothing bounced.. anything that was invalid was dropped and logged.
[23:03:44] <hparker> That's what matters
[23:04:14] <hparker> Dropped?
[23:04:44] <Dominian> Yeah.. if the To: is invalid.. aka doesn't exist for the domain.. it just ignores the email
[23:04:56] <Dominian> giving uhhhh..550 iirc
[23:05:11] <hparker> Right, then it bounces back tot he forged From:
[23:05:50] <Dominian> hmm alright
[23:06:11] <pickcoder> how does it bounce if the message isn't queued?
[23:06:22] <Dominian> pickcoder: thank you.. that's what I was trying to say.
[23:06:36] <hparker> Where's it going to bounce it to?
[23:06:37] <pickcoder> pre-data checks don't queue anything
[23:06:45] <pickcoder> the MTA sending the e-mail has to hand up
[23:06:45] <Dominian> thank you pickcoder
[23:06:46] <pickcoder> hang up
[23:06:52] <Dominian> that's what I was trying to say
[23:07:00] <hparker> If it's coming from your backup MX with an invalid To:
[23:07:01] <Dominian> it doesn't bounc anything.. it sends a 550 user doesn't exist and drops the connection
[23:07:16] <hparker> What's your backup MX do with it?
[23:07:22] <Dominian> gets rid of it
[23:07:25] <pickcoder> hparker: the backup mx should accept it in the first place
[23:07:29] <pickcoder> gah
[23:07:30] <pickcoder> shouldn't
[23:07:42] <pickcoder> unless it's an open mail server
[23:07:50] <Dominian> pickcoder: well the backup mx.. as it is configured right now.. just queues everything and forwards it after the main mx comes back up
[23:07:50] <hparker> pickcoder: I know, Dominian is saying there's no need to validate recipient on the backup
[23:07:59] <pickcoder> eep
[23:08:04] <pickcoder> yeah.. bad mojo there
[23:08:04] <Dominian> I didn't say anything like that
[23:08:04] <hparker> Everything....
[23:08:15] <hparker> Dominian: Then how'd we get here?
[23:08:29] <pickcoder> Dominian: do you sync valid recipients?
[23:08:44] <hparker> <Dominian> hparker: if it forwards to non-existent addresses postfix drops the mail.
[23:08:49] <Dominian> pickcoder: negative.. backup mx is configured.. from what I'm told to accept email destined for my domain
[23:09:07] <hparker> Everything to your domain?
[23:09:08] <Dominian> hparker: How does that translate to "validate recipient on the backup"?
[23:09:09] <pickcoder> so you'll take anything as long as the domain is correct?
[23:09:19] <hparker> Do you keep a list of valid recipients on it?
[23:09:20] <Dominian> pickcoder: correct
[23:09:23] <pickcoder> that's not a good idea
[23:09:28] <Dominian> pickcoder: maybe so
[23:09:38] <pickcoder> you're asking for bounce hell as hparker mentioned
[23:09:52] <Dominian> Well, as I said, hasn't happened eyt.
[23:10:07] <Dominian> but I need to look at syncing the recipient table up with the backup mx anyway.
[23:10:10] <Dominian> so I see where you are going.
[23:10:48] <hparker> ;)
[23:11:08] <Dominian> ya know... I just realized something....
[23:11:13] <Dominian> I thought I still had a backup mx....
[23:11:16] <Dominian> I don't
[23:11:26] <Dominian> I removed it about 6 months ago when I did a rebuld..
[23:11:36] <Dominian> because I saw no freakin' point having it.
[23:11:37] <Dominian> lol
[23:11:39] <Dominian> nevermind
[23:11:40] <hparker> lol
[23:11:42] <sn00p-> anybody familiary with postfixadmin?
[23:11:52] * hparker feeds Dominian to rob0
[23:11:52] <Dominian> hparker: I had to go through my changelog files before I found that little gem
[23:12:01] <Dominian> hparker: actually it was rob0 who talked me into getting rid of it.
[23:12:03] <sysmonk> :))))
[23:12:07] <Dominian> sn00p-: yes.. use it on my box
[23:12:10] *** pirho has joined #postfix
[23:12:16] <hparker> lol
[23:12:20] <sysmonk> rofl
[23:12:21] <sysmonk> ;)))
[23:12:22] <hparker> sn00p-: I use it as well
[23:12:29] <sn00p-> Dominian i'm having problems with config.inc.php it gives me Parse error: syntax error, unexpected T_STRING in /usr/local/www/postfixadmin/config.inc.php on line 56
[23:12:31] *** pirho has quit IRC
[23:12:32] <pickcoder> I use vim
[23:12:47] <Dominian> sn00p-: what's on line 56?
[23:12:50] <sn00p-> Dominian but it says on line 56 // If you want to automagically generate paswords set this to 'YES'.
[23:13:02] <Dominian> then you are missing something else above that line
[23:13:05] <Dominian> like a } or something
[23:13:10] <Dominian> what did you change in the file?
[23:13:15] *** pirho has joined #postfix
[23:13:22] <sn00p-> just the data baes and info
[23:13:47] <sn00p-> am I suppose to change anything?
[23:14:01] <Dominian> $CONF['database_host'] = 'localhost';
[23:14:02] *** Roobarb has quit IRC
[23:14:13] <Dominian> make sure that every entry you change has the '<words here';
[23:14:22] *** Roobarb has joined #postfix
[23:14:31] <sn00p-> $CONF['database_host'] = 'localhost';
[23:14:44] <Dominian> I mean for everything
[23:14:46] <Dominian> not just that last one
[23:14:55] <Dominian> sn00p-: can you paste the stuff you changed to a patebin?
[23:14:55] <sn00p-> local host change to my domain?
[23:15:04] <Dominian> no
[23:15:08] <Dominian> I was using it as an example.
[23:15:31] <sn00p-> I guess I should try and get postfix working
[23:15:35] <sn00p-> it wont send mail
[23:15:38] <Dominian> uhhh
[23:15:44] <Dominian> Yeah.. first get postfix working then worry about postfixadmin.
[23:15:47] <sn00p-> something wrong with mysql files
[23:15:48] * rob0 can't eat another bite, after all that fried sepski
[23:15:58] <sn00p-> config files
[23:16:05] <Dominian> sn00p-: what tutorial are you using?
[23:16:25] *** stellina has quit IRC
[23:16:26] <sn00p-> I'm not using any I couldn't find one I'm justlooking at a friends setup
[23:16:42] <sn00p-> i wanna run postfix, postfixadmin, dovecot mysql
[23:16:56] <Dominian> hehe
[23:17:05] <Dominian> http://wiki.slackadelic.com/doku.php/howto:mailserver
[23:17:07] <Dominian> have fun
[23:17:16] <Dominian> Ignore the mailscanner/mailwatch stuff if you don't want to use that.
[23:17:27] <Dominian> but that tutorial does mysql,postfix,postfixadmin,dovecot
[23:17:41] <Dominian> and it was done on slackware so the commands needed may be different for you
[23:17:55] <sn00p-> thats not for freebsd tho
[23:18:10] <Dominian> so?
[23:18:17] <Dominian> It should still be close to what ou need
[23:18:30] <Dominian> It will at least give you some what of an idea of what you are getting yourself into :)
[23:18:43] <Dominian> although need to add af ew updates to it.. nothing to do with postfix though.
[23:18:44] *** lkthomas-home has joined #postfix
[23:19:06] <lkthomas-home> does default postfix use maildrop ?
[23:19:09] <sn00p-> I read the DATABASE_MYSQL.TXT
[23:19:12] <sn00p-> in dovecot
[23:19:17] <sn00p-> grr
[23:19:20] <sn00p-> postfixadmin
[23:19:22] <sn00p-> whwere ever it is
[23:20:36] <sn00p-> am I forgetting something I have to edit the aliass file
[23:21:47] <sn00p-> hmm
[23:21:55] <sn00p-> I think is a typpo
[23:23:10] <lkthomas-home> actually, if I didn't define mailbox_command, nothing will be use by postfix, am I correct ?
[23:23:44] <Dominian> wrong
[23:24:01] <Dominian>  The mailbox_command parameter specifies the optional external
[23:24:05] <Dominian> to use
[23:24:18] <Dominian> You don't need that unless you plan on passing mail through procmail or somethng of the sort first
[23:24:39] <lkthomas-home> Dominian: so it wouldn't use procmail and maildrop by default
[23:24:56] <Dominian> don't think so.
[23:25:03] <lkthomas-home> hmm
[23:25:38] <rob0> Postfix defaults to no external delivery agent. Debian, and perhaps others, OTOH, do set mailbox_command in their packaged main.cf.
[23:25:42] <lkthomas-home> out of topic question, does procmail read any default config file if ~/.procmailrc does not exists ?
[23:25:56] <hparker> /etc/procmail
[23:25:57] <rob0> maybe /etc/procmailrc
[23:25:58] * cpm defaults to the rob0 delivery agent, rda
[23:26:11] <hparker> really dumbass?
[23:26:11] <Dominian> mknod /dev/rob0 ?
[23:26:12] <rob0> cpm never gets any mail, either!
[23:26:21] <lkthomas-home> right, hmm
[23:26:34] *** lclimber has quit IRC
[23:27:06] <lkthomas-home> rob0: can I define a test account to use procmail while other still using no external delivery agent ?
[23:27:34] * rob0 decides to allow lkthomas-home's test account :)
[23:27:37] <lkthomas-home> if I use mailbox_command to define procmail, the whole server will follow
[23:27:53] <rob0> several ways ...
[23:27:56] <lkthomas-home> which is something I don't want before I understand it
[23:28:07] * lkthomas-home listening to rob0
[23:28:29] <rob0> 1. mailbox_command_maps  2. ~/.forward  3. something I haven't thought of yet :)
[23:28:48] <lkthomas-home> blah, hmm
[23:28:52] <sn00p-> Dominian yes, I got it to work
[23:28:55] <lkthomas-home> let me google for first option
[23:28:55] <rob0> Okay, s/several/two/
[23:28:59] <sn00p-> Dominian had user name and password setup wrong
[23:29:00] <Dominian> sn00p-: cool
[23:29:06] <sn00p-> for the databases
[23:29:07] <sn00p-> lol
[23:29:22] <rob0> Google not. Consult thou thine own postconf(5) manual.
[23:29:48] <rob0> #2 is in local(8) manual.
[23:30:02] <lkthomas-home> rob0: OK, sounds easy. test at domain dot com /usr/bin/procmail -a "$EXTENSION"
[23:30:28] <Dominian> couldn't you define that as a config file and put everything in there?
[23:30:33] <Dominian> hmm guess not
[23:30:43] <lkthomas-home> Dominian: huh
[23:30:55] <Dominian> nevermind
[23:30:56] <Dominian> I"m tired.
[23:30:57] <lkthomas-home> I mean for mailbox_command_maps config file
[23:30:59] <Dominian> heading home... back in a bit
[23:31:03] <Dominian> lkthomas-home: there you go
[23:31:05] <rob0> mailbox_command_maps = hash:/etc/postfix/mailbox_command_maps
[23:31:06] <lkthomas-home> thanks Dominian
[23:31:11] <lkthomas-home> yeah rob0
[23:31:12] *** olinux has joined #postfix
[23:31:13] <rob0> later Dominian
[23:31:17] <lkthomas-home> that's what I am talking about
[23:31:41] <lkthomas-home> then I can having fun with procmail without messing around the whole server
[23:32:26] <rob0> echo `which procmail` > ~/.forward
[23:33:09] <lkthomas-home> what happen if it doesn't exists ?
[23:33:15] <lkthomas-home> I mean ~/.forward
[23:34:00] *** hemry has joined #postfix
[23:34:26] <lkthomas-home> rob0: the problem is that I create account by postfixadmin, and I can't just insert those file to each account one by one
[23:35:44] <rob0> Ah, well that changes absolutely everything.
[23:36:27] <rob0> Postfixadmin is a frontend to mysql and virtual(8). All the settings we are discussing only apply to local(8) delivery.
[23:36:40] <lkthomas-home> opppsss
[23:36:45] <lkthomas-home> hell yeah
[23:36:47] <lkthomas-home> damn
[23:36:53] <rob0> There's no good way to do procmail without Unix accounts.
[23:37:00] <lkthomas-home> FUC*!
[23:37:07] <lkthomas-home> THAT
[23:38:30] <lkthomas-home> well
[23:38:35] <lkthomas-home> another choice is maildrop
[23:39:05] <rob0> I think some people have messed around with using a pipe(8) transport, but yes, maildrop is better. It's maintained code, too; procmail is not.
[23:39:50] <lkthomas-home> really ?
[23:39:53] <lkthomas-home> procmail is dead ?!
[23:40:05] <rob0> 2001 or so, I think.
[23:40:20] <lkthomas-home> LOL
[23:40:38] <lkthomas-home> so maildrop is the mainstream delivery agent ?
[23:41:24] <rob0> Damn ... 2001/Sep/11 on the tarball!
[23:41:45] <lkthomas-home> haha
[23:42:01] <rob0> It must have been under development in the World Trade Center.
[23:42:34] <lkthomas-home> woo
[23:42:39] <lkthomas-home> might be
[23:43:15] <lkthomas-home> one question
[23:43:26] <lkthomas-home> it have to define user on maildrop pipe option
[23:43:30] <lkthomas-home> what user should I use ?
[23:43:51] <lkthomas-home> I use postfix:postfix for my mail dir
[23:44:01] <lkthomas-home> I mean, for virtual user maildir
[23:44:05] <rob0> I never messed with pipe nor maildrop. No, it should not be postfix.
[23:44:07] <lkthomas-home> should I use user=postfix ?
[23:44:12] <lkthomas-home> hmm
[23:44:15] *** rootsvr has quit IRC
[23:45:39] <rob0> Postfixadmin says to use postfix for virtual_uid/gid_maps, but Wietse says to use one or more separate accounts.
[23:46:01] <rob0> I tend to trust Wietse's judgement.
[23:46:02] * hparker used 1001/1001 iirc
[23:47:11] <lkthomas-home> hparker: well, I define virtual uid = postfix, gid = postfix
[23:47:35] <lkthomas-home> so it should be ok to use user=postfix on maildrop pipe
[23:55:41] *** Mazon is now known as mazon
[23:55:56] *** Mez has joined #postfix

top