[00:00:44] <donald> and then they get a connection refused. Thats also ok.
[00:00:57] <pickcoder> they get a timeout
[00:01:36] *** Knoedel2 has quit IRC
[00:01:53] *** keffer has joined #postfix
[00:02:03] <jeev> yea it's still in the queue
[00:02:12] <jeev> shit man, proper DNS, spf, dkim..
[00:02:16] <jeev> have sent them mail before
[00:02:26] *** LordLamer has quit IRC
[00:02:36] <lennard> being a postmaster really is a shitty job these days
[00:02:54] <jeev> certificate verification failed for edge1.choicepoint.net[66.241.37.242]:25: untrusted issuer /C=US/O=Entrust.net/OU=www.entrust.net/CPS incorp. by ref. (limits liab.)/OU=(c) 1999 Entrust.net Limited/CN=Entrust.net Secure Server Certification Authority
[00:02:56] <jeev> is that my fault
[00:02:57] <jeev> or theirs ?
[00:03:25] <adaptr> they don't trust the issuer
[00:03:35] <jeev> i dont? or THEY dont
[00:03:40] <adaptr> is it backed by verisign or somesuch ?
[00:03:45] <jeev> mine is self signed
[00:03:53] <adaptr> then no, they don't trust it
[00:03:54] <lennard> jeev: is it a smtpd-log or a smtp-log?
[00:04:05] <jeev> ok, cause it looks like mine is denying it
[00:04:08] <jeev> is what a smtp-log
[00:04:14] <jeev> that's what my maillog shows me
[00:04:43] <dam85> thanks everybody
[00:04:44] <dam85> bye bye
[00:04:48] *** dam85 has quit IRC
[00:05:00] <adaptr> I have to concur with lennard on this one
[00:05:03] <sobriquet> hey now... dont hate on spammers... they're only trying to make a living.  Everyone is entitled to try to make a living.
[00:05:08] <adaptr> he was very good
[00:05:08] <lennard> and I bet it has smtp(d)?[0-9]+: in front of it
[00:05:22] <adaptr> sobriquet: spammers should be executed withotu parole
[00:05:30] <adaptr> you're barking up the wrong tree here
[00:05:31] <lennard> heh
[00:05:37] <jeev> you think the 20 dollar godaddy cert is good enough ?
[00:05:42] <sobriquet> pretty hard to earn parole when you're dead adaptr
[00:05:47] <lennard> I'd like to see executing with parole
[00:05:52] <adaptr> jeev: if theirs is backed by a roto CA, yes
[00:05:56] <adaptr> *root CA
[00:06:42] *** LordLamer has joined #postfix
[00:07:16] <jeev> hmm
[00:07:49] *** sobriquet has quit IRC
[00:08:18] <adaptr> walk the tree, I think openssl can show you the whole lineage
[00:08:19] *** sjmudd has joined #postfix
[00:09:17] <jeev> theyr'e there own CA
[00:09:42] <adaptr> how do you know ?
[00:11:37] <pickcoder> I doubt they are
[00:11:54] *** sjmudd has quit IRC
[00:12:51] <jeev> https://www.godaddy.com/gdshop/ssl/ssl.asp
[00:12:55] <jeev> look at the actual loaded cert
[00:13:41] <lennard> I think go daddy is usually recognized
[00:15:06] *** pulsar has quit IRC
[00:16:46] <pickcoder> Starfield Services
[00:16:53] <pickcoder> is the root CA
[00:17:26] <pickcoder> unless I'm reading it wrong
[00:20:49] <pickcoder> I don't see how they can sell certs for $20 and still have to perform the usual company investigation
[00:20:53] <pickcoder> s/company/ident
[00:21:29] <jeev> oh
[00:21:33] <jeev> hey do the basic stuff
[00:21:37] <jeev> no check
[00:21:41] <jeev> they took the price to $30 !
[00:22:05] <lennard> *ponder*
[00:22:09] <lennard> to drink more beer or not to
[00:22:16] <adaptr> more beer!
[00:22:20] <adaptr> no-brainer
[00:22:37] <jeev> oh well
[00:22:40] <jeev> 3rd attempted passed i thnk
[00:22:42] <jeev> the mailserver took it
[00:22:57] * pickcoder would rather have marqaritas
[00:23:04] <jeev> alcohol is for LOSERS!
[00:23:13] <adaptr> jup, that's me
[00:23:13] <cafuego> liar
[00:23:20] <jeev> i dont drink
[00:23:21] <jeev> ;)
[00:24:05] <cafuego> not knowingly, I'm sure
[00:24:15] *** alibo` has quit IRC
[00:24:16] <cafuego> most fruit juice ferments a little bit after opening
[00:24:24] <jeev> ;)
[00:24:36] <pickcoder> hopefully it's not consumed @ 50 proof of more
[00:24:40] <pickcoder> s/of/or
[00:24:52] <cafuego> if it is, death would be a good indicator ;-)
[00:24:56] <pickcoder> (unless it's homemade wine..)
[00:26:22] *** quieteyes has left #postfix
[00:26:50] * pickcoder has always wondered if real fruit juice is in Mt Dew due to its amazing open-bottle shelf life
[00:27:26] <cafuego> pickcoder: I can confirm real fruit juice doesn't taste like crap, unlike mt dew
[00:27:26] <jeev> hey
[00:27:28] <jeev> you guys watch
[00:27:31] <jeev> Little People, Big World
[00:27:32] <jeev> those little guys
[00:27:38] <jeev> they used to take up half of the basketball courts at the YMCA
[00:27:39] <jeev> for like a year
[00:27:42] <jeev> i used to always get mad at them
[00:27:46] <pickcoder> cafuego: meh.. I do the Dew.
[00:27:48] <jeev> but on their tv show, i like watching it
[00:27:51] <jeev> they ALWAYS drink mr. deww
[00:27:54] <jeev> who the fuck drinks mt dew?
[00:27:57] <jeev> that was when we were <10
[00:28:06] <jeev> i dont thin anyone in california drinks mountain dew
[00:28:10] <lennard> I don't know, I think they don't sell it here in .nl
[00:28:15] <cafuego> i think it's somewhat more popular in the states than elsewhere
[00:28:28] <pickcoder> Sundrop is popular here in the south
[00:28:30] <lennard> except after going through customs at the airport - and then only some of the time
[00:28:37] <pickcoder> I need the caffeine
[00:28:40] <cafuego> seems to mainly be water + sugar
[00:28:49] <cafuego> pickcoder: for caffeine there is coffee
[00:29:01] <pickcoder> no one can make coffee here and I don't like making it
[00:29:09] <pickcoder> besides it gives me heartburn
[00:29:27] <pickcoder> Maybe one day I'll try Bawls
[00:29:28] <cafuego> nexium ftw
[00:29:56] <cafuego> pickcoder: if you're in the us i'll have to agree on "no one can make coffee here" ;-)
[00:30:19] <lennard> I hear the huge mugs we have at the office are a result of that
[00:30:23] <pickcoder> my wife worked at a coffee bar so I usually let her make it
[00:30:25] <jeev> oh my god i'm missing TOM LEYKUS
[00:30:54] <lennard> americans cant make proper strength coffe, therefor they need huge mugs, then we get to use those and put it full of normal coffee
[00:31:07] <cafuego> lennard: Mugs are fine, as long as you fill them with say 8 shots of espresso and NOT a single shot of grey water.
[00:31:44] <lennard> cafuego: what countrys defenition of espresso? :)
[00:31:50] <cafuego> lennard: italy
[00:31:57] <lennard> hmm
[00:32:03] <lennard> thats porbably overdoing it a bit :)
[00:32:13] <cafuego> what we call a "short black" in .au
[00:32:16] <lennard> but, yes, definatly better than gray water
[00:32:26] <lennard> grey*
[00:32:33] <lennard> I think
[00:32:42] <lennard> ugh, too tired to properly speak english :P
[00:32:57] <cafuego> Zeg maar gewoon grijs dan.
[00:33:28] <lennard> suer, that works :P
[00:39:10] <lunaphyte> coffee is vile, and tom leykis is an ass-clown. :)
[00:39:40] <lennard> meh, both coffee and beer are just facts of life
[00:39:50] <lennard> although beer probably is the only one I drink too much of
[00:40:12] <lunaphyte> agreed, unfortunately.  beer falls in the same category for me as coffee.
[00:40:44] <lennard> which means you don't drink too much of either?
[00:41:05] <adaptr> I'm getting confused - beer also invokes ass-clowniness ?
[00:41:53] <lunaphyte> lennard: none for me, thanks.
[00:42:06] <lennard> heh
[00:42:27] <lennard> principe, common sense, disgust, or allergy?
[00:42:51] <lunaphyte> hmm.  2 and 3, i think mostly.
[00:42:59] <lennard> mmkay
[00:43:14] <lennard> I sometimes consider not drinking beer
[00:43:30] <lennard> its just easier to drink it though :P
[00:43:55] <cafuego> and a fair spot cheaper than say martinis
[00:43:59] <adaptr> instead of considering ?
[00:44:20] *** pulsar has joined #postfix
[00:44:39] <lennard> drinking is easier than both considering and not drinking :P
[00:45:02] * adaptr sets the topic to #achtervoegsel
[00:45:15] <lennard> ... achtervoegsel?
[00:45:19] <lennard> wth is that about :P
[00:45:31] <adaptr> does this even resemble an english channel anymore ?
[00:45:56] <lennard> dunno, it does seem to be the leading language :)
[00:46:29] <lennard> thats partially because most dutchies secretly consider the language superior over their own, I suspect, though
[00:46:47] <cafuego> I sure don't.
[00:47:01] <adaptr> I'm used to it
[00:47:23] * pickcoder would talk in binary if it wasn't so time consuming
[00:47:28] <cafuego> I speak it every day, but also every day I find english somewhat limiting compared to dutch.
[00:47:46] <lennard> you do?
[00:47:49] <adaptr> pickcoder: I will speak binary way sooner than you speak dutch
[00:47:49] <cafuego> if only those damn new hollanders would all learn it.
[00:47:57] <adaptr> yes, manhattan--
[00:48:15] <pickcoder> adaptr: you can be sure of that too
[00:48:20] <cafuego> that's new amsterdam, i think you'll find
[00:48:32] <adaptr> cafuego: isn't that where most of them..congregate ?
[00:48:40] <lennard> most times when I encounter limits I cannot for the life of me figure out a proper dutch description for an english word
[00:48:53] <lennard> then again - I hardly ever actually *speak* English
[00:48:55] <cafuego> adaptr: no, new holland is more widely known as australia these days ;-)
[00:48:56] <lennard> only type it
[00:49:41] <adaptr> lennard: I speak English every day - to an Indian programmer, whose English is waay worse than my - well, pick any language I don't speak
[00:49:58] <lennard> oh boy
[00:50:05] <lennard> I have people to do that for me
[00:50:25] <lennard> not much better though, they can only relay their best guess-work of the conversation to me :)
[00:50:34] <adaptr> well, he's only 10 feet away, having someone do that for me would be rather inefficient
[00:50:38] *** F6F has quit IRC
[00:50:44] <lennard> ah
[00:50:56] <adaptr> the rest of his team is in Banga valley
[00:51:01] <lennard> the Indians I have 'contact' with all are actually in India :)
[00:52:16] *** cilquirm has joined #postfix
[00:53:41] *** roe has joined #postfix
[00:55:29] *** donald has quit IRC
[00:55:38] *** pickcoder has quit IRC
[00:58:04] *** randra has quit IRC
[00:59:14] *** war9407 has quit IRC
[01:27:12] *** jra has joined #postfix
[01:31:26] *** saurabhb has joined #postfix
[01:35:26] *** s0what has quit IRC
[01:47:40] *** gutocarvalho has quit IRC
[01:53:25] *** deadpigeon has quit IRC
[01:54:20] *** verywiseman has quit IRC
[01:54:42] *** verywiseman has joined #postfix
[01:57:54] *** ming_zym has joined #postfix
[02:00:11] *** pitakill has joined #postfix
[02:06:24] *** boolean has quit IRC
[02:11:51] *** githogori has quit IRC
[02:13:01] *** edman007 has left #postfix
[02:14:32] *** linguini has joined #postfix
[02:16:13] *** felipe_ has quit IRC
[02:16:22] *** verywiseman has quit IRC
[02:21:33] *** verywiseman has joined #postfix
[02:22:20] *** hparker|laptop has joined #postfix
[02:37:49] *** pirho has quit IRC
[02:43:19] *** orzel has left #postfix
[02:47:18] *** xpoint has quit IRC
[02:53:10] *** cilquirm has quit IRC
[02:54:47] *** cilquirm has joined #postfix
[02:55:40] *** hparker has quit IRC
[02:55:45] *** jra has left #postfix
[02:55:49] *** hparker|laptop is now known as hparker
[02:57:06] *** verywiseman has quit IRC
[03:11:47] *** magyar_ has quit IRC
[03:15:16] *** linguini has quit IRC
[03:15:49] *** linguini has joined #postfix
[03:18:08] <linguini> I'm confused; wouldn't most postfix users want postfix built with STARTTLS and AUTH support?
[03:22:13] <lunaphyte> huh?
[03:32:01] *** hparker has quit IRC
[03:40:18] *** hparker has joined #postfix
[03:58:23] <growltiger> not necessarily
[03:58:30] <growltiger> i lived fine without those
[03:59:47] <cite> Good morning.
[04:06:01] *** pulsar has quit IRC
[04:11:47] *** Southron has joined #Postfix
[04:12:57] *** alxgomz has joined #postfix
[04:13:52] <alxgomz> i'm stuck to a postfix server not accepting additionnal connection because all 100 smptd procces are busy..... what should i do?
[04:14:04] *** linguini has quit IRC
[04:14:09] *** roe has quit IRC
[04:21:12] <cite> alxgomz: Perhaps http://www.postfix.org/STRESS_README.html can help you.
[04:22:29] *** _mavrick61 has quit IRC
[04:23:36] *** _mavrick61 has joined #postfix
[04:24:11] <hparker> alxgomz: using any RBLs?
[04:24:33] <alxgomz> yes...
[04:26:15] *** felipe_ has joined #postfix
[04:32:16] *** Niemi has quit IRC
[04:35:46] *** githogori has joined #postfix
[04:36:58] <alxgomz> hparker: rbl can cause this?
[04:42:29] <hparker> Not unless DNS is slow... RBL usually helps cut the amount of connections down
[04:49:57] *** Tapnzout has joined #postfix
[04:56:55] <sahil> read the stress and tuning documents available at the postfix web site.
[05:08:29] *** jens_ has joined #postfix
[05:09:25] *** Tapout has quit IRC
[05:24:36] *** jense has quit IRC
[05:32:06] *** mmcji has joined #postfix
[05:32:42] <mmcji> howdee
[05:32:45] <mmcji> when i am asked for the: "Root and postmaster mail recipient" during install, do I put a regular linux user here?
[05:33:09] <mmcji> i am installing postfix on ubuntu 8.10 server
[05:36:16] *** goldfischli has joined #postfix
[05:37:09] *** goldfisc1li has quit IRC
[06:03:45] *** roe has joined #postfix
[06:11:24] *** Southron has quit IRC
[06:19:06] *** ovb_ has quit IRC
[06:19:10] *** ovb_ has joined #postfix
[06:21:31] *** verywiseman has joined #postfix
[06:24:14] *** hparker has quit IRC
[06:28:31] *** tshine is now known as tshine_zzz
[06:32:19] *** UQlev has joined #postfix
[06:49:50] *** niki has quit IRC
[07:08:04] *** drummerboy has joined #postfix
[07:08:06] <drummerboy> hi
[07:09:47] <drummerboy> i try to install postfix and dovecot on debian 4 and use this howto (which refers to this channel): http://workaround.org/articles/ispmail-etch/index.html.en#checking-the-user-s-maildir (i use the german version)
[07:11:07] <drummerboy> in the section ?Checking the user's maildir? it says there should be a folder /home/vmail/example.com/john (in my case it's a real address)
[07:11:35] <drummerboy> but there?s no folder for the domain, and therefore no subfolders, too
[07:12:31] <drummerboy> this howto says if anything goes wrong i should run ?postfix check?, so i did this
[07:12:50] <drummerboy> it says:
[07:13:00] <drummerboy> postfix/postfix-script: warning: /var/spool/postfix/lib/libnss_compat-2.3.6.so and /lib/libnss_compat-2.3.6.so differ
[07:13:00] <drummerboy> postfix/postfix-script: warning: /var/spool/postfix/lib/libnss_nis-2.3.6.so and /lib/libnss_nis-2.3.6.so differ
[07:13:00] <drummerboy> postfix/postfix-script: warning: /var/spool/postfix/lib/libnss_dns-2.3.6.so and /lib/libnss_dns-2.3.6.so differ
[07:13:05] <drummerboy> what can i do?
[07:13:37] <drummerboy> (i'm relatively new to server administration and linux o:-) )
[07:15:07] <f3ew> copy the files from /lib ionto /var/spool/postfix/lib
[07:20:07] <drummerboy> done. postfix check returns nothing, which seems to be good
[07:20:55] <drummerboy> but was this the cause for the folders to not be there?
[07:21:04] <f3ew> send the user an email
[07:22:27] <drummerboy> done
[07:22:44] <drummerboy> i have 2 accounts in the user-db
[07:22:58] <drummerboy> i sent a mail to both of them
[07:23:29] <f3ew> now see if the maildirs exist?
[07:23:37] *** roe has quit IRC
[07:24:07] <drummerboy> ?ls /home/vmail/? still says nothing
[07:25:25] *** pitakill has quit IRC
[07:26:47] <drummerboy> ah, in the spam-folder of the account from where i sent the mails i have an answer from the server
[07:27:18] *** jonez has quit IRC
[07:27:53] <drummerboy> ?[?] your message could not be delivered [?]?
[07:27:53] <drummerboy> ?<*** at ***** dot de>: unknown user: "***"?
[07:32:43] <f3ew> There you go
[07:32:50] <f3ew> See your logs
[07:33:26] <drummerboy> this one? /var/log/mail.log
[07:33:50] <f3ew> ys
[07:33:51] <f3ew> yes
[07:33:54] <drummerboy> ok
[07:37:04] <drummerboy> trivial-rewrite[32077]: warning: do not list domain *****.de in BOTH mydestination and virtual_mailbox_domains
[07:37:09] <drummerboy> aaaha
[07:37:11] *** UQlev has quit IRC
[07:37:28] <drummerboy> looks like the cause?
[07:38:41] <drummerboy> i don't want to get any normal mails to local system users
[07:39:09] <drummerboy> they should all be delivered to virtual accounts
[07:42:12] *** fury_ has joined #postfix
[07:43:37] *** cilquirm has quit IRC
[07:45:13] <drummerboy> i deleted my domain from mydestination
[07:45:23] <drummerboy> have i to restart something?
[07:47:26] *** Niemi has joined #postfix
[07:48:01] <drummerboy> ok, i did /etc/init.d/dovecot restart and /etc/init.d/postfix restart
[07:48:07] <drummerboy> it seems to work now
[07:48:13] <drummerboy> there are subdirs :)
[07:50:09] *** _fury has quit IRC
[07:56:35] *** Fallenou has joined #postfix
[07:58:37] *** Fallenou has quit IRC
[07:58:41] <drummerboy> what are common periods for a ssl certificate?
[08:07:42] *** |_Knoedel_| has joined #postfix
[08:29:07] *** bhagat has joined #postfix
[08:48:31] *** |_Knoedel_| has quit IRC
[09:06:32] *** |_Knoedel_| has joined #postfix
[09:16:09] *** loddafni1 has joined #postfix
[09:16:35] *** loddafnir has joined #postfix
[09:17:53] *** sophokles has joined #postfix
[09:18:42] <drummerboy> looks like my server works :) thanks
[09:21:23] *** amrit|wrk is now known as amrit|afk
[09:29:18] <tobias-> Is it possible to "bind" postfix sendmail-binary so it uses another from IP?
[09:37:08] *** harobed has joined #postfix
[09:42:36] *** Signum has quit IRC
[09:44:22] *** amrit|afk is now known as amrit
[09:44:41] *** knoba has quit IRC
[09:46:25] *** brancaleone has joined #postfix
[09:48:07] *** knoba has joined #postfix
[09:48:14] <sysmonk> tobias-: the sendmail binary itself?
[09:48:18] <sysmonk> or postfix itself?
[09:52:10] <tobias-> sysmonk: sendmail itself i guess
[09:52:26] <tobias-> sysmonk: which connects directly to a server
[09:52:28] <sysmonk> then i don't think it's possible
[09:52:35] <tobias-> damn
[09:52:40] <sysmonk> unless you want to have a seperate postfix instance
[09:53:03] <tobias-> i'm handling my postfix instance on another IP on my server and I want everything that has to do with mail to go from that IP aswell
[09:53:26] <sysmonk> ah
[09:53:32] <sysmonk> then that's not sendmail binary related
[09:53:37] <sysmonk> !smtp_bind_address
[09:53:38] <knoba> sysmonk: "smtp_bind_address" : a configuration parameter in the main.cf: An optional numerical network address that the SMTP client should bind to when making a connection.
[09:53:39] <sysmonk> tobias-: ^^
[09:53:58] <tobias-> i'll try it ;)
[09:54:00] <tobias-> thx
[09:54:24] <sysmonk> i thought you want postfix to send from one ip, but the 'sendmail' binary itself - from another :)
[09:54:34] <sysmonk> that's not possible without seperate instances
[09:54:47] <sysmonk> but if you want all mail to be sent from some ip - then it's possible with smtp_bind_address
[09:59:04] *** war9407 has joined #postfix
[09:59:31] *** Signum has joined #postfix
[10:06:43] <tobias-> sysmonk: works fine, thanks :)
[10:10:30] *** Testmaster has quit IRC
[10:11:24] *** magyar has quit IRC
[10:11:45] *** magyar has joined #postfix
[10:12:53] *** phnord has joined #postfix
[10:17:45] *** saurabhb has quit IRC
[10:19:26] *** ErikRose_ has joined #postfix
[10:21:01] <ErikRose_> I'm at my wits' end. Does anyone know what to do about a "SASL PLAIN authentication failed: authentication failure" every time I try to authenticate to SMTP over SSL?
[10:22:08] <ErikRose_> I've been frobbing everything I can think of all night, finally figured out that openssl s_client was renegotiating because I used a capital R when typing RCTP TO, but now I'm stuck here.
[10:24:29] <sep> ErikRose_, :)
[10:24:49] <ErikRose_> You again!
[10:24:50] <ErikRose_> ;-)
[10:25:12] <ErikRose_> Thanks; you gave me my first laugh all night. :-)
[10:25:32] <ErikRose_> Ladies and gentlemen, I give you sep, king of the capital R.
[10:25:41] <sep> ErikRose_, postfix in debian run chrooted afaik, are you sure it can reach/read your user database ?
[10:25:54] <ErikRose_> You're full of good ideas.
[10:26:05] <ErikRose_> Let's see here...
[10:26:06] <sep> ErikRose_, tehy just take a while to come to me :)
[10:26:31] <ErikRose_> Well, I've been banging on this for 3 weeks, so have yourself an ego boost.
[10:26:45] <sep> ^^
[10:27:09] <ErikRose_> (postfix for only 4 hours of that, though)
[10:29:04] <ErikRose_> Have I been misspelling the *#$*#$ smtp_sasl_path all night long?
[10:29:08] <ErikRose_> smtp*D*_sasl_path
[10:29:17] *** mmcji has quit IRC
[10:30:22] <ErikRose_> Son of a...
[10:30:55] <ErikRose_> It works.
[10:31:49] <ErikRose_> WOOOHOOOOOOO!!!!
[10:31:54] * ErikRose_ dances around the room.
[10:32:05] <ErikRose_> Out, out, you demons of inaccurate spelling!
[10:32:16] *** Niemi has quit IRC
[10:32:18] <ErikRose_> THANK YOU, sed!
[10:32:22] <ErikRose_> sep!
[10:32:29] *** rprasad has joined #postfix
[10:32:49] <ErikRose_> Wow, if you ever need any Zope/Plone/Python doctoring, just call!
[10:35:18] <sep> :)
[10:35:32] <ErikRose_> Now to get stuff back into maildirs.
[10:36:36] *** brancaleone has quit IRC
[10:45:27] *** amrit is now known as amrit|zzz
[10:46:57] *** brancaleone has joined #postfix
[10:47:06] <ErikRose_> Done. Woo!
[10:47:17] <ErikRose_> I can has mail server again!
[10:48:31] *** sep has quit IRC
[10:48:49] *** sep has joined #postfix
[10:54:26] *** gstaniak has joined #postfix
[10:54:40] <gstaniak> hi
[10:55:27] <rprasad> hi
[10:55:47] <gstaniak> is there a way to use a milter (any pre-queue filter) so that postfix would dynamically decide whether it will allow relaying mail to a specific domain?
[10:57:35] <rprasad> what do you want to acheieve
[10:57:45] <rprasad> may not really require a milter
[10:59:11] *** Juspion has joined #postfix
[11:01:57] <gstaniak> rprasad: well, as i said i need to dynamically decide whether to relay or not. specifically, i need to to check whether the postfix machine is configured as an mx for this domain, and if it is so, allow relaying.
[11:03:06] *** pendlar2 has joined #postfix
[11:04:24] <pendlar2> Hi folks.. I've got a postfix running on my laptop, to handle outgoing mail from my mail client. I have to use different smtp relayhosts at work and at home (and also other sites). Can I set up several relayhosts? Is this a stupid way of doing it anyway? :-P
[11:04:29] *** randra has joined #postfix
[11:09:00] *** pendlar2 is now known as pendlaren
[11:10:47] *** Juspion has quit IRC
[11:15:16] <deface> why do you 'have' to ?
[11:15:40] <deface> and yes, stupid way
[11:15:57] <deface> for starters, having a mobile postfix setup ... stupid
[11:16:13] <deface> use the mailx package from your distro
[11:17:25] <rprasad>  gstaniak:  thats a waste of resources , whay cant you have a relaydomains file , if I may ask
[11:20:03] <gstaniak> rprasad: long story. large mail server at an isp that used to be mx for a lot of domains is being phased out. i want the server to not relay as soon as it is removed from the mx list of a domain, without having to manually modify the relay_domains list a lot of times
[11:20:49] *** idle-boy has left #postfix
[11:21:44] <rprasad> u would better do a cached copy of relaydomains. Send 451 to all domains you dont have in relaydomains
[11:21:57] <rprasad> then parse logs and populate the relaydomains
[11:21:58] *** idle-boy has joined #postfix
[11:22:01] <rprasad> after MX lookup
[11:23:12] <rprasad> else you would have to have real code at a milter
[11:25:51] <gstaniak> rprasad: you mean using an external tool? something that would regularly go through logs, check for recipient domains, check if the server is listed as an MX for them and then re-populate the relaydomains list?
[11:26:10] *** rprasad has quit IRC
[11:27:47] *** rprasad has joined #postfix
[11:28:35] <rprasad> gstanaik: parsing logs thru external script is best for u
[11:28:52] *** ErikRose_ has quit IRC
[11:32:05] <gstaniak> rprasad: ok, thx
[11:32:07] *** Niemi has joined #postfix
[11:43:06] <pendlaren> deface: thanks. I installed postfix as a test since I have some trouble with the smtp-server at work, and usually smtp-servers doesn't allow relaying if you're not on the inside of their network...
[11:45:23] <pendlaren> deface: So instead of having alpine (mail client) try a list of smtp servers, I thought I perhaps could let postfix do that job in the background
[11:46:04] <deface> sound redundant to me
[11:46:07] <deface> sounds*
[11:46:21] <deface> your just using sendmail, not postfix
[11:46:27] <deface> ie. mail is not delivered to your laptop
[11:46:36] <deface> your just using the sendmail binary
[11:46:42] <pendlaren> ah... yes, that's correct, but I thought postfix was a sendmail implementation?
[11:46:55] <deface> yes, but there is also sendmail
[11:46:59] <deface> w/ out the mta
[11:47:18] <pendlaren> ah... that's what I wanted to do, yes
[11:47:26] *** master_void has joined #postfix
[11:47:31] <deface> what distro ?
[11:47:38] <pendlaren> ubuntu
[11:47:47] <deface> i think its the mailx package as well
[11:48:05] *** gstaniak has quit IRC
[11:48:29] <pendlaren> mailx                           - A simple mail user agent
[11:48:32] <pendlaren> seems like
[11:48:34] *** pirho has joined #postfix
[11:48:46] <deface> try that on for size
[11:49:01] <pendlaren> I'll try to swap postfix with that one!
[11:49:26] *** cpm has joined #postfix
[11:59:35] *** growltiger_ has joined #postfix
[12:01:26] *** ming_zym has quit IRC
[12:05:59] *** bigtone has joined #postfix
[12:09:17] *** Tex-Twil has joined #postfix
[12:09:29] *** Tex-Twil has left #postfix
[12:10:10] <bigtone> what is the common wisdom of including reject_invalid_hostname (or reject_invalid_helo_hostname) in smtpd_recipient_restrictions?  I know that some ISPs fail this check, so I risk losing mail (at least until their MTA bounces).  If I remove it, what is likely to happen to my spam load?
[12:10:18] <bigtone> (broad general question, I know)
[12:10:32] *** Niemi has quit IRC
[12:14:52] <dragonheart> could use a warnifreject directive to guage the impact
[12:16:18] *** growltiger has quit IRC
[12:19:31] *** roe has joined #postfix
[12:19:41] *** drummerboy has quit IRC
[12:22:12] *** verywiseman has quit IRC
[12:27:20] *** gutocarvalho has joined #postfix
[12:29:20] *** jra has joined #postfix
[12:31:55] <pendlaren> deface: I got exim4 (that's what got installed when I installed mailx on ubuntu) running quite easily, and it pushes out e-mail through an smtp, but I can't find anything on how I setup multiple smtp servers...but I guess this is the wrong channel for that... :) Thanks for your help, anyway!
[12:36:36] *** pirho has quit IRC
[12:37:23] *** pirho has joined #postfix
[12:40:15] <bigtone> dragonheart: thanks
[12:43:13] *** roe has quit IRC
[12:45:15] *** muecke77 has joined #postfix
[12:48:16] *** muecke77 has quit IRC
[13:01:03] *** kk_CHN has joined #postfix
[13:04:34] *** weedar has joined #postfix
[13:14:10] *** PaSzCzUs has quit IRC
[13:15:39] *** s0what has joined #postfix
[13:20:18] *** mrfree has joined #postfix
[13:20:38] <mrfree> hi all
[13:21:51] <mrfree> some local daemons use ssmtp to send email to hostmaster via postfix, but it doesn't work  http://pastebin.com/d55e43f2a
[13:23:10] <mrfree> any idea?
[13:23:34] <mrfree> postfix seems to work well using other client
[13:27:53] <randra> any here use maia mailguard or mailzu!?
[13:31:09] *** Dakylla has joined #postfix
[13:31:16] <Dakylla> hi
[13:32:15] <Dakylla> i"ve a working virtual domain setup. i would like to have my maildir folder defined with relative paths in my sql table
[13:32:41] <Dakylla> how can i set postfix so that it has a default/prefix maildir for all mailboxes please ?
[13:33:25] *** euphidime has quit IRC
[13:40:14] *** euphidime has joined #postfix
[13:42:56] *** Tex-Twil has joined #postfix
[13:43:10] *** Tex-Twil has left #postfix
[14:00:33] *** Dakylla has quit IRC
[14:08:22] *** vice-versa has quit IRC
[14:20:12] *** verywiseman has joined #postfix
[14:29:39] *** kk_CHN has quit IRC
[14:33:26] *** roe has joined #postfix
[14:35:37] *** eanxgeek has joined #postfix
[14:36:33] *** rprasad has quit IRC
[14:41:59] *** weedar has quit IRC
[14:42:15] *** weedar has joined #postfix
[14:49:12] <lunaphyte_> mrfree: how about solving the problems that postfix specifically mentions in those log entries?
[14:49:40] <lunaphyte_> !mysql
[14:49:40] <knoba> lunaphyte_: "mysql" : http://www.postfix.org/MYSQL_README.html is helpful in configuring postfix to talk to a mysql server.
[14:49:53] <lunaphyte_> oh, he left.
[14:50:25] <robtone_> right.
[14:51:18] <mrfree> lunaphyte_, which problem in particular?
[14:52:10] <lunaphyte_> i guess if it were me i'd fix all of them.
[14:53:36] <mrfree> lunaphyte_, I noticed other local clients (windows) doesn't work... probably because they use their netbios name
[14:53:47] *** deuterium has joined #postfix
[14:54:12] <lunaphyte_> i'd imagine so.
[14:55:12] <mrfree> lunaphyte_, I think it should be related with my last config rewrite :)
[14:55:49] <mrfree> I'd like to disable  reject_unknown_sender_domain  for sasl-auth users
[14:56:10] <lunaphyte_> why on earth would you do that?
[14:56:20] <mrfree> reject_unknown_helo_hostname
[14:56:37] <mrfree> ops I think it should be the prob
[14:56:37] <mrfree> :)
[14:57:42] <lunaphyte_> are you saying that you mean reject_unknown_helo_hostname rather than reject_unknown_sender_domain?
[14:59:00] <mrfree> yup, I've added reject_unknown_helo_hostname to smtpd_helo_restrictions I think it could be useful for the other smtp server but the client
[15:01:37] <mrfree> can I add something like permit_sasl_authenticated *before* reject_unknown_helo_hostname in the smtpd_helo_restrictions?
[15:01:48] <mrfree> to skip this check for my users client...
[15:02:10] <lunaphyte_> !tias
[15:02:11] <knoba> lunaphyte_: "tias" : Try It And See
[15:02:23] <mrfree> lunaphyte_ :)
[15:04:19] <mrfree> lunaphyte_, I mean... adding permit_sasl_authenticated in the smtpd_helo_restrictions has effects only for smtpd_helo_restrictions or can it be evaluated in some risky matter?
[15:04:21] <mrfree> manner
[15:06:27] <jra> just use smtpd_recipient_restrictions and forget about the rest
[15:09:11] <mrfree> reading the logs I found a lot of smtp server sending their hostname (without domain) for the help
[15:09:12] <mrfree> o
[15:09:23] <mrfree> I think I should disable this restriction
[15:11:14] <jra> I think I stop >80% of spam just with reject_non_fqdn_helo_hostname
[15:13:17] *** roe has quit IRC
[15:24:41] *** UQlev has joined #postfix
[15:26:36] *** Southron has joined #Postfix
[15:26:55] <mrfree> jra, reject_non_fqdn_helo_hostname only checks if the helo is "well-formed", right? So should be better to check if the hostname exists with reject_unknown_helo_hostname, shouldn't it?
[15:27:27] <mrfree> with a cost of a dns query I suppose
[15:31:17] *** h2o_backup_nick1 has joined #postfix
[15:34:04] <jra> I only check for unknown_sender_domain, non_fqdn_helo_hostname catches enough for my taste... don't want to risk any false positives
[15:37:35] *** tombar has joined #postfix
[15:39:28] *** mrichman has joined #postfix
[15:39:51] <mrichman> Can someone tell me why I get a cert validation error (and an untrusted connection) here: ov 20 09:46:37 securemail-qa postfix/smtp[14030]: certificate verification failed for cluster12.us.messagelabs.com[85.158.136.227]:25: untrusted issuer /C=GB/O=Trustis Limited/OU=Trustis FPS Root CA
[15:41:19] <mrichman> Do I need to install the Trustis FPS Root CA somewhere?
[15:46:54] *** xpoint has joined #postfix
[15:47:28] *** deuterium has quit IRC
[15:53:44] *** stas` has quit IRC
[16:05:20] <lunaphyte_> mrichman: it's likely installed, just not found by postfix.
[16:07:04] *** TheBirdMan has joined #postfix
[16:07:28] <TheBirdMan> what keyword would i look for in a postfix maillog to know if an email sent out successfully?
[16:08:05] <cpm> status=sent
[16:08:41] *** cilquirm has joined #postfix
[16:10:23] <mrichman> i concatenated their root CA .pem file to my smtpd_tls_CAfile, but I still get the untrusted warning
[16:10:53] *** hparker has joined #postfix
[16:11:02] <mrichman> lunaphyte:
[16:13:11] <TheBirdMan> i guess my next question is, is there a generic/ambiguous keyword that indicates that something went wrong from maillog?
[16:13:59] *** nicku has joined #postfix
[16:14:00] <mrichman> TheBirdMan: does mail.err help?
[16:14:40] *** cilquirm has quit IRC
[16:15:26] <nicku> hello.. is it possible to get postfix to authenticate with pam
[16:16:04] <cedric33> hi i have more mail  on postqeue  i have this error delivery temporarily suspended: connect to 127.0.0.1[127.0.0.1]: Connection refused)
[16:16:14] <cedric33> if i want to restart the send of this mail
[16:16:28] <cedric33> i don't know the command to do that if you can help me please thanks
[16:17:44] <mrichman> cedric33: i think the postsuper command can requeue emails
[16:18:41] <cedric33> postsuper can tell to postfix to try to resend the mail thanks
[16:19:08] *** weedar has quit IRC
[16:20:34] <UQlev> nicku: you should use sasl2 or dovecot with postfix to authenticate against pam
[16:20:46] *** h2o_backup_nick1 is now known as deuterium
[16:21:29] <nicku> UQlev: sorry,, i meant i need to authenticate against winbind
[16:22:12] <UQlev> nicku: smtp-auth?
[16:26:38] *** pendlaren has left #postfix
[16:29:01] <nicku> UQlev: can i make it to use Windows ads password db
[16:32:41] <nicku> I have samba to authenticate against winbind and it gets the passwords from ads.
[16:32:55] <nicku> now i need to get the same with postfix
[16:34:47] <UQlev> nicku: I gues postfix+ldap should work with ads
[16:35:50] *** weedar has joined #postfix
[16:36:07] <UQlev> nicku: probably you will need postfix+dovecot+ldap
[16:37:37] *** tombar has quit IRC
[16:38:45] *** tombar has joined #postfix
[16:47:02] *** tshine_zzz is now known as tshine
[16:49:57] <cedric33> hi i have a problem on my qeue postfix i have 8000 mails i don't know if it's my modification do that i modify on master.conf  smtp-amavis     unix    -       -       n       -       6       lmtp
[16:50:12] <cedric33> sorry i use 20 process smtp-amavis
[16:50:37] <cedric33> i modify now to 10 i have the same problem do you thinks the best number of process is thanks
[16:51:14] <mrichman> Does postfix cache MX records?
[16:51:14] *** pitakill has joined #postfix
[16:51:58] *** TheBirdMan has quit IRC
[16:52:18] *** TheBirdMan has joined #postfix
[16:52:22] <cedric33>  i don't know  how i can see that
[16:52:37] <cedric33> mrichman : how i can see that
[16:52:38] <cedric33> please
[16:52:53] <jra> postfix uses whatever you're using in /etc/resolv.conf
[16:53:00] <mrichman> cedric33: sorry i dont know
[16:54:16] *** _bt has quit IRC
[16:54:44] <cedric33> i see the check amavis is very long and the number of process i don't know if it's good
[16:54:56] <nicku> jra: do you know howto postfix + ads authenticate
[16:56:55] <jra> nope
[16:58:58] <mrichman> jra: for some reason my mail isnt being delivered....i updated the MX record to correct it, but i still dont see the mails going out
[16:59:08] <mrichman> jra: no errors and no delivery...weird
[16:59:40] <cedric33> nobody have any idea of my problem please?
[16:59:51] <mrichman> cedric33: please dont beg ;)
[17:00:21] <cedric33> mrichman : please help me :) :)
[17:00:27] *** |_Knoedel_| has quit IRC
[17:02:16] <mrichman> My email is handled by a third party via MX record on the same domain as my local postfix box. I get this message when I try to send myself a test message from the postfix box: Nov 20 11:02:04 securemail-qa postfix/local[15442]: 39692383B6: to=<mrichman at wealthmsi dot com>, relay=local, delay=0.24, delays=0.17/0.02/0/0.06, dsn=2.0.0, status=sent (delivered to command: procmail -a "$EXTENSION")
[17:03:00] <mrichman> I don't want "local" delivery, but relay to the server in MX. I think the issue is the same domain as the postfix server. How can I tell it to relay out for its own domain to avoid local delivery?
[17:04:29] *** slushpupie has quit IRC
[17:11:45] *** weedar has quit IRC
[17:12:47] <mrichman> I need to make it use the MX record if it cant find a local user for delivery
[17:20:31] *** githogori has quit IRC
[17:22:26] *** linguini has joined #postfix
[17:23:40] *** drummerboy has joined #postfix
[17:29:55] *** Knoedel2 has joined #postfix
[17:31:28] *** UQlev has quit IRC
[17:33:39] *** SARGuy has joined #postfix
[17:37:45] <bpgoldsb> I have a postfix 2.3.8 server using Dovecot's LDA.  This server has user foo and alias bar.  foo is a member of the alias bar.  When I send an email with recipient 'foo' and 'bar', foo's mailbox gets two copies.  Both copies in the mailbox are identical (verified with diff -u file1 file2).  This wasn't the behavior on my older Postfix 2.1.x setup (which used Maildrop).  Anyone know why it might be different?
[17:40:40] *** mrfree has quit IRC
[17:43:14] *** phnord has quit IRC
[17:48:25] *** roe has joined #postfix
[17:49:28] <cedric33> if the qeue postfix is full 8000 mails is it normal to check all is very long ?
[17:49:34] *** eanxgeek|laptop has joined #postfix
[17:54:41] *** master_void has quit IRC
[18:02:24] *** sypher has joined #postfix
[18:07:12] *** jonez has joined #postfix
[18:19:26] *** Spec has quit IRC
[18:22:41] *** _mavrick61 has quit IRC
[18:27:42] *** havvg has joined #postfix
[18:31:25] *** magyar_ has joined #postfix
[18:36:29] *** Tabmow has quit IRC
[18:36:46] *** Tabmow has joined #postfix
[18:39:38] *** githogori has joined #postfix
[18:44:06] *** slushpupie has joined #postfix
[18:46:43] <cedric33> hi is it possible to flush a postqueue for a domain name
[18:48:59] *** drummerboy has left #postfix
[18:49:01] <roe> a virtual domain?
[18:49:31] <cedric33> yes
[18:49:38] <cedric33> because i have a big problem
[18:49:58] *** sypher has quit IRC
[18:50:03] <cedric33> i have on my postqueue 9000 mails and it very long to send the mail i don't know why
[18:50:44] *** c00l2sv has joined #postfix
[18:51:14] <roe> and what makes you think that requeuing them manually will fix the issue?
[18:51:40] <roe> there is a way to requeue messages fitting a regexp, but I forget how
[18:51:44] *** tombar has quit IRC
[18:51:56] *** Tapout has joined #postfix
[18:52:04] <cedric33> i use postqueue -p to flush all mail but nothink
[18:52:10] <cedric33> i have the same numbers of mails
[18:52:24] <Trengo> you mean -f dont you?
[18:52:25] <roe> postqueue is also "site" aware, however I am not sure how sites are assigned
[18:52:52] <cedric33> sorry yes postqueue -f
[18:53:06] *** c00l2sv has quit IRC
[18:53:07] *** stas has joined #postfix
[18:53:20] <Trengo> maybe its undeliverable spam
[18:53:31] <Trengo> look at postqueue -p ?
[18:53:52] <cedric33> i use qshape -s  i see the domain name it's a good domaine name not a spam
[18:54:02] <cedric33> when i see with postqueue -p i see the good mail
[18:54:35] <cedric33> i see amavis-new have 100% of CPU to check if it's a spam or no
[18:56:19] <roe> depending on your hardware and rules, that could be completely normal
[18:56:43] <Trengo> look at postqueue -p it should show the reason why its not delivering
[18:58:00] <cedric33> Trengo : i see but i have not a error
[18:58:13] <cedric33> 78739B8318*   60868 Thu Nov 20 13:24:11 .... not error
[19:01:52] <cedric33> i try to flush a domain name with postqueue -s test.Fr i have postqueue: fatal: Flush service is not configured for destination
[19:07:23] *** vertigo- has quit IRC
[19:07:49] *** s0what has quit IRC
[19:09:10] <cedric33> nobody have any idea why when i try to flush a site with postqueue i have postqueue: fatal: Flush service is not configured for destination
[19:09:15] <cedric33> thanks for your help
[19:09:42] <roe> don't assume sites = virtual domains
[19:09:51] *** Tapnzout has quit IRC
[19:10:49] <cedric33> roe : ho thanks
[19:11:28] <roe> I do not know what "sites" mean, you will probably have to read some documentation
[19:13:26] <cedric33> yes i think
[19:13:33] <roe> cedric33, the resident experts in this cannel seem to be busy at their day jobs
[19:14:16] *** Zeit|awy_ has joined #postfix
[19:15:29] <cedric33> yes
[19:15:30] <cedric33> :)
[19:16:05] <cedric33> my server work for more time  just now i have this problem
[19:16:07] <cedric33> strange
[19:20:19] *** Zeit|awy has quit IRC
[19:21:38] <roe> I would say an update to an sa rule may be the cause to your performance issues, there is a timing flag to send to amavis to see where it is getting stuck, but again, I forget... I need to start making notes for myself
[19:21:45] *** Tapnzout has joined #postfix
[19:21:56] <lunaphyte_> !fast_flush_domains
[19:21:56] <knoba> lunaphyte_: "fast_flush_domains" : a configuration parameter in the main.cf: Optional list of destinations that are eligible for per-destination logfiles with mail that is queued to those destinations.
[19:22:02] <lunaphyte_> cedric33: ^^
[19:22:04] *** verywiseman has quit IRC
[19:22:31] <roe> is that a new feature?  I don't remember seeing that in the past
[19:23:02] *** EasilyOdd has quit IRC
[19:23:49] *** harobed has quit IRC
[19:24:42] *** deadpigeon has joined #postfix
[19:24:49] <cedric33> lunaphyte : thanks i must modify my main.cf i search the line to add on my file thanksssssss
[19:25:24] <lunaphyte_> that's been in there for ages.  at least since 2000, afair.
[19:25:51] *** TGM has joined #postfix
[19:26:05] <roe> really, that seems like an invaluable feature for a  virtual mail setup
[19:26:11] <cedric33> lunaphyte :; have you any idea of my probleme please
[19:26:16] <cedric33> beucause me i have not
[19:26:23] <roe> granted I don't run a virtual mail setup, but I still would like to think I would have heard of it
[19:26:48] <lunaphyte_> as i recall, it was intended more for etrn related operations.
[19:27:21] <lunaphyte_> the flush service, that is.
[19:28:05] *** hever has joined #postfix
[19:28:32] *** hever has quit IRC
[19:28:53] <roe> !tell roe etern
[19:28:54] <knoba> roe: Error: No factoid matches that key.
[19:29:03] <lunaphyte_> you must balance amavis' feeding with it's output.
[19:29:14] *** niki has joined #postfix
[19:29:46] *** Tapout has quit IRC
[19:30:07] *** stas has quit IRC
[19:30:29] *** eanxgeek|laptop has quit IRC
[19:30:53] <cedric33> lunaphyte : how?
[19:32:53] *** Ryushin has joined #postfix
[19:37:04] *** gutocarvalho has quit IRC
[19:40:15] *** amrit|zzz is now known as amrit|wrk
[19:47:41] *** carl- has joined #postfix
[19:48:33] <roe> for sufficiently large sites, do people still use cat/less/grep or something else to investigate issues in mail.log
[19:50:13] *** rouri has joined #postfix
[19:50:19] <lunaphyte_> sure
[19:51:39] *** Tapnzout has quit IRC
[19:52:12] <roe> I find it frustrating at times
[19:52:19] <lunaphyte_> why?
[19:52:24] <cedric33> but if on mail.log you have not error log
[19:54:32] *** jeev has quit IRC
[19:54:37] *** loddafnir has quit IRC
[19:54:41] *** Tapout has joined #postfix
[19:57:48] *** brancaleone has quit IRC
[20:07:58] *** randra has quit IRC
[20:12:18] *** Spec has joined #postfix
[20:13:08] *** bhagat has quit IRC
[20:13:38] *** bhagat has joined #postfix
[20:14:09] *** linguini has quit IRC
[20:14:11] *** hever has joined #postfix
[20:15:57] <roe> well, there are common issues that I need to lookup and  sifting through lots of clutter to find a straight forward question takes longer than it should, I guess I could save my greps and build little python/perl scripts, but I am lazy
[20:17:18] <lunaphyte_> aha.  the truth is revealed.
[20:22:32] *** s0what has joined #postfix
[20:25:28] *** modelnine has joined #postfix
[20:27:47] <modelnine> Is setting "local_recipient_maps = $alias_maps" bad/broken? Why? (I know it's "broken", because I'm seeing infinitely queued mails destined for the $mydestination domains) I'm trying to setup a system that will only accept mail for the names defined in the alias table[s], which are all themselves destinations to external hosts.
[20:28:01] *** hever has quit IRC
[20:29:55] <modelnine> Just to add to that: the destination address for the account in the alias map itself is a relaydomain (and defined in a map queried relay_domains), and has an entry in the transport_maps table.
[20:30:42] <roe> lunaphyte_, I fully admit that I could solve my own issue, I was just wondering if someone else already solved my issue for me :)
[20:31:42] <bpgoldsb> I have a postfix 2.3.8 server using Dovecot's LDA.  This server has user foo and alias bar.  foo is a member of the alias bar.  When I send an email with recipient 'foo' and 'bar', foo's mailbox gets two copies.  Both copies in the mailbox are identical (verified with diff -u file1 file2).  This wasn't the behavior on my older Postfix 2.1.x setup (which used Maildrop).  Anyone know why it might be different?
[20:35:27] <cedric33> i have a last question i see on the log i have more mails of one virtual domain can i check the number of  mail i receiv  for this  day please
[20:36:01] <modelnine> cedric33: try a grep "@mydomain>" | wc -l
[20:36:06] <modelnine> on the mail log.
[20:37:49] *** TGM has quit IRC
[20:38:24] <cedric33> thanks
[20:40:05] <cedric33> i think with qshape i can do it . With grep ... i can have an line of 13d.fr but error or other i am not shure it send line email
[20:41:31] <modelnine> cedric33: you can. Simply check for the actual line format you want to find (signalling a successful delivery), tack that into a regex, and do a wc -l.
[20:41:42] <modelnine> I didn't say that "@mydomain>" would be sufficient. :-)
[20:42:15] <cedric33> :)
[20:42:26] <cedric33> ok  i add regex thanks
[20:42:51] *** sypher has joined #postfix
[20:44:09] *** pulsar has joined #postfix
[21:00:22] *** denis_ has joined #postfix
[21:02:06] *** roe has quit IRC
[21:11:22] *** modelnine has quit IRC
[21:19:06] *** slushpupie has quit IRC
[21:19:21] *** felix-da-catz_zz is now known as felix-da-catz
[21:19:33] *** slushpupie_ has joined #postfix
[21:20:08] *** rour has joined #postfix
[21:20:33] *** TheBirdMan has quit IRC
[21:21:06] *** Ryushin has quit IRC
[21:33:03] *** rour is now known as ruba
[21:41:56] *** BuenGenio has joined #postfix
[21:43:39] *** alxgomz has quit IRC
[21:44:06] *** rouri has quit IRC
[21:50:42] *** BuenGenio_ has joined #postfix
[21:53:24] *** shabbs has joined #postfix
[21:53:53] <shabbs> can someone tell me what the delays=x/y/z/w values mean in the logs?
[21:54:03] *** BuenGenio has quit IRC
[21:56:34] <Dominian> Postfix logs additional delay information as "delays=a/b/c/d" where a=time before queue manager, including message transmission; b=time in queue manager; c=connection setup time including DNS, HELO and TLS; d=message transmission time.
[21:57:14] *** cpm has quit IRC
[21:57:40] <shabbs> thanks
[21:57:50] <Dominian> taken from google groups ;)
[21:59:42] *** pitakill has quit IRC
[22:00:11] <shabbs> I must have been searching for the wrong phrases then because I got nothin from google :P
[22:02:23] <shasta> shabbs, http://www.postfix.org/postconf.5.html#delay_logging_resolution_limit
[22:03:10] *** brancaleone has joined #postfix
[22:03:48] *** UQlev has joined #postfix
[22:06:03] *** eanxgeek has quit IRC
[22:06:33] *** pickcoder has joined #postfix
[22:09:12] <shabbs> so, I'm having this interesting issue with a gateway mail server. it does spam and virus checking through amavis (sa/clamav) reinserts the message on 10024 and then forwards either inside or outside depending on which way the mail goes.
[22:09:42] <shabbs> the problem is that seemingly randomly messages coming in get deferred
[22:09:58] <shabbs> then at a later date when the queue retries they go through
[22:10:20] <shabbs> so sometimes people get 3-4 messages all at once that they should've received 20 minutes ago
[22:10:32] <roe_> get defered from amavis or from the destination server?
[22:11:01] <shabbs> the log says "conversation with [internal server] timed out while sending end of data"
[22:11:13] <roe_> on the 10024 port?
[22:11:41] <shabbs> I'll pastebin the log of a message from start to finish
[22:13:02] <shabbs> http://pastebin.com/m7e8de3a0
[22:13:55] <pickcoder> PIX
[22:14:31] <shabbs> we've checked the fw config, first off it's version 8.1 or 8.3 second there is no smtp fixup
[22:15:08] <pickcoder> it's timing out with the 10..10.10.3 machine not amavis
[22:15:13] <pickcoder> s/../.
[22:16:16] <shabbs> any other setting that would affect mail through pix?
[22:16:31] <pickcoder> when it resends using the PIX work-around it goes through
[22:19:00] <shabbs> there are several messages in the queue that seem stuck permenantly. juding by the from address they're spam anyway, but they have the same timed out message, and no matter how many times postfix retries they still end up getting deferred
[22:19:22] <pickcoder> !smtp_pix_workaround_threshold_time
[22:19:23] <knoba> pickcoder: "smtp_pix_workaround_threshold_time" : a configuration parameter in the main.cf: How long a message must be queued before the PIX firewall "<CR><LF>.<CR><LF>" bug workaround is turned on.
[22:19:43] <pickcoder> 0 makes it active upon first attempt
[22:20:41] <pickcoder> default is 500s queue  time
[22:21:00] <shabbs> I'll give it a try
[22:22:20] *** BuenGenio has joined #postfix
[22:22:44] *** BuenGenio_ has quit IRC
[22:23:02] <pickcoder> if the deferrals are constantly happening then you should see an immediate effect if that's the problem
[22:23:42] <pickcoder> you can put the queue on hold and then release it to flush the queue and reset the queue timing
[22:23:57] <pickcoder> you still have to restart postfix though
[22:26:28] *** FFForever-BNC has joined #postfix
[22:26:47] <FFForever-BNC> what do i put for  tls_random_source?
[22:27:25] <shabbs> yeah, I restarted and flushed, now I'm waiting to see if they'll end up back in the deferred queue. what seemed odd was that it wasn't happening to all messages. there'd be stretches for like an hour where all messages would go through immediately
[22:27:43] <shasta> FFForever-BNC, why do you want to change the defaults for your system?
[22:28:13] <FFForever-BNC> i get this error =\ Nov 19 14:09:20 mail postfix/smtpd[25081]: fatal: unsupported dictionary type: dev
[22:28:30] <FFForever-BNC> its dev:/dev/urandom
[22:28:45] <shasta> once again, why did you change the defaults?
[22:29:10] <FFForever-BNC> i didn't the idiot before me did....
[22:29:14] <shasta> (and the error is pretty much self-explanatory)
[22:29:24] <FFForever-BNC> whats the default
[22:29:52] <jra> depends on where you compile
[22:30:01] <FFForever-BNC> its an ubuntu server
[22:30:17] <jra> then it's dev:/dev/urandom
[22:30:31] <FFForever-BNC> thats what it is....
[22:30:53] <shasta> man 5 postconf will tell you how to get the defaults
[22:31:42] <shasta> oops, my bad, man 1 postconf
[22:32:49] *** UQlev has quit IRC
[22:35:22] <shabbs> hrm, just had a whole bunch of them go back in the deferred queue
[22:36:09] <shabbs> and this last one has delay=0.07
[22:36:19] <shabbs> so somethin ain't right
[22:37:26] *** FFForever-BNC has left #postfix
[22:38:29] <pickcoder> what is smtp_pix_workarounds set to?
[22:39:39] <shabbs> according to postconf -n it's not because it doesn't show up, but according to the config file disable_esmtp, delay_dotcrlf
[22:40:05] <shabbs> but it also didn't show in postconf -n when I specifically set it to nothing, which I found odd
[22:40:50] *** scort has joined #postfix
[22:40:54] <pickcoder> postconf mail_version?
[22:41:40] <shabbs> heh, 2.3.8 which would explain why it doesn't do anything
[22:41:48] <pickcoder> yup
[22:41:49] <shabbs> silly me
[22:41:54] <jra> \o/
[22:43:03] <pickcoder> Debian?
[22:43:12] <shabbs> ubuntu
[22:43:15] *** FFForever-BNC has joined #postfix
[22:43:24] <pickcoder> and it still has 2.3.8?
[22:43:31] <shabbs> 7.04
[22:43:51] <pickcoder> I'm not up to speed on ubuntu releases
[22:43:53] <pickcoder> I run Debian
[22:44:14] <pickcoder> +Fedora+RHEL
[22:44:15] <shabbs> 8.04 is latest release
[22:44:29] <shabbs> so I'm 2 releases behind
[22:44:30] <lennard> isnt intrepid 8.10?
[22:44:31] <pickcoder> Lenny (testing) has 2.5.5 which is inline with Fedora 9
[22:45:08] *** carl- has quit IRC
[22:45:44] <shabbs> wow, am I really 3 behind? interpid 8.10, hardy 8.04?
[22:45:50] <shabbs> intrepid rather
[22:45:58] <shabbs> time to upgrade
[22:46:04] <pickcoder> shabbs: welcome to Ubuntu
[22:46:06] <pickcoder> :P
[22:46:13] <shabbs> :P
[22:46:25] <shabbs> thanks for the help. if I still have issues after the upgrade I'll come back :)
[22:46:30] <lennard> well, you could upgrade to 8.04, hardy is a LTS release :)
[22:46:34] <pickcoder> no prob
[22:46:40] <shabbs> yeah, that's what it'll be
[22:46:46] <shabbs> thanks again!
[22:46:49] *** shabbs has quit IRC
[22:47:00] *** denis_ has quit IRC
[23:08:49] *** brancal has joined #postfix
[23:16:18] <mrichman> Is there a way to specify an MX record to use if a local mailbox does not exist?
[23:20:07] *** brancaleone has quit IRC
[23:23:16] *** jens_ has quit IRC
[23:39:26] *** Knoedel2 has quit IRC
[23:43:41] *** s0what has quit IRC
[23:49:24] *** BuenGenio has quit IRC
[23:49:55] *** libervisco has joined #postfix
[23:50:59] *** BuenGenio has joined #postfix
[23:51:04] *** havvg has quit IRC
[23:51:18] *** verywiseman has joined #postfix
[23:52:12] *** aka has joined #postfix
[23:52:50] *** FFForever-BNC has left #postfix
[23:53:51] *** brancal has quit IRC
[23:54:47] <SARGuy> have an interesting problem... i have mail coming into PF, which is configured to lookup all address' in the virtual address table...  problem is, some work, some don't.  i.e. I can send a message to aaa at domain dot com (located in the virtual address table) and it's good..  send a message to bbb at domain dot com and it's returned as "unknown in virtual alias table"..  send a message to ccc at domain dot com and it ends up in a loop to the front end server...
[23:54:58] <SARGuy> ideas?  more explanation needed?
[23:55:51] <hyper__ch> SARGuy: is bbb at domain dot com in the virt table?
[23:56:08] <SARGuy> no.. sorry...  in that example, only aaa at domain dot com is in the table
[23:56:19] <SARGuy> so aaa@ and bbb@ work correctly..  ccc@ doesn't
[23:57:18] <hyper__ch> well, I'm not really familiar with postfix... I thought I'd just ask the obvious thing.... sometimes I get lost in details so much that I forget about obvious things :)





top