October 24, 2013 
< | 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 | >
October 24, 2013 [00:00:12] <kithpom> adaptr: an upside down one ¿ at the beginning can be.[00:03:47] <adaptr> since it must be matched by a normal one at the end, no.[00:03:49] *** Bry8Star has quit IRC[00:03:51] <mocx> adaptr: i'm reading a tutorial that says this: "Postfix requires us to have a Fully Qualified Domain Name or FQDN. Thats our full domain name, including our mail prefix, as our server name. To find out what your current server name is, you can use the command hostname:"[00:03:58] <adaptr> !tell mocx tutorial[00:03:58] <knoba> mocx: "tutorial" : A very common problem is that some people prefer to follow a step-by-step tutorial that shows them how to setup their server w/out reading the documentation or understanding what they are doing. If something goes wrong, they have no clue whatsoever about where to find hints, and they sometimes decide to start from scratch using a different tutorial. This is not The Proper Way.[00:04:00] <thumbs> mocx: why are you reading tutorials?[00:04:13] <thumbs> mocx: tutorials are for experts[00:04:29] <mocx> however I've heard it only cares about a PTR record[00:04:47] <thumbs> mocx: stop reading that tutorial[00:04:52] <thumbs> !fqdn[00:04:52] <knoba> thumbs: "fqdn" : the 'Fully-Qualified Domain Name'. It consists of the hostname part and the domain part. 'www.postfix.org' is a FQDN whereas 'www' is not. Only DNS resolvable, fully-qualified domain names (FQDNs) are permitted when domain names are used in SMTP.[00:04:58] <thumbs> erm, not that one.[00:05:29] <thumbs> !qrdns[00:05:29] <knoba> thumbs: Error: "qrdns" is not a valid command.[00:05:33] <thumbs> !fqrdns[00:05:33] <knoba> thumbs: Error: "fqrdns" is not a valid command.[00:05:36] <adaptr> thumbs: iHerd you like PTRs, so iMade a PTR to PT to UR PTR[00:06:26] <kithpom> time to go back to bed. Thanks again adaptr rob0[00:07:22] <mocx> well thanks for dancing around the question and providing me with opinions[00:07:24] *** mocx has left #postfix[00:07:46] <thumbs> you're very welcome, new postfix user.[00:07:47] <matthewt> another satisfied customer[00:08:06] <thumbs> and next time, don't follow tutorials[00:08:09] <adaptr> he got what he paid for[00:08:22] <thumbs> tutorials lead to pain, and pain leads to suffering[00:08:37] <adaptr> suffering leads to hate, and Star Wars 7.[00:08:40] <kithpom> I didn't get to welcome him to the internet :([00:09:15] <rob0> I gave him his probable answer, but that was ignored.[00:10:11] *** Bry8Star has joined #postfix[00:10:23] <adaptr> I loved how thumbs failed so utterly at repeating what you'd just said[00:10:33] <rob0> yes, that was precious[00:10:36] <adaptr> sssht best not make him feel bad about that[00:10:41] <kithpom> it's bullying like this that will lead him to shooting up his middle school.[00:10:55] <adaptr> if he is still in middle school he doesn't need a mail server[00:11:18] <kithpom> if he is in middle school working on his mail server it explains the bullying[00:11:19] <rob0> thumbs did that many years ago. He's now shooting up nursing homes.[00:11:53] <thumbs> oh, !fcrns[00:11:59] <thumbs> yeah, I'm distracted.[00:11:59] <adaptr> hahahaha[00:12:06] <adaptr> !fcrns[00:12:06] <knoba> adaptr: Error: "fcrns" is not a valid command.[00:12:41] *** kithpom has quit IRC[00:14:09] <UQlev> !fcrdns[00:14:09] <knoba> UQlev: "fcrdns" : http://en.wikipedia.org/wiki/Forward_Confirmed_reverse_DNS : your IP address should resolve to $myhostname, which in turn should resolve back to your IP. This is very important if you want big sites to accept your mail. If you can't have it from your ISP, see !relayhost[00:29:28] *** davlefou_ has joined #postfix[00:32:57] *** davlefou__ has quit IRC[00:37:27] *** jarif has quit IRC[00:44:07] *** UQlev has quit IRC[00:48:31] *** higuita has quit IRC[00:55:56] *** twb has joined #postfix[00:56:34] *** trusktr has joined #postfix[01:03:23] *** higuita has joined #postfix[01:15:37] *** MaximusColourum has quit IRC[01:29:50] *** davlefou__ has joined #postfix[01:33:27] *** davlefou_ has quit IRC[01:34:52] *** mattcen has quit IRC[01:37:41] *** mattcen has joined #postfix[01:58:38] *** hallamigo has quit IRC[02:15:24] *** rowinofwin has joined #postfix[02:16:53] *** rowinofwin has quit IRC[02:17:10] *** rowinofwin has joined #postfix[02:17:20] <rowinofwin> Hi guys, I have a new Freepbx server which I am trying to use voicemail to email on, the issue seems to be with the postfix server/firewall. When I try to telnet out from this new server to a known working mail server on 25 it hangs, if I do the same from another system on the same network no issue[02:17:36] <rowinofwin> I have tried stopping iptables, but no change[02:17:48] <rowinofwin> telnet on 80 and 465 are fine, so just 25[02:18:43] <lunaphyte> show actual data[02:18:56] <lunaphyte> pastebin telnet attempts[02:20:13] <rowinofwin> http://pastebin.com/Tnx1eNig[02:20:25] <rowinofwin> the last one is the 25 connection, still going now[02:20:42] <rowinofwin> telnet: connect to address xx.xx.xx.xx: Connection timed out[02:20:53] <rowinofwin> after one minute[02:20:59] <lunaphyte> pastebin iptables -vnL --lin[02:21:21] <rowinofwin> this mail server accepts connections on 26 though as well as 25[02:22:19] <rowinofwin> http://pastebin.com/KMJMz1EE[02:22:28] <rowinofwin> and this is with iptables stopped[02:22:47] <rowinofwin> it is the same started[02:23:10] <lunaphyte> well, then something else is blocking your connection attempts.[02:23:12] *** zerick has quit IRC[02:23:14] <rowinofwin> can I maybe change the default port?[02:23:18] <lunaphyte> can you telnet to *any* mail servers?[02:23:37] <rowinofwin> I can telnet on 26 to this mail server, but not on 25 to any mail server[02:23:54] <lunaphyte> what is "this mail server"?[02:24:03] <thumbs> lunaphyte: that server![02:24:17] <lunaphyte> ack![02:24:20] <rowinofwin> the mail server I will be commonly sending to[02:24:34] <rowinofwin> from the freepbx server to the mail server in a DC[02:24:44] <rowinofwin> the freepbx is the machine which cannot send[02:24:50] <lunaphyte> what sort of internet connection does this freepbx server have?[02:25:03] <rowinofwin> standard business asdl 2+[02:25:19] <lunaphyte> perhaps your isp is blocking port 25 outgoing[02:25:25] <rowinofwin> and I can telnet to the same external mail server on port 25 from another machine here on the same connection[02:25:37] <lunaphyte> oh, well that's different then.[02:25:40] <rowinofwin> yeah[02:25:44] <rowinofwin> really confusing[02:25:52] *** rowinofwin is now known as MIA_rowinofwin[02:26:07] <lunaphyte> two computers both on the same internet connection - one can reach a public internet mail server on port 25, the other can't?[02:29:12] *** MIA_rowinofwin is now known as rowinofwin[02:29:19] <rowinofwin> ha! you are correct[02:29:32] <rowinofwin> the other admin set up a route to our backup connection[02:29:52] <rowinofwin> so outgoing port 25 is being routed to that when it comes from the freepbx box[02:30:02] <rowinofwin> so it is limited to that internet connection[02:30:27] <lunaphyte> so the two computers are not using the same internet connection?[02:30:42] *** davlefou_ has joined #postfix[02:30:52] <rowinofwin> aparently they are for everything except a few differently routed ports, 25 inclusive[02:30:59] <rowinofwin> so that is the difference[02:31:01] <rowinofwin> testing now[02:31:12] <rowinofwin> yep[02:31:16] <rowinofwin> switched over and all is well[02:32:13] <lunaphyte> that's good.[02:32:30] <lunaphyte> btw - you don't need postfix installed on your freepbx server just to send email.[02:33:06] <lunaphyte> postfix is mail routing software - for putting on computers that are mail servers. mail routing software doesn't go on computer that just need to send email.[02:33:13] <lunaphyte> *go on computers[02:33:46] *** davlefou__ has quit IRC[02:44:54] *** rowinofwin is now known as MIA_rowinofwin[02:53:51] *** MIA_rowinofwin is now known as rowinofwin[03:07:05] *** rowinofwin is now known as MIA_rowinofwin[03:08:42] *** MIA_rowinofwin is now known as rowinofwin[03:10:00] *** rowinofwin is now known as MIA_rowinofwin[03:15:28] *** MIA_rowinofwin is now known as rowinofwin[03:15:41] *** rowinofwin is now known as MIA_rowinofwin[03:25:02] *** jelly has quit IRC[03:25:28] *** jelly has joined #postfix[03:30:55] *** davlefou__ has joined #postfix[03:34:03] *** davlefou_ has quit IRC[03:55:27] *** MIA_rowinofwin is now known as rowinofwin[03:57:07] *** rowinofwin is now known as MIA_rowinofwin[04:10:58] *** MIA_rowinofwin is now known as rowinofwin[04:21:02] *** rowinofwin is now known as MIA_rowinofwin[04:23:52] *** MIA_rowinofwin is now known as rowinofwin[04:31:37] *** davlefou_ has joined #postfix[04:32:57] *** rowinofwin is now known as MIA_rowinofwin[04:34:13] *** davlefou__ has quit IRC[04:36:50] *** MIA_rowinofwin is now known as rowinofwin[04:46:03] *** rowinofwin is now known as MIA_rowinofwin[05:00:49] *** MIA_rowinofwin is now known as rowinofwin[05:13:46] *** Phoenixz has quit IRC[05:15:44] *** jarif has joined #postfix[05:16:43] *** JamieH has quit IRC[05:16:55] *** Phoenixz has joined #postfix[05:17:40] *** JamieH has joined #postfix[05:18:08] *** rowinofwin is now known as MIA_rowinofwin[05:24:45] *** MIA_rowinofwin is now known as rowinofwin[05:30:50] <jimpop> Will "nested" access tests work? I have ..., check_sender_access, check_recipient_access .... where check_recipient_access utilizes a restriction class that calls check_sender_access. (i.e. verifying sender is a mailman subscriber)[05:31:52] *** davlefou__ has joined #postfix[05:31:54] <jimpop> the first check_sender_access is a hash lookup, the second check_sender_access is a tcp_table lookup[05:35:05] *** davlefou_ has quit IRC[05:38:17] <rob0> jimpop, right. That's the point of restriction classes, to be able to conditionally apply a restriction.[05:40:06] <jimpop> rob0: ? but will the 2nd call to check_sender_access occur (assuming the 1st check_sender_access call was DUNNO)?[05:41:51] <rob0> My magic crystal ball already knows the goal: to reject list posts from non-subscribers! :)[05:42:01] <jimpop> and just to be clear, the 1st call to check_sender_access and check_recipient_access is from a restriction class called from smtpd[05:42:21] <jimpop> rob0: yes... and loooong overdue ;-)[05:43:02] <rob0> so if the RCPT is a list (check_recipient_access) you check the sender against the subscriber list (check_sender_access)[05:44:27] *** trusktr has quit IRC[05:44:30] <jimpop> yes, but just prior to check_recipient_access, i also call a check_sender_access hash to filter out various invalid senders[05:44:32] <rob0> so I think you might be doing the wrong lookup first. First look up the list posting address as recipient.[05:44:37] <rob0> oh[05:45:22] <rob0> but is the first check_sender_access involved somehow?[05:45:32] <jimpop> it is inline[05:45:37] <rob0> or is it just another restriction beforehand?[05:45:46] <jimpop> just another restriction before[05:46:20] <jimpop> which makes me believe that it should be ok.[05:46:28] <rob0> yes[05:46:45] <jimpop> k[05:47:59] <jimpop> i wrote a simple mailman lookup (is_member) that I now call from master spawn(8), and then use a tcp table with check_sender_access[05:48:07] <jimpop> so that it's all "real time"[05:52:56] *** rowinofwin is now known as MIA_rowinofwin[05:54:11] <rob0> hmm. I think I would have had the Mailman add/delete subscriber function trigger a rebuild of a hash: map.[05:54:28] <pj> sometimes it's just easier to write a policy daemon.[05:54:53] <rob0> still, has to get the subscriber list out of Mailman[05:55:26] <pj> yeah, but the advantage of a policy daemon over a tcp table is you get all the info about the messsage for the policy daemon to work with instead of just a single key.[05:55:45] <pj> so you can put all your complex logic into the policy daemon and call it once.[05:56:13] <rob0> all are valid approaches ... just depends which one you feel more comfortable implementing[05:56:21] <pj> yep, true[05:56:31] <pj> I was just tossing that out there as another way to do it.[05:57:14] *** MIA_rowinofwin is now known as rowinofwin[06:03:29] <jimpop> well the mailman check is either Yes or No (OK or REJECT)... so that works well with check_sender_access[06:05:55] <pj> jimpop: what is involved in checking mm?[06:06:38] <jimpop> rob0: the problem with the add/del subscriber function, is that it is a mailman component (imbedded deep inside Mailman's own python code) that would be generating postfix formatted hash files. (I think that is crossing too much from one area into another)[06:07:35] <jimpop> pj: it's a very striped down copy of bin/find_mmember[06:07:46] <jimpop> *stirpped[06:07:49] <jimpop> doh[06:08:06] <pj> so your tcp daemon needs to fork that for every time it does the check?[06:08:24] <jimpop> yes.[06:08:49] <jimpop> but... it's a list server, so it only forks for every incoming list post (~30/day?)[06:09:07] <pj> ahhhh, so it's not high traffic enough to be an issue.[06:09:13] <jimpop> correct[06:09:40] <jimpop> the advantage is that if someone subscribes, they can immediately post[06:10:32] <pj> the disadvantage is that someone that tries to spam the list rapidly could DDOS the server by forcing it to fork so much.[06:10:47] <pj> well, DOS it[06:10:59] <jimpop> hopefully postscreen will take care of that[06:11:05] <pj> but I'm guessing that ... yeah[06:11:06] *** TomHome has joined #postfix[06:11:17] <pj> you have other anti-spam in front that should prevent that (hopefully).[06:11:33] <jimpop> also, the check_recipient check is way down the list of all the checks[06:11:39] <jimpop> yes[06:12:08] <pj> right, but it would (or should) come before more expensive anti-spam checks, like spamassassin.[06:12:57] <jimpop> that would be nice.. but I'm using spammass-milter, which pretty much hits right off the bat[06:13:57] <pj> yeah, that's what I was referring to, if you were using amavisd or something like that then SA wouldn't even be run until post-queue.[06:14:59] <jimpop> i may be wrong... i roll-up all my inbound restricions into a restriction class (smtpd_client_restrictions=$inbound_restrictions)[06:15:30] <jimpop> and things like reject_non_fqdn_hostname hit before spamassassin gets called[06:16:14] <jimpop> so yes, the check_recipient_access check will occur (after all the other checks) before spamassassin[06:16:27] <pj> yeah, well SA is a pretty expensive test, so it's best left until last, imo.[06:17:33] <jimpop> i guess the miters are at the end of the stack, just before queue[06:18:10] <jimpop> the only reason i don't use Amavis is because i want to reject[06:18:56] <pj> well, it makes sense, the recipient restrictions would happen at the RCPT TO stage, but the milter can't run until END OF DATA.[06:19:22] <jimpop> ahh, yes that explains it[06:19:23] <pj> ...mainly because the milter needs the data for deep content inspection.[06:19:46] *** jarif has quit IRC[06:20:41] <pj> I'm not sure if the milter would run before or after the data restrictions, though, I am curious about that.[06:20:50] * pj tries to find out in the docs[06:21:07] <jimpop> i would join you but my eyes are sore[06:21:10] <jimpop> :P[06:22:23] <pj> at a complete guess it probably runs after all recipient checks have passed.[06:23:35] <pj> obviously it's a non-issue for non_smtp_milters[06:23:45] <pj> +d[06:26:46] *** rowinofwin is now known as MIA_rowinofwin[06:27:01] *** trusktr has joined #postfix[06:27:51] *** MIA_rowinofwin is now known as rowinofwin[06:32:33] *** davlefou_ has joined #postfix[06:35:24] *** davlefou__ has quit IRC[06:47:20] *** rowinofwin is now known as MIA_rowinofwin[06:51:59] *** MIA_rowinofwin is now known as rowinofwin[07:05:08] *** kirin` has quit IRC[07:06:37] *** kirin` has joined #postfix[07:32:51] *** davlefou__ has joined #postfix[07:34:14] *** davlefou_ has quit IRC[07:36:05] *** chrisp15 has joined #postfix[07:37:36] *** err-or has quit IRC[07:42:51] *** Phoenixz has quit IRC[07:56:37] *** olegfusion has joined #postfix[08:00:41] *** rowinofwin is now known as MIA_rowinofwin[08:05:47] *** err-or has joined #postfix[08:13:31] *** steven4455 has quit IRC[08:14:50] *** trusktr has quit IRC[08:25:23] *** Bronze has quit IRC[08:34:12] *** davlefou_ has joined #postfix[08:36:50] *** davlefou__ has quit IRC[08:37:09] *** tld has joined #postfix[08:40:23] *** trusktr has joined #postfix[08:46:29] *** TomHome has quit IRC[08:47:00] *** TomHome has joined #postfix[08:54:17] *** jarif has joined #postfix[09:00:58] *** zorg1 has joined #postfix[09:01:12] *** cyrilz_ has joined #postfix[09:03:52] *** cyrilz_ has quit IRC[09:06:59] *** wdp has joined #postfix[09:07:00] *** wdp has quit IRC[09:07:00] *** wdp has joined #postfix[09:23:39] *** trusktr has quit IRC[09:30:24] *** abiteofcode has joined #postfix[09:33:28] *** abiteofcode has left #postfix[09:33:45] *** davlefou__ has joined #postfix[09:37:27] *** davlefou_ has quit IRC[09:52:08] *** Hydrazine has quit IRC[09:53:16] *** Hydrazine has joined #postfix[10:16:41] *** sharky_ has quit IRC[10:19:15] *** zafu has joined #postfix[10:26:28] <zafu> hi, I have a virtual_alias_maps with a pcre entry like "/^(info|contact|etc)@/ localuser" and it delivers info at anydomain dot com to localuser even though 'anydomain' is not in virtual_alias_domains, is that normal?[10:32:37] *** honestly has quit IRC[10:32:59] *** twb has quit IRC[10:33:53] *** Cromulent has joined #postfix[10:34:14] *** davlefou_ has joined #postfix[10:37:20] *** davlefou__ has quit IRC[10:41:26] *** wizbit has joined #postfix[10:42:20] <wizbit> i run postfix on a vps and also postfix on my home server, at the moment i relay email from my home server using my ISPs SMTP, is there a way to bypass the ISPs SMTP and relay it directly onto my postfix VPS?[10:43:02] <wizbit> when i try and do that at the moment the VPS logs say relay not allowed[10:43:03] *** davlefou_ has quit IRC[10:43:04] <sep> you can configure your home postfix to use your vps as it's smarthost just like you did for your isp's server[10:43:21] <wizbit> yep i did that but the VPS postfix log says relay not allowed[10:43:26] <wizbit> or something along those lines[10:43:28] *** davlefou has joined #postfix[10:43:31] <sep> on the vps you must allow your homeserver to relay. (i normaly do that with smtp auth)[10:44:13] <wizbit> what are the smtp settings for main.cf ?[10:44:13] <sep> if you have a static ip at home you can allow on ip.[10:44:22] <wizbit> home has dynamic ip[10:44:36] *** olegfusion has quit IRC[10:45:25] <sep> http://www.postfix.org/SASL_README.html[10:45:31] <sep> i like the dovecot method[10:47:06] <wizbit> bloody heck this looks like a mission to setup[10:48:43] <wizbit> is there a easier way to just put something in main.cf and allow the dynamic hostname allow relay, probably not as secure but it doesnt matter too much[10:48:51] <sep> the webpage have 4 alternative ways to do it. just do the postfix method. unless you allready have a imap server configured[10:49:09] <wizbit> the vps runs postfix and dovecot[10:49:39] <sep> so it sould be a piece of cake[10:52:08] <wizbit> does that mean my homebox will be requested for a password everytime i send a email via mutt?[10:55:14] <wizbit> sep: http://www.cyberciti.biz/faq/postfix-smtp-authentication-for-mail-servers/[11:22:00] *** davlefouAMD has joined #postfix[11:38:01] *** davlefou has quit IRC[11:38:03] *** err-or has quit IRC[11:41:46] *** err-or has joined #postfix[11:51:57] *** davlefou has joined #postfix[11:52:54] <sep> wizbit, it means you must configure your sasl auth in your homebox postfix smarthost configuration[12:23:20] <wizbit> hmmm i cant figure this out[12:23:38] <wizbit> my postfix uses dovecot for sasl[12:23:45] <wizbit> these settings do nothing: http://www.freelock.com/kb/postfix-relayhost[12:24:00] <wizbit> i think they only work with cyrus[12:24:16] <wizbit> so how can i get dovecot to read a simple username and password file[12:25:03] *** BlackDex_ has joined #postfix[12:26:53] <BlackDex_> Hello.. Is it possible to have mails which come from localhost use a different outgoing IP than mails which are relayed by authenticated users?[12:27:02] *** Cromulent has quit IRC[12:29:21] *** BlackDex_ is now known as BlackDex[12:35:13] *** davlefou_ has joined #postfix[12:38:49] *** davlefou has quit IRC[12:43:38] *** olegfusion has joined #postfix[12:52:51] *** RadoQ has quit IRC[12:54:09] <sep> BlackDex, normaly i'd use virtual and separate servers for that. altho it is possible to run separate instances of postfix i think the added overhead of a additional vm host is better then the added complexity and the non standard solution.[13:01:04] <wizbit> can cyrus and dovecot work along with eachother?[13:01:14] <BlackDex> Well.. i don't have an option for vm's[13:01:51] <wizbit> dovecot doesnt seem to have a way of storing a simple text file with username and password so it will automatically auth a remote smtp server[13:05:58] *** fraff has joined #postfix[13:07:32] *** [diablo] has joined #postfix[13:08:38] *** donmichelangelo has quit IRC[13:09:13] *** donmichelangelo has joined #postfix[13:13:52] <fraff> !welcome[13:13:52] <knoba> fraff: "welcome" : Welcome to #postfix! If you're new here, or to IRC, first read the channel topic (/topic). It has important instructions on how to ask good questions. You will get more and better help if you follow those instructions. Good Luck![13:17:50] *** staticsafe-znc has quit IRC[13:18:15] *** jaybe has quit IRC[13:18:38] *** midnightmagic has quit IRC[13:18:40] *** staticsafe has quit IRC[13:19:03] *** kenyon has quit IRC[13:19:03] *** Ulver has quit IRC[13:19:03] *** crane has quit IRC[13:19:27] *** gencha has quit IRC[13:19:28] *** tapout has quit IRC[13:20:29] *** gencha has joined #postfix[13:20:30] *** jaybe has joined #postfix[13:20:45] *** staticsafe-znc has joined #postfix[13:20:54] *** kenyon has joined #postfix[13:21:08] *** staticsafe has joined #postfix[13:21:55] *** tapout has joined #postfix[13:22:31] *** Ulver has joined #postfix[13:22:50] *** crane has joined #postfix[13:23:42] *** midnightmagic has joined #postfix[13:25:29] *** Zelest has quit IRC[13:25:38] *** Zelest has joined #postfix[13:27:17] *** TomHome has quit IRC[13:35:58] *** davlefou__ has joined #postfix[13:36:08] <fraff> hi all, I need to use sender_dependent_relayhost_maps but it's overriden by a entry in transport_maps, is it possible to avoid that ?[13:39:31] *** davlefou_ has quit IRC[13:48:23] *** tuxick has quit IRC[13:55:09] *** Tuxick has joined #postfix[13:56:36] *** wdp has quit IRC[14:06:23] *** Section1 has joined #postfix[14:09:35] <lunaphyte_> !tell fraff welcome[14:09:35] <knoba> fraff: "welcome" : Welcome to #postfix! If you're new here, or to IRC, first read the channel topic (/topic). It has important instructions on how to ask good questions. You will get more and better help if you follow those instructions. Good Luck![14:15:32] *** mattcen has quit IRC[14:16:28] *** robinho86 has joined #postfix[14:16:43] *** mattcen has joined #postfix[14:23:30] *** MaximusColourum has joined #postfix[14:39:14] *** erob has joined #postfix[14:39:27] *** davlefou__ has quit IRC[14:51:38] *** davlefou__ has joined #postfix[14:53:17] *** erob has quit IRC[14:53:52] *** erob has joined #postfix[14:55:03] *** erob has quit IRC[14:56:06] *** davlefouAMD has quit IRC[15:04:36] *** davlefouAMD has joined #postfix[15:09:14] *** wizbit has quit IRC[15:14:13] <fraff> lunaphyte_: what's wrong with my question ?[15:16:10] <lunaphyte_> sigh[15:16:16] <lunaphyte_> read the instructions. follow them.[15:17:58] <rob0> fraff, postconf.5.html#sender_dependent_relayhost_maps : "This information is overruled with ..."[15:18:50] *** Bronze has joined #postfix[15:19:49] <fraff> rob0: yes, I've read that, that's why my question is: "is it possible to override transport(5) with sender_dependent_relayhost_maps ?"[15:20:31] <rob0> Yes, and if a feature is not documented, it's safe to assume it is not implemented.[15:20:39] <fraff> It may be with multiple instance anyway, but I don't want to do that, is there an other solution ?[15:21:05] <rob0> Since I don't know what you want to do, I can't recommend ways to do it.[15:22:25] *** nbg has quit IRC[15:26:01] *** FreshFeesh has joined #postfix[15:28:06] *** FreshFeesh has quit IRC[15:36:42] *** davlefou_ has joined #postfix[15:37:33] <fraff> rob0: hope this is more readable http://pastebin.com/QADs2zG0[15:38:48] <lunaphyte_> if you ignore the instructions for getting help, most people will probably just ignore you[15:40:11] *** davlefou__ has quit IRC[15:40:27] <rob0> !relay_host[15:40:27] <knoba> rob0: Error: "relay_host" is not a valid command.[15:41:25] <fraff> rob0: yes, "relayhost", sorry[15:41:37] <rob0> All I can suggest is that your setup seems unnecessarily complicated, but I guess that's typical when MSexchange is part of the requirements.[15:42:00] <rob0> I'd focus on fixing the content filter, improving its performance.[15:43:16] <fraff> do you mean that antispam.lan.domain.tld should be faster ?[15:44:35] *** FreshFeesh has joined #postfix[15:47:00] <rob0> right, 30 minutes for an email is absurd[15:48:03] <lunaphyte_> there was someone in here not long ago with a system that had a huge backlog in the mail queue and was taking forever to process messages. was that you?[15:48:52] <fraff> no, wasn't me[15:49:06] *** FreshFeesh has left #postfix[15:49:13] *** Poetics has joined #postfix[15:49:27] <lunaphyte_> oh, well, regardless, for further help, follow the instructions in the factoid that was shared with you.[15:53:31] *** Poetics has quit IRC[15:53:59] <fraff> It's gonna be faster to get an answer somewhere else than understand what's wrong with ma question (specialy when the answer is: get a faster machine)[15:54:09] <fraff> have a good day/evening[15:54:13] *** fraff has left #postfix[15:54:30] <lunaphyte_> it's so weird.[15:55:01] <lunaphyte_> am i honestly missing something?[15:56:46] *** wdp has joined #postfix[15:57:49] <rob0> 30 minutes does not mean "get a faster machine", it means "fix what's broken".[15:58:14] <rob0> anyway, new /ignore material identified.[15:59:16] <lunaphyte_> i'd like to understand where the disconnect was for him - why it seems he didn't understand that he should read the topic[16:00:09] *** donmichelangelo has quit IRC[16:00:37] *** donmichelangelo has joined #postfix[16:05:13] <rob0> Well, I kind of see that he was asking not how to solve a problem with Postfix, but how to implement a feature which does not exist.[16:05:37] <rob0> If you're not having a Postfix problem, logs and configs don't help.[16:06:13] <lunaphyte_> it sounded to me like yet another instance of trying to solve a problem the wrong way.[16:06:28] <rob0> indeed, very much so[16:07:02] <lunaphyte_> logs and configs help me see what the actual problem is.[16:07:48] <rob0> "I have a crappy and broken proprietary antispam thing, in front of a crappy MSexchange backend. I put a Postfix in front of the mess. How can I make Postfix fix the problems with the junk behind it?"[16:07:59] <lunaphyte_> :)[16:08:25] <rob0> The whole thing reeks of lack of Clue.[16:08:37] <lunaphyte_> yeah[16:09:10] <lunaphyte_> this whole smtp[16:09:12] <lunaphyte_> meh[16:09:17] <rob0> Postscreen and amavisd-new would have solved the problem and made the workaround unnecessary.[16:09:26] *** freezey has joined #postfix[16:09:29] <lunaphyte_> this whole smtp_generic_maps exercise i've been doing has been fun.[16:09:38] <rob0> (But that requires some clue to setup and maintain.)[16:11:41] <lunaphyte_> i'm in the final stretch, which is the regex part :\[16:12:35] <rob0> hehe[16:13:43] <lunaphyte_> there are these sort of recursive groupings, for lack of a better term, which make this is challenge for my brain.[16:14:37] <lunaphyte_> one "internal" domain name is valid on the internet, so should only have rewriting done if other internal domain names are used.[16:14:41] *** jarif has quit IRC[16:14:53] *** Cromulent has joined #postfix[16:15:48] <lunaphyte_> and then i'm not to my next crusade, which will be [hopefully] rewriting particular envelope senders hitting submission which are frequent queue clogging offenders.[16:16:24] *** jarif has joined #postfix[16:17:27] *** [diablo] has quit IRC[16:37:28] *** davlefou__ has joined #postfix[16:38:50] *** sputnik has quit IRC[16:39:47] *** chrisp15 has quit IRC[16:41:10] *** davlefou_ has quit IRC[16:53:33] *** sputnik has joined #postfix[16:54:07] *** freezey has quit IRC[16:54:42] *** freezey has joined #postfix[16:59:11] *** freezey has quit IRC[17:00:40] *** freezey has joined #postfix[17:06:48] *** fabian` has joined #postfix[17:24:11] *** kiri has quit IRC[17:25:59] *** kiri has joined #postfix[17:27:44] *** Motoko has joined #postfix[17:36:32] *** steven4455 has joined #postfix[17:37:10] *** UQlev has joined #postfix[17:37:52] *** davlefou_ has joined #postfix[17:38:56] *** UQlev has quit IRC[17:40:18] *** Cromulent has quit IRC[17:40:56] *** davlefou__ has quit IRC[17:41:08] *** RadoQ has joined #postfix[17:59:06] *** MacWinner has joined #postfix[17:59:29] *** hnperez has joined #postfix[17:59:49] *** hnperez has quit IRC[18:01:48] *** alnr has quit IRC[18:02:28] *** freezey has quit IRC[18:03:02] *** freezey has joined #postfix[18:07:18] *** freezey has quit IRC[18:09:05] *** hallamigo has joined #postfix[18:16:41] <lunaphyte_> i'm not understanding the if endif regexp stuff.[18:18:06] <lunaphyte_> how do i say if /^(.*)@.*\.example\.com$/ but not /^(.*) at foo\ dot example\.com$/ then $1 at example dot com?[18:21:18] <lunaphyte_> ah, maybe i have it[18:22:27] *** kiri has quit IRC[18:23:00] *** kiri has joined #postfix[18:27:16] <jelly> that one might even be doable with negative lookahead or lookbehind instead of two checks[18:28:25] *** alnr has joined #postfix[18:29:40] *** SvenOostenbrink has joined #postfix[18:30:08] *** donmichelangelo has quit IRC[18:30:41] *** donmichelangelo has joined #postfix[18:32:15] *** kiri has quit IRC[18:34:39] *** davlefouAMD has quit IRC[18:35:01] <lunaphyte_> well, i had the order backwards, but how do you mean?[18:36:09] <lunaphyte_> right now i'm doing this: http://dpaste.com/1428199/ , which seems to work[18:37:40] *** kiri has joined #postfix[18:38:28] *** davlefou__ has joined #postfix[18:41:24] *** davlefou_ has quit IRC[18:42:02] *** davlefouAMD has joined #postfix[18:42:50] *** cali has quit IRC[18:44:26] *** sandvige has joined #postfix[18:45:34] *** freezey has joined #postfix[18:49:56] *** cali has joined #postfix[19:11:36] *** kiri has quit IRC[19:14:39] *** kiri has joined #postfix[19:19:26] *** sharky has joined #postfix[19:23:04] *** freezey has quit IRC[19:23:37] *** freezey has joined #postfix[19:26:57] *** freezey_ has joined #postfix[19:27:45] *** freezey has quit IRC[19:28:53] <adaptr> does every regex have to match the entire input key ?[19:29:45] <patdk-wk> regex assumes .* at the start and end, unless you use ^ and $[19:30:23] <adaptr> *in the context of postfix regex lookup tables*, does every regex used ( in the matching and if parts) have to match the ENTIRE input?[19:30:43] <adaptr> ACTUAL, NORMAL, PCRE regexes do not. they are always substring matches[19:31:19] <adaptr> in which you CAN anchor the beginning and end, but it's by no means required[19:31:21] <rob0> The expression '/./' matches everything.[19:31:29] <adaptr> excellent.[19:31:49] <patdk-wk> just /something/ matchs just fine like regex should[19:31:54] <adaptr> I would then shorten your if to if example.com if ! foo then ..[19:32:06] <adaptr> because I am in favour of readability before anything else[19:33:31] <rob0> but ... 'foo' will match things like 'food.com'[19:37:11] <adaptr> no, it won't. you've already matched example.com[19:39:36] *** davlefou__ has quit IRC[19:40:23] *** davlefou__ has joined #postfix[20:01:10] *** Hydrazine has quit IRC[20:14:17] <jelly-home> lunaphyte_: something like /^(.*)@.*(?<!foo)\.example\.com$/[20:14:52] <jelly-home> but that will unmatch jelly at barfoo dot example.com as well[20:16:39] <jelly-home> sorry, my http://store-xkcd-com.myshopify.com/products/i-know-regular-expressions t-shirt has just been washed[20:19:07] <adaptr> jelly-home: so include the @[20:19:14] <adaptr> or exclude the dot.[20:19:47] <jelly-home> or do a lookahead instead /^(.*) at ((?!foo)| dot *)\.example\.com$/[20:20:15] <jelly-home> no, that does not work[20:20:18] <adaptr> excluding the dot is less painful[20:39:44] *** davlefou_ has joined #postfix[20:42:14] *** davlefou__ has quit IRC[20:43:30] *** freezey_ has quit IRC[20:44:02] *** freezey has joined #postfix[20:49:01] *** freezey has quit IRC[20:52:43] *** sandvige has quit IRC[20:54:09] *** jackbravo has joined #postfix[20:54:21] <jackbravo> !getting_help[20:54:21] <knoba> jackbravo: "getting_help" : before asking your question, provide a pastebin which includes relevant log data and your config. see !pastebin, !relevant_logs and !showconfig for instructions on doing this.[20:54:49] <jackbravo> !relevant_logs[20:54:49] <knoba> jackbravo: "relevant_logs" : Relevant logs are mail.* syslog Postfix logs (NOT verbose, see !verbose) which show the entire handling of a single mail which illustrates the issue with which you want help. Random selections from your mail log might not do. IMAP/POP3 daemons and external delivery agents typically log to the same facility (mail); those are usually not relevant here.[21:03:55] *** trusktr has joined #postfix[21:06:00] *** freezey has joined #postfix[21:14:35] *** master_of_master has joined #postfix[21:17:31] *** master_o1_master has quit IRC[21:19:59] *** zerick has joined #postfix[21:20:48] *** amospalla has quit IRC[21:29:28] *** trusktr has quit IRC[21:31:35] *** trusktr has joined #postfix[21:31:39] *** olegfusion has quit IRC[21:39:38] *** davlefou__ has joined #postfix[21:41:37] *** trusktr has quit IRC[21:42:37] *** pyco has quit IRC[21:42:54] *** davlefou_ has quit IRC[21:42:59] *** pyco has joined #postfix[21:43:00] *** pyco has joined #postfix[21:56:37] <dhammond> does header_checks filter outbound mail or just inbound?[21:57:54] <patdk-wk> what is the difference?[21:57:57] <patdk-wk> it filters mail[21:58:04] <patdk-wk> there is no such thing as outbound or inbound[21:58:12] <lunaphyte_> what is "outbound" mail?[21:59:11] <lunaphyte_> outbound mail must be inbound mail that is also outbound mail?[21:59:35] <lunaphyte_> so inbound mail is mail that is only inbound, but not outbound, and outbound mail is mail that is both inbound and outbound?[21:59:37] <patdk-wk> not sure why you want to filter email that is on it's way out[21:59:44] <patdk-wk> normally yo uwant to filter it on it's way in[21:59:47] <dhammond> hm maybe i just dont understand how the process works from start to finish. Im trying to prevent my server from attempting to make connections (that subsequently fail) to example.com when emails are sent from my application during unit testing[21:59:54] <patdk-wk> and since all email comes IN and goes OUT[22:00:05] <lunaphyte_> oh, was that you who asked the ther day?[22:00:09] <dhammond> yeah[22:00:09] <lunaphyte_> *the other[22:00:23] <lunaphyte_> did you see my reply?[22:00:37] <dhammond> still havent figured it out. i thought header_checks might work but I still saw traffic to example.com in the tcpdump file[22:00:37] <lunaphyte_> step one is simply to stop generating such mail.[22:00:42] <dhammond> let me check[22:00:46] <lunaphyte_> step two is to simply reject such mail.[22:01:07] <lunaphyte_> check_recipient_access[22:01:24] <lunaphyte_> but let's focus on step one.[22:01:37] <lunaphyte_> what is generating this undesirable mail?[22:03:11] <dhammond> hmm my application sends confirmation emails after a user registers. I dont want to use real email addresses for this so I unit test with userXXX at example dot com[22:03:55] <lunaphyte_> nice to see someone being courteous :)[22:04:01] <dhammond> Im auditing the external network traffic and example.com always shows up. I dont want to whitelist the IP if possible but rather just "drop" or silently ignore the attempt to send the email[22:04:06] <waldi> dhammond: use your own test-domain[22:04:18] <lunaphyte_> that would be the best, yeah.[22:04:20] <waldi> and don't allow anything else sent over the test-setup[22:05:12] <patdk-wk> heh, that sounds simple[22:05:29] <patdk-wk> check_sender_access .... (example.com DROP)[22:05:42] <patdk-wk> actually, that would be check_recipient_access[22:06:07] <dhammond> patdk-wk: let me read up on that. that sounds like the winner[22:06:20] <patdk-wk> it's the simplist, not saying it's the best :)[22:06:29] *** Section1 has quit IRC[22:06:41] <patdk-wk> pesonally, I would route them all to a single email account[22:06:48] <dhammond> waldi: do you mean a real external domain?[22:06:50] <patdk-wk> so if I wanted to do more indepth testing, I could[22:07:13] <lunaphyte_> it doesn't have to be valid on the internet, no.[22:07:26] <lunaphyte_> just set up an address for testing.[22:08:00] <dhammond> ah that might work too[22:08:22] <dhammond> ok well ive got some reading to do now. hopefully this works. thank you guys :)[22:09:33] *** KaiForce has joined #postfix[22:15:51] *** freezey has quit IRC[22:17:41] *** thelamest has quit IRC[22:18:17] *** mintux has joined #postfix[22:18:29] <mintux> if i want to have dkim i should install amavis ?[22:21:14] *** SvenOostenbrink has quit IRC[22:22:51] <patdk-wk> if yo uwant[22:22:54] <patdk-wk> or some other dkim package[22:24:46] <mintux> amavis is so hard and when it enable i can not send email to gmail or ...[22:27:26] <patdk-wk> do what?[22:27:39] <patdk-wk> well, either learn how to use amavis[22:27:42] <patdk-wk> or learn something else[22:27:55] <patdk-wk> or, if you wish to complain, don't do anything, just don't do it here[22:27:57] <mintux> if i can install only dkim it's so suitable for me[22:28:10] <patdk-wk> !dkim[22:28:10] <knoba> patdk-wk: "dkim" : DomainKeys Identified Mail (DKIM) is a method for email authentication that allows an organization to take responsibility for a message in a way that can be validated by a recipient. this is typically implemented in postfix by means of a milter. alternatively, existing content filters (e.g. amavis) may also have their own implementation mechanism.[22:28:17] <patdk-wk> hmm[22:28:47] <mintux> i found something called opendkim that seems doesn't need to use amavis[22:30:17] *** oGG has joined #postfix[22:31:15] *** freezey has joined #postfix[22:32:12] *** trusktr has joined #postfix[22:35:12] <Shinobi> My mail keeps timing out on 25 for all major email carriers... hotmail, yahoo, gmail[22:35:22] <Shinobi> Not on a blacklist[22:35:24] <waldi> and it does not for others?[22:35:49] <waldi> can you connect mail.thinkmo.de on port 25? (my server, so i know for sure it works)[22:35:54] <Shinobi> waldi: I don't know where else to end it to besides one of my emails to test[22:36:10] <Shinobi> ok[22:37:57] <Shinobi> waldi: do I just send email to this domain? How do I connect otherwise?[22:38:06] <waldi> nc, telnet[22:39:16] <Shinobi> no[22:39:33] <waldi> well. then you have to ask your isp[22:42:00] *** davlefou_ has joined #postfix[22:42:08] *** amospalla has joined #postfix[22:42:48] *** davlefou__ has quit IRC[22:50:21] <Shinobi> so if I do: telnet mydomain.com 25 ... does it look up the mx record since it's port 25 or does it just look up a cname or a record[22:50:37] *** KaiForce has quit IRC[22:50:56] <lunaphyte_> it resolves the hostname. that's it.[22:51:56] <waldi> Shinobi: i doubt it. mydomain.com does not have a cname[22:57:10] <Shinobi> waldi: I think it is a routing issue for me to connect to your server. It says network unreachable[22:57:59] <waldi> Shinobi: then try port 587[22:59:05] <Shinobi> ok[22:59:07] <Shinobi> works[22:59:33] <Shinobi> got a 220[23:00:20] <waldi> so your provider blocks port 25[23:00:50] <Shinobi> can I run a tracert by port?[23:03:04] <Shinobi> yes, it's blocked. this makes testing impossible[23:03:08] <Shinobi> from here[23:05:27] <waldi> so, remove the block[23:05:30] <adaptr> or, indeed, ever sending mail.[23:06:00] *** sharky_ has joined #postfix[23:08:37] <mintux> why i got refused to talk to me: [TS03] All messages from <my server ip> will be permanently deferred when i want to send email to yahoo my ip is new and has not sent spam from it[23:08:53] *** sharky has quit IRC[23:08:59] <waldi> mintux: please show the _complete_ message[23:09:01] <adaptr> !tell mintux getting_help[23:09:01] <knoba> mintux: "getting_help" : before asking your question, provide a pastebin which includes relevant log data and your config. see !pastebin, !relevant_logs and !showconfig for instructions on doing this.[23:09:06] <mintux> ok[23:10:01] <mintux> http://codepad.org/KpEBTeVs[23:10:34] *** Alexwijn has joined #postfix[23:11:47] <mintux> waldi: http://codepad.org/KpEBTeVs[23:16:09] <adaptr> mintux: and did you do what they suggested ?[23:17:08] <mintux> i setup dkim and reverse DNS and spf record but i don't sent the support form[23:17:39] <mintux> yet[23:19:04] *** GNU\colossus has quit IRC[23:20:15] <waldi> "This message indicates Yahoo is seeing a high volume of emails from your IP address, which is a characteristic of unsolicited, bulk emailing."[23:20:26] <thumbs> wee[23:22:08] <mintux> you mean this is possible sent bulk email from this ip before i take it?[23:23:22] <Alexwijn> Hi[23:23:28] <Alexwijn> I'm trying to setup Postfix and Dovecot...But since I send a email from my gmail to the test account on that server[23:23:43] <Alexwijn> And I keep getting spammed from Google Mail server: http://pastebin.com/xX3rRmDR[23:26:17] <Alexwijn> So what I can do to fix this?[23:27:14] *** mintux has quit IRC[23:29:26] <Shinobi> does gmail support mta - mta tls[23:30:01] <thumbs> oh, it's the TLS fellow again[23:30:10] *** donmichelangelo has quit IRC[23:30:23] <thumbs> as far as your answer, it depends, but they'll advertise it.[23:30:48] *** donmichelangelo has joined #postfix[23:31:31] <Alexwijn> Does nobody regonize my issue?[23:31:36] *** GNU\colossus has joined #postfix[23:33:34] <adaptr> Alexwijn: you could: 1. block the IP, or 2. ask Google to fix it[23:33:52] <Alexwijn> Why would I block Google Mail Server?[23:34:13] <Alexwijn> Or is this just one mail server of them are is so anoying?[23:34:23] <adaptr> unless, of course, asclub-management.eu is you[23:34:28] <Alexwijn> that is me[23:34:38] <adaptr> then you caused the flood[23:34:49] <Alexwijn> How to fix that?[23:35:01] <adaptr> sigh[23:35:16] <adaptr> are you sure you should be running a public MTA if you don't understand what is happening ?[23:35:30] <adaptr> basic causes: bad hostname/myhostname/mydomain settings, bad routing[23:35:35] <Shinobi> thumbs: just seeing if i can test w/o having to buy service just yet[23:35:36] <adaptr> bad DNS somewhere[23:35:41] <adaptr> "whatever"[23:35:44] <Alexwijn> ah[23:35:50] <jackbravo> !pastebin[23:35:50] <knoba> jackbravo: "pastebin" : (#1) see !paste, or (#2) a pastebin site lets you easily share logs and configuration. Examples are dpaste.org, fpaste.org, or pastebin.ca. Please avoid ad-supported sites such as pastebin.com if possible.[23:35:50] <Alexwijn> if that's fixed then it should be over[23:36:03] <adaptr> Alexwijn: at this point, you don't know. you need to find out.[23:36:20] *** freezey has quit IRC[23:36:25] <Alexwijn> Because I do have a bad hostname at the moment, I'm waiting till DNS to propagate[23:36:43] *** freezey has joined #postfix[23:37:24] <adaptr> DNS. does. not. PROPAGATE.[23:37:36] <adaptr> urgh. soo tired[23:37:42] * thumbs propagates to adaptr[23:37:49] <adaptr> that sounds nasty[23:37:57] <adaptr> do it again[23:38:35] <Alexwijn> Never heard of propagation dns?[23:39:00] <thumbs> Alexwijn: DNS doesn't propagate[23:39:13] <Alexwijn> So DNS propagation does not exisit?[23:39:18] <adaptr> it does not.[23:39:23] <Alexwijn> lol[23:39:36] *** TRUNASUCI has joined #postfix[23:40:23] <Alexwijn> Let me write some wiki info here for you[23:40:44] *** davlefou__ has joined #postfix[23:40:48] <thumbs> Alexwijn: please don't.[23:40:50] <adaptr> please don't[23:40:52] <Alexwijn> hm[23:40:54] *** mintux has joined #postfix[23:40:59] <thumbs> Alexwijn: DNS expires, it doesn't propagate[23:41:10] <adaptr> this kind of attitude doesn't bode well for your troubleshooting skills, either[23:41:32] <Alexwijn> I know DNS Propagation is a rather inaccurate term used to describe the process of DNS resolution[23:41:35] <Alexwijn> But still[23:41:39] <Alexwijn> Everyone using it[23:41:55] <adaptr> no, it has nothing to do with DNS resolution. at all.[23:42:28] <Alexwijn> Right[23:43:05] <Alexwijn> It is just a term that everyone use[23:43:08] <Alexwijn> That's it[23:43:11] <adaptr> and ?[23:43:15] <adaptr> it is incorrect.[23:43:33] <Alexwijn> For you maybe[23:43:45] <adaptr> this is not open to discussion, Alexwijn . please drop it.[23:43:59] *** davlefou_ has quit IRC[23:44:42] *** mintux has left #postfix[23:44:54] *** Alexwijn has left #postfix[23:45:16] <adaptr> them egos. so fragile![23:45:33] <adaptr> thumbs: propagate me again, come on.. you know you want to.[23:45:41] * thumbs props[23:45:52] <adaptr> eeewww that sounds downright zexual[23:54:19] *** Alexwijn has joined #postfix[23:55:45] <Alexwijn> If my mail server received my test email but at the queue once receive it says that the connection is refused with my hostname[23:55:50] <Alexwijn> What should I do[23:55:54] <Alexwijn> (The hostname is working now)[23:56:10] <thumbs> Alexwijn: show the literal log entries[23:56:15] <Alexwijn> To what is he connecting? To Dovecot?[23:56:24] <thumbs> that demonstrate your issue[23:57:00] <Alexwijn> ok...?