[00:02:55] *** jeffspeff2 has joined #postfix
[00:03:05] *** jeffspeff has quit IRC
[00:04:15] *** standel has quit IRC
[00:04:56] *** jeffspeff2 has quit IRC
[00:05:26] *** jeffspeff2 has joined #postfix
[00:06:42] *** Fallenou has quit IRC
[00:09:50] <r8> Are the RFC requirements for hostnames spread across RFC 1035 _and_ RFC 1033?
[00:13:09] <r8> Wait... I was just having trouble reading the BNF in 1035. The restriction I didn't see explicitly stated was that hostnames cannot be all-numeric.
[00:15:40] <r8> Hostnames also cannot start with a number, is that right?
[00:19:01] <rob0> I don't think that's true. At least, not in terms of BIND, I know named can serve all-numeric name components.
[00:19:23] *** danbeck has quit IRC
[00:19:44] <r8> Maybe one of those areas where RFC != reality. Or I'm reading RFC 1035 wrong. Or 1035's been superseded.
[00:20:01] *** fredrik` has joined #postfix
[00:21:00] <vice-versa> I seem to remember that being an early domain name restriction that no longer exists
[00:26:59] *** pirho has quit IRC
[00:28:11] *** fredrik` has quit IRC
[00:46:40] *** technoid_ has quit IRC
[00:47:07] *** Ryushin has joined #postfix
[00:52:28] *** MaD^MaRe` has joined #postfix
[00:57:38] *** pitakill has quit IRC
[01:01:32] *** war9407 has quit IRC
[01:02:19] *** pingouin_ has quit IRC
[01:03:34] *** pingouin has joined #postfix
[01:14:44] *** livid has joined #postfix
[01:18:10] *** pingouin_ has joined #postfix
[01:19:36] *** R__ has quit IRC
[01:23:28] *** Zblakany has quit IRC
[01:25:01] *** pingouin_ has quit IRC
[01:25:08] *** pingouin_ has joined #postfix
[01:25:58] *** knoba has quit IRC
[01:26:17] *** knoba has joined #postfix
[01:28:58] *** pingouin has quit IRC
[01:31:57] *** glad_work has joined #postfix
[01:47:26] <r8> So it seems to me that just what constitutes a valid hostname is rather confused. (Aside from the fact that DNS labels aren't necessarily hostnames, though hostnames are DNS labels.)
[01:48:12] <r8> Whether a hostname must begin with a letter. Whether a hostname may be entirely numerical. Whether a hostname may contain an underscore.
[01:48:22] <r8> On this last point it seems the consensus is no.
[01:48:36] <r8> I mean, aside from You Know Who.
[01:49:07] <glad_work> ok if i'm using a postfix box to be a relay to my exchange box (inbound mail) and i have a second exchange box that i want to be configured as the relay failover/load balanced - can i manage that with the transport file?
[01:49:10] *** styx-tdo has joined #postfix
[01:50:14] <styx-tdo> hi, i have a weird problem, google did not show anyhthing useful. how can i prevent mail from apache/php (e.g. buggy forums) to be sent via the local postifx? - basically, block all local sent mail not having a domain name part in the sender?
[01:52:17] <styx-tdo> i want all mail that needs append_at_myorigin to not be accepted
[01:52:23] <styx-tdo> or just dropped
[01:54:39] *** Ryushin has quit IRC
[01:57:57] <styx-tdo> as this doesn't use all the smtp stuff, all these restrictions do not apply :(
[01:59:55] *** Broken|work__ has joined #postfix
[02:03:58] *** jeffspeff2 has quit IRC
[02:05:55] *** livid has quit IRC
[02:06:50] <seekwill> I'm guessing you don't want to implement smtpauth?
[02:08:45] *** cilly has quit IRC
[02:09:45] *** Motoko-chan has quit IRC
[02:13:23] <thumbs> seekwill: wake up!
[02:16:48] *** Broken|work_ has quit IRC
[02:21:34] *** jeffspeff has joined #postfix
[02:22:09] *** _xous has joined #postfix
[02:22:49] <glad_work> k - i think i have the balance thing figured out - my next issue is that i now get a 554 when i'm trying to send to a valid domain...
[02:23:00] <glad_work> (that i'm trying to forward to)
[02:28:00] *** xous has quit IRC
[02:30:28] *** ack_syn has joined #postfix
[02:30:35] <ack_syn> hello
[02:32:08] <xpoint> big problem :)
[02:32:43] <ack_syn> haha
[02:33:42] <ack_syn> hey, what system do you use to make statistics of send/receive e-mails, to se all sent/receive messages, in short, to interpret the maillog log in a web interface. I'm already using Isoqlog, mailgraph and mrtg. Is there anything else?
[02:34:46] <ack_syn> oh, I'm also using "post-la" a brazilian project. But it's full of bugs. I have fixed some ..
[02:35:35] <ack_syn> s/send/sent && s/receive/received .:. sorry my poor english
[02:35:40] <xpoint> google postfix-logwatch
[02:36:32] <xpoint> or pflogsumm.pl
[02:36:34] <ack_syn> is there any you know without search ? i.e. a postfix interface do administrate => postfixadmin.
[02:36:46] <ack_syn> oh, mailgraph use this script
[02:37:00] <xpoint> what ?
[02:37:18] <ack_syn> do you KNOW some solution?
[02:37:24] <ack_syn> or must I search for one?
[02:37:27] <xpoint> on what ?
[02:37:49] <ack_syn> solution to my problem
[02:37:55] <xpoint> 42
[02:38:14] <ack_syn> what ?
[02:39:42] <rob0> 0x2A
[02:39:50] <glad_work> hmm .. what causes your 220 to take 5-10 seconds to appear?
[02:40:44] *** jeffspeff has quit IRC
[02:41:24] <glad_work> if i telnet to 127.0.0.1 25 it appears instantly... but if i telnet to the ip of the machine (even locally) it takes a while....
[02:41:59] <xpoint> glad_work, bad router
[02:42:19] <xpoint> og dns server on localhost
[02:42:51] <glad_work> router should be fine
[02:43:25] <glad_work> since i'm telnetting from my postfix box to the internal ip of the postfix box
[02:44:18] <xpoint> could still be a bad router
[02:44:50] *** githogori has quit IRC
[02:45:13] <glad_work> could - but i can almost garantee that it isn't
[02:45:51] <xpoint> check its routes and or firewall
[02:46:15] *** Draecos has quit IRC
[02:46:43] *** jeffspeff has joined #postfix
[02:47:55] <ack_syn> glad_work, make the test using a client machine in the same IP networking than the 'postfix box'
[02:49:08] *** jeffspeff has quit IRC
[02:50:05] *** jeffspeff has joined #postfix
[02:52:31] <glad_work> ack_syn - already done that - same thing
[02:53:11] *** jeffspeff has quit IRC
[02:53:23] <ack_syn> so the problem must be in the gw, isnt it 'clear' for you?
[02:53:27] *** jeffspeff2 has joined #postfix
[02:53:53] <ack_syn> if same thing == test was fine, then the problem is in the gw
[02:54:16] <glad_work> k - i'll have the guy review the arp table
[02:55:13] *** jeffspeff2 has quit IRC
[02:55:41] *** jeffspeff2 has joined #postfix
[02:57:08] <ack_syn> xpoint, logwatch isn't a good "logwatch". I'd like to see all the sent/received messages on my mta. Do you know a script for this?
[02:57:20] <ack_syn> post-la do it, but it's a little poor
[02:57:40] <ack_syn> "poor" = lot of bugs
[02:59:02] <xpoint> i only use pflogsumm.pl and postfix-logwatch so dont need others here :)
[02:59:13] *** jeffspeff has joined #postfix
[02:59:44] <xpoint> cat logfile | pflogsumm.pl -d today | less
[03:00:19] <xpoint> cat logfile | postfix-logwatch --detail=20 | less
[03:00:34] <xpoint> ymmv
[03:01:37] <glad_work> ack_syn - it isn't the router
[03:01:54] <glad_work> resolve.conf was setup to use the labbed ns
[03:03:14] <ack_syn> xpoint ill try
[03:04:57] <ack_syn> sorry 'labbed' dont know this word
[03:05:18] <glad_work> build/testing environment
[03:05:38] <lunaphyte> hopefully you really meant resolv.conf
[03:05:52] <glad_work> i did ... just not focusing on what i'm typing here
[03:05:54] <glad_work> stupid deadlines
[03:06:12] <lunaphyte> what's a deadline?
[03:06:41] <ack_syn> anyways, if the problem consit in 'it appears instantly' and you tried from a client telnet IP 25, dns would not influence, IMO
[03:07:22] <glad_work> it does since its doiing reverse dns
[03:07:31] <lunaphyte> !chroot
[03:07:32] <knoba> lunaphyte: "chroot" : The fifth column in master.cf, if not n , means that the Postfix process described on that line runs in a chroot, see !debug , !queue_directory and files in the examples/chroot-setup subdirectory of the Postfix source archive which show examples of a Postfix chroot environment on a variety of systems
[03:07:47] <glad_work> and localhost is in the /etc/hosts file
[03:07:50] <ack_syn> to the first connect appears instantly? I dont think so...
[03:08:37] <glad_work> its working now - and the issue was reverse dns (or lack of dns altogether hehe)
[03:09:04] <ack_syn> I guess the reverse search is done when you try rcpt to command. So it verifies the reverse rcpt to MTA
[03:09:11] <glad_work> no
[03:09:15] <glad_work> it does reverse dns upon connect
[03:09:18] <ack_syn> Oops, s/MTA/MX
[03:09:38] <lunaphyte> smtpd_client_restrictions
[03:11:04] <ack_syn> smtpd_recipient_restrictions
[03:11:17] <ack_syn> :|
[03:11:53] <lunaphyte> both, but in his case, the former.
[03:11:54] <Shamgar> Hey, quick question. In a .forward+extension file, how do I tell it to just deliver it to a local file? I'm trying to piece it together from the local manpage, but apparently I'm just not seeing it.
[03:12:02] <Shamgar> i.e. to a maildir I mean
[03:13:12] <ack_syn> hehe, no make sense search the revers from a host IP
[03:13:19] <ack_syn> reverse
[03:13:39] *** jeffspeff2 has quit IRC
[03:14:13] <ack_syn> not every IP has reverse configured
[03:16:11] <lunaphyte> yes, a fair amount of false positives results if you employ that check.
[03:16:15] <Shamgar> nm, I found it.
[03:17:11] <lunaphyte> but even if you're not using that restriction, postfix still does dns lookups by default, for logging purposes.
[03:18:07] <Shamgar> or not. Ok, this is confusing. The doc says that if you want to deliver to a maildir you just put /path/to/maildir/ in the .forward.
[03:18:16] <Shamgar> I did that, and postfix tried to send it to /path/to/maildir/@mydomain
[03:19:40] <ack_syn> lunaphyte, look what he said. "if i telnet to 127.0.0.1 25 it appears instantly... but if i telnet to the ip of the machine (even locally) it takes a while...."
[03:20:07] <lunaphyte> yes.
[03:20:09] <ack_syn> I never seen a telnet takes a while because of a reverse lookup
[03:20:23] <lunaphyte> the telnet itself didn't.
[03:20:29] <lunaphyte> if you want to split hairs, that is.
[03:20:58] <ack_syn> 'split hairs' lol
[03:21:15] <Shamgar> doh, nm. I found the problem. Has to start with a / and I started with ./
[03:21:19] <ack_syn> so I told that guy, the problem isnt a reverse lookup
[03:21:37] <ack_syn> and it really wasnt
[03:21:52] <lunaphyte> sure it was.
[03:21:56] <ack_syn> :|
[03:22:08] <rob0> Shamgar: :)
[03:22:19] <ack_syn> ok, so my english doesnt let me understand why.
[03:22:31] <ack_syn> I mean, my _poor_ english
[03:22:50] <lunaphyte> when he connected using the address of the physical interface, postfix tried to contact a resolver to retrieve the ptr record for the source address, but wasn't able to.
[03:23:24] <glad_work> and spent 8 seconds timing out
[03:23:29] <glad_work> before sending a 220
[03:23:35] <lunaphyte> 8 seconds?!
[03:24:04] <glad_work> something like that
[03:24:06] <lunaphyte> the most dangerous 8 seconds in smtp!
[03:24:08] <lunaphyte> :p
[03:24:44] <glad_work> ok now to figure out why mail is looping onto itself and not being relayed to my server
[03:24:54] <lunaphyte> !loopback
[03:24:55] <knoba> lunaphyte: "loopback" : 'Mail loops back to myself' means that your Postfix wanted to send out the mail to the internet but then discovered that the DNS says your mail server should be responsible. Most likely you forgot to list your domain in mydestination or virtual_(alias|mailbox)_domains
[03:25:43] <ack_syn> well. I have never had this problem in a testing environment
[03:26:14] <lunaphyte> i'm sure the engineers who built that bridge that collapsed said the same thing.
[03:26:41] <glad_work> i take it transport_maps is depreciated?
[03:27:04] <rob0> Where did you read that?
[03:27:18] <lunaphyte> why would you say that?
[03:27:52] <glad_work> # TRANSPORT MAP | # | # See the discussion in the ADDRESS_REWRITING_README document.
[03:28:33] <rob0> Does ADDRESS_REWRITING_README say that transport_maps is deprecated?
[03:29:44] <glad_work> teach me for not reading
[03:32:33] *** ack_syn has quit IRC
[03:36:54] <glad_work> ok in my noobness i've failed to figure out how to get postfix to relay specific domains to my exchange server :/
[03:37:30] <lunaphyte> !basic
[03:39:47] <lunaphyte> oh - better:
[03:39:50] <lunaphyte> !examples
[03:40:10] <lunaphyte> see the postfix email firewall/gateway section
[03:40:50] * vice-versa found an email harvesting bot
[03:42:10] *** nphase_ has joined #postfix
[03:42:42] <vice-versa> appears to be an owned Windows Server 2003 boxen, go figure
[03:56:17] *** mavrick61 has quit IRC
[03:57:25] *** mavrick61 has joined #postfix
[04:00:10] *** Juspion has joined #postfix
[04:05:15] *** nphase_ has quit IRC
[04:06:15] *** danbeck has joined #postfix
[04:07:34] *** danbeck has quit IRC
[04:10:43] *** danbeck has joined #postfix
[04:11:35] *** danbeck has quit IRC
[04:27:18] <glad_work> . /etc/postfix/master.cf | comment out the local delivery agent?
[04:27:57] <glad_work> which one is that
[04:28:29] *** githogori has joined #postfix
[04:28:34] <glad_work> bah ... i'm getting lazy at this time of day
[04:28:42] <glad_work> asking before googling
[04:33:01] *** googlah has quit IRC
[04:35:20] *** memetic has quit IRC
[04:37:44] *** xp_prg is now known as xp_gone
[04:38:54] *** R__ has joined #postfix
[04:39:14] *** R__ has quit IRC
[04:39:52] *** xpoint has quit IRC
[04:44:58] <glad_work> alright - how do i get relaydomains into an external file? (this time i have googled)
[04:54:58] *** pickcoder has joined #postfix
[05:01:55] *** Juspion has quit IRC
[05:06:59] *** jeffspeff has quit IRC
[05:07:56] *** xous has joined #postfix
[05:19:29] *** _xous has quit IRC
[05:21:56] *** seekwill has quit IRC
[05:22:20] *** goldfisc1li has joined #postfix
[05:22:33] *** seekwill has joined #postfix
[05:27:49] <glad_work> ok with the transport file (for relaying specific domains) how do i get nlb/failover setup?
[05:35:46] *** goldfischli has quit IRC
[05:49:23] *** Motoko-chan has joined #postfix
[05:58:05] <mavrick61> I have try to find some documentation how to setup fallback relay to graveyard postfix server, any one who know?
[05:59:35] *** saurabhb has joined #postfix
[06:08:28] <seekwill> graveyard?
[06:13:17] *** pickcoder has quit IRC
[06:19:11] <vice-versa> something got lost in translation
[06:21:18] <r8> engrish.com
[06:25:35] <vice-versa> meh who knows
[06:25:42] * vice-versa orders some Chinese take-away from 'Translate server error'
[06:37:47] *** jeffspeff has joined #postfix
[06:42:25] <r8> A bare IP registers as a non-FQDN rather than an invalid hostname?
[06:45:56] <vice-versa> in what context, reject_invalid_helo_hostname?
[06:46:41] <r8> I've got both reject_invalid_helo_hostname and reject_non_fqdn_hostname (that order) and the second is what shows up in the log as what's triggered.
[06:48:13] <vice-versa> ip addresses are valid in helo/ehlo but must be encloded in []
[06:49:10] <r8> When they're not bracketed, I'm seeing reject_non_fqdn_hostname getting triggered rather than the preceding reject_invalid_helo_hostname.
[06:51:31] <vice-versa> are they in the same restriction stage?
[06:51:46] <r8> Yes.
[06:52:06] <r8> smtpd_recipient_restrictions
[06:54:39] <r8> I'm not conversant in C but it looks by the code that both of those restrictions check valid_hostname(), which I'm guessing is where the problem is being identified. (reject_non_fqdn_helo_hostname also checks for the existence of a '.')
[06:56:43] <r8> Oh, maybe there's an exception in reject_invalid_helo_hostname for something called "valid_hostaddr()"...
[06:58:52] <r8> "&& !valid_hostaddr(test_name, DONT_GRIPE)) /* XXX back compat */" Hrm. "back compat".
[06:59:13] *** tombar_ has quit IRC
[07:00:39] <r8> I'm guessing this is related to my earlier grousing about the RFCs being unclear. Ultimately it seemed like they were saying that an all-numeric hostname was bogus.
[07:01:52] <f3ew> yes :)
[07:02:04] <r8> My gut tells me this is a sensitive issue.
[07:02:21] <f3ew> nope
[07:03:59] <r8> Maybe we can get a config or compile parameter that says treat numeric hostnames as bogus? strict_rfcs_810_920_952_1035_1123_2181_hostnames?
[07:18:32] <r8> Or maybe I can just add some local patches to my pkgsrc. Do you know why it's saying "back compat"? Any idea what breaks if I forego this?
[07:21:05] <vice-versa> I would guess backward compatibility with expected behaviour in previous postfix versions
[07:22:47] <r8> Hm. That makes sense.
[07:26:00] *** amrit|wrk is now known as amrit
[07:26:04] *** ming_zym has joined #postfix
[07:27:31] *** hparker has quit IRC
[07:34:39] *** madrescher has joined #postfix
[07:47:02] <seekwill> So I have this problem... postfix only listens to localhost, and not eth0...
[07:47:31] <f3ew> postconf inet_interfaces
[07:47:45] <seekwill> :)
[07:47:47] <seekwill> Thanks!
[07:48:23] <seekwill> I didn't know it defaults to listening to localhost
[07:50:49] <f3ew> It doesn't
[07:50:50] <sysmonk> it doesn't ...
[07:50:53] <f3ew> It's a distro thing
[07:50:59] <seekwill> Stupid CentOS :)
[07:51:19] <seekwill> But I understand
[08:04:59] <milligan_> I'm getting a lot of double bounces. A spammer sends an email with a spoofed adress, I send a reply to the spoofed address that the account they tried to reach doesn't exist, and then I get a reply back, that the spoofed address I tried to reach doesn't exist. Is there any way to get rid of that ?
[08:05:45] <seekwill> Check recipients before accepting the message
[08:06:04] <milligan_> I do.
[08:06:27] <seekwill> From your description, you're sending an out of band bounce
[08:06:41] <seekwill> When the spoofer sends in RCPT TO: , you should check then, and reject.
[08:06:42] <sysmonk> my notebook is dead ;/
[08:06:46] <seekwill> sysmonk: :(
[08:06:54] <seekwill> I need fooooooooood
[08:06:56] <sysmonk> something went wrong with the display
[08:07:00] <seekwill> brb
[08:07:21] <sysmonk> when i start it, it doesn't show anything, and after some time it gives one long beep and two short beeps,
[08:07:27] * sysmonk is clueless how to fix that one
[08:07:50] <milligan_> seekwill, I could drop replies to non-existant accounts .. but that wouldn't be very polite to authentic senders .. if they typo an address or something.
[08:08:11] <seekwill> It actually would be VERY good for people who typo's
[08:08:19] <seekwill> Catchall's are bad
[08:08:20] <seekwill> brb
[08:08:39] <sysmonk> yeah, catchalls are evil
[08:08:47] <milligan_> I don't have a catchall. I simply send a reply, that an account doesn't exist if they try to reach a non-existant user.
[08:09:04] *** phnord has joined #postfix
[08:09:13] <sysmonk> wee
[08:09:17] <sysmonk> bounes are more evil
[08:09:19] <sysmonk> bounces*
[08:10:36] <milligan_> so I shouldn't bounce ?
[08:15:56] <f3ew> reject!
[08:16:22] <sysmonk> yeah
[08:16:27] <sysmonk> and that's what postfix does by default.
[08:21:08] <milligan_> Yes, it does reject.. but that generates an email that's attempted delivered to the other end, doesn't it? I guess that might be a problem in the other end, and not in mine ?
[08:21:51] *** magyar has quit IRC
[08:22:21] <seekwill> Oh my
[08:22:49] <seekwill> milligan_: When your server rejects on RCPT, the legit sender's server will generate a bounce message to the sender
[08:23:47] <milligan_> yeah. And when the sender doesn't exist, their bounce will generate a reject, that's delivered back to me.
[08:23:56] <seekwill> No it wont
[08:24:10] <seekwill> A bounce has a null from
[08:24:18] <seekwill> And there is no "you"
[08:28:03] <seekwill> milligan_: Your email is foo at a dot com. You want to send an email to me at bar at b dot com. @a.com talks to @b.com, saying he has a message for baar. @b.com says "nope! u fail!", and @a.com sends email to foo at a dot com saying baar at b dot com does not exist.
[08:28:55] <seekwill> A spammer "MTA" won't process that bounce.
[08:28:56] <milligan_> yeah .. I think I found the problem here.
[08:29:06] <seekwill> Which is?
[08:29:48] <seekwill> BTW, how did you configure this "friendly" message?
[08:30:38] <milligan_> It's my quota plugin that's sending bounces .. not the "user doesn't exist"
[08:30:43] <milligan_> So I am in fact bouncing :-\
[08:30:46] <seekwill> oh
[08:31:04] <f3ew> milligan_, modify the quota plugin to update a DB instead
[08:31:05] <seekwill> You should stop that "user doesn't exist" too :)
[08:31:14] <f3ew> then Postfix can reject overquota immediately
[08:31:27] <milligan_> seekwill, the "user doesn't exist" is handled correctly :)
[08:32:01] <seekwill> milligan_: if you say
[08:35:28] <seekwill> Is there any way to have SA scan the email during the smtp process?
[08:35:54] <seekwill> ... before accepting the message?
[08:38:37] <sysmonk> before queue filtering
[08:38:40] <sysmonk> !before_queue
[08:38:40] <knoba> sysmonk: Error: "before_queue" is not a valid command.
[08:38:43] <sysmonk> doh
[08:39:03] <sysmonk> www.postfix.org/SMTPD_PROXY_README.html
[08:39:19] *** Fallenou has joined #postfix
[08:39:53] <seekwill> ah, thanks
[08:40:01] *** amrit is now known as amrit|zzz
[08:43:25] *** jeffspeff has quit IRC
[08:43:36] *** JaccoH has quit IRC
[08:54:33] * sysmonk opens up a bottle of beer
[08:54:36] <sysmonk> so, good morning guys!
[08:54:38] <sysmonk> :)
[09:02:28] <glad_work> morning + beer = alco
[09:02:56] *** sophokles has joined #postfix
[09:07:16] *** client52123 has joined #postfix
[09:07:28] <client52123> Hi all
[09:08:33] <sysmonk> glad_work: alcowhat?!
[09:08:34] <sysmonk> ;)
[09:08:59] <sysmonk> vacation + morning + beer = good!
[09:09:23] <sysmonk> workday + morning + beer = paxmel
[09:09:44] * sysmonk doesn't know how's it called in english, but russian people will understand me :)
[09:10:19] <client52123> i am new to postfix
[09:10:59] *** client52123 is now known as roy52
[09:12:12] *** lumpek has joined #postfix
[09:12:41] <roy52> now i am getting error unknown user when i send mail thru telnet
[09:12:45] *** DarklyCute has joined #postfix
[09:12:55] <roy52> can anyone plz help me
[09:12:57] <sysmonk> roy52: /topic
[09:13:09] <sysmonk> oh noes, please, not 'plz' guys
[09:13:26] <roy52> thx :)
[09:13:30] <seekwill> sysmonk: !!!! plz help m3!
[09:14:28] <sysmonk> seekwill: you forgot plzplzplzplz
[09:14:31] <seekwill> oh
[09:14:39] <sysmonk> and <arrow up>enter (x5)
[09:15:39] *** lumpek has quit IRC
[09:15:51] *** lumpek has joined #postfix
[09:15:59] <roy52> ?
[09:16:10] <seekwill> lol
[09:16:54] *** SeJo has quit IRC
[09:17:10] *** SeJo has joined #postfix
[09:19:08] *** Zblakany has joined #postfix
[09:19:44] <seekwill> Gah, this is weird.
[09:20:12] *** brancaleone has joined #postfix
[09:20:37] <seekwill> My message gets bounced because of "User unknown in virtual alias table". But it's getting bounced from another postfix instance somehow...
[09:21:13] <sysmonk> seekwill: bounced? or rejected?
[09:21:21] <seekwill> bounced.
[09:21:33] *** mcblady has quit IRC
[09:21:37] <seekwill> Which is what confuses me
[09:21:48] <sysmonk> so it passes through virtual_*_maps, but can't store it, right?
[09:21:58] <sysmonk> what do you use for storing?
[09:21:59] <seekwill> I'm not sure...
[09:22:01] <seekwill> maildir
[09:22:09] <sysmonk> seekwill: /topic !
[09:22:10] <sysmonk> ;)))
[09:22:13] <sysmonk> logs, postconf -n....
[09:22:18] <roy52> status=bounced (unknown user: "test")
[09:22:19] <seekwill> I'm working on that...
[09:22:52] *** mcblady has joined #postfix
[09:23:25] <seekwill> oh
[09:23:27] <seekwill> gah
[09:24:21] <sysmonk> roy52: learn to read, please
[09:24:29] <sysmonk> sorry, that should have been 'plzplz'
[09:24:30] *** roy52 has quit IRC
[09:25:02] <seekwill> oooh
[09:25:06] <seekwill> I know
[09:25:14] <sysmonk> seekwill: plzplzplzplz pastebin
[09:25:14] <sysmonk> ;)
[09:25:25] <seekwill> No, I don't want to show you my postfix n00bz
[09:25:39] <sysmonk> seekwill: you're a lol!!!!
[09:25:44] <seekwill> :(
[09:26:23] *** no_maam has quit IRC
[09:26:34] *** no_maam has joined #postfix
[09:27:22] <seekwill> sysmonk: Did I tell you who I worked for?
[09:27:31] <sysmonk> nope
[09:27:34] <seekwill> oh
[09:27:39] <sysmonk> oh, WORKED or WORK ?
[09:27:39] <sysmonk> ;)
[09:27:42] <seekwill> Work :) haha
[09:27:46] <seekwill> I still work there
[09:28:03] <sysmonk> yeah, i think you did
[09:28:06] <seekwill> oh
[09:28:12] <sysmonk> although i don't remember if you explictly told the company name
[09:28:12] <seekwill> ssshhh
[09:28:17] <seekwill> oh
[09:28:34] * sysmonk greps the logs
[09:29:22] *** Motoko-chan has quit IRC
[09:30:20] *** Trengo has quit IRC
[09:30:27] *** loompek has quit IRC
[09:30:31] *** loompek has joined #postfix
[09:31:07] <sysmonk> haha
[09:31:13] <sysmonk> i like the grep
[09:31:34] <sysmonk> lemme paste it to your pm
[09:31:38] <seekwill> heh
[09:32:03] <sysmonk> very informative grep output :))))
[09:32:10] <seekwill> I laugh a lot
[09:32:19] <f3ew> hmmm?
[09:32:20] *** Knoedel2 has joined #postfix
[09:32:20] <seekwill> I remember that conversation!
[09:32:32] <f3ew> grep reject: /var/log/maillog | perl -pe 's/^.+ to=(.*)$/$1/' | cut -f1 -d ' ' | tr -d \> | cut -f2 -d @ |sort | uniq -c | sort -k1 -n
[09:32:40] <f3ew> New favorite command
[09:34:54] <Knoedel2> is it possible to call amavis like my greylisting daemon over smtpd_restriction_classes ?
[09:34:55] *** lumpek has quit IRC
[09:35:36] <Knoedel2> so enable/disable amavis for individuell mailaccounts
[09:36:37] <sysmonk> f3ew: ;/
[09:36:43] <sysmonk> f3ew: your command sucks
[09:37:28] <sysmonk> f3ew: hmmm... wait...
[09:37:45] <sysmonk> f3ew: you're getting the uniq -c helo's which were rejected?
[09:38:00] <f3ew> domains in the to=
[09:39:10] <f3ew> I know
[09:40:17] <sysmonk> then maybe it's allright
[09:40:57] <sysmonk> although i'd do to=<.*\ at ( dot *)> proto=.*/
[09:41:12] <sysmonk> but hey, it's unix, 10000 ways to do one thing :P
[09:41:35] <f3ew> I know
[09:41:43] <f3ew> but that was a really quick oneliner
[09:48:08] <seekwill> sysmonk: arg, that wasn't it..
[09:48:10] <seekwill> here you go
[09:49:37] <sysmonk> seekwill: pastebin master.cf
[09:49:53] * f3ew cries
[09:49:54] *** Jax has joined #postfix
[09:49:57] <seekwill> oh
[09:50:00] <seekwill> aye
[09:50:32] <seekwill> I forgot about this file hehe
[09:50:37] <f3ew> flippy network
[09:51:38] <f3ew> perl -ne 'if (/^Aug 29.+reject:/) { /^.+ to=<.+ at ( dot *)> proto=.*$/; print $1,"\n" }' < /var/log/maillog | sort | uniq -c | sort -n -k 1
[09:51:45] <f3ew> Now to reduce that further
[09:53:57] *** DarklyCute has quit IRC
[09:54:02] <sysmonk> seekwill: why do you have will at gotstuc dot com will at gotstuc dot com in virtual_alias_maps ?
[09:54:18] <seekwill> sysmonk: That's what the book says? :)
[09:54:27] <sysmonk> seekwill: um, what book?
[09:54:31] <seekwill> The Book of Postfix
[09:54:41] <seekwill> The Book
[09:54:53] <sysmonk> which page?
[09:55:23] <sysmonk> btw, this one? 2. no.starch.the.book.of.postfix.state.of.the.art.message.transport.mar.2005.ebook.ddu.pdf
[09:56:20] <f3ew> yes
[09:56:37] <f3ew> sysmonk, that prevents the address from hitting a catchall
[09:56:57] <seekwill> Oh, I read the page kinda wrong... Page 191. I only need that will@ will@ twice if I use a catchall (which I am not)
[09:57:04] <seekwill> f3ew: :)
[09:57:09] <sysmonk> yeah
[09:57:13] <sysmonk> but he doesn't have catchall
[09:57:17] <sysmonk> that's why i'm asking
[09:57:23] <seekwill> ah
[09:57:45] *** war9407 has joined #postfix
[09:57:54] *** xming_ has joined #postfix
[09:58:18] <seekwill> So is that a bad thing?
[09:58:31] <sysmonk> nah, just asking
[09:58:40] <seekwill> oh
[09:58:45] <sysmonk> seekwill: virtual_alias_domains has gotstuc.com
[09:59:06] <seekwill> So I need to add virtual_* to the master.cf?
[09:59:57] <seekwill> ok?
[10:00:28] <sysmonk> nono :) wait i'm looking at it
[10:01:48] <sysmonk> imho you have to take out gotstuc.com from virtual_alias_domains
[10:02:43] <seekwill> Hmm.. it's not master.cf, the part about 127.0.0.1:10025? Maybe I need to add stuff to that to tell it about virtual users?
[10:02:55] <sysmonk> seekwill: no, it has that stuff from main.cf
[10:03:00] <seekwill> oh
[10:03:16] <sysmonk> you overwrite the stuff in master.cf with -o
[10:03:28] <sysmonk> everything else is 'invoked' from main.cf
[10:03:35] <seekwill> If I take it out of virtual_alias_domains, where do I list it as a domain?
[10:03:37] <seekwill> ah
[10:03:56] <sysmonk> uh
[10:03:57] <sysmonk> waiiit
[10:04:05] <seekwill> :)
[10:04:07] <sysmonk> where is your virtual_mailbox_domains ?!
[10:04:51] <seekwill> Book didn't say to put one?
[10:05:06] <sysmonk> book used local delivery?
[10:05:24] <seekwill> I'm confused :)
[10:05:33] <sysmonk> which page is it?
[10:05:35] <sysmonk> i'll look at it
[10:05:47] <seekwill> 190-191
[10:06:23] *** Draecos has joined #postfix
[10:06:46] <seekwill> oh, pg 196...
[10:06:46] <sysmonk> seekwill: it only talks about aliases
[10:06:49] <sysmonk> yeah
[10:06:54] <seekwill> ah
[10:06:55] <sysmonk> page 196 talks about virtual delivery
[10:07:49] <seekwill> Does that mean I don't need virtual_alias_*?
[10:07:57] <seekwill> Just virtual_mailbox_base|maps?
[10:07:58] <sysmonk> if you want to have aliases - yes
[10:08:38] <seekwill> I'm confused with virtual_alias_domains and virtual_mailbox_domains
[10:08:47] <seekwill> I thought...they were the same...
[10:09:16] <sysmonk> seekwill: don't ALIAS and MAILBOX differ to you ? :)
[10:09:33] <seekwill> umm...
[10:10:34] *** xming has quit IRC
[10:14:01] *** hark has joined #postfix
[10:15:08] *** F6F has joined #postfix
[10:37:23] *** rootsvr has joined #postfix
[10:46:33] *** hark has quit IRC
[10:46:40] *** hark has joined #postfix
[10:46:50] *** rootsvr has quit IRC
[10:52:57] *** xming_ has quit IRC
[10:53:56] *** iKs has quit IRC
[10:56:22] *** denis_ has joined #postfix
[10:57:21] *** loompek has quit IRC
[10:57:30] *** loompek has joined #postfix
[10:57:34] *** xming has joined #postfix
[11:01:41] *** cyclops has joined #postfix
[11:01:53] <cyclops> hi all
[11:01:58] <cyclops> I'm back!!
[11:06:40] <seekwill> Great! I can go to sleep now
[11:06:42] <seekwill> nite
[11:06:47] <seekwill> THANKS SYSMONK!
[11:06:49] *** seekwill has quit IRC
[11:13:49] <f3ew> lol
[11:40:36] *** xming has quit IRC
[11:44:09] *** goldfisc1li is now known as goldfischli
[11:44:16] *** drindt has joined #postfix
[11:46:49] *** aron is now known as Aron
[11:54:56] *** leyoda has joined #postfix
[11:55:32] *** Jax has quit IRC
[11:57:09] *** leyoda has left #postfix
[11:59:05] *** Draecos has quit IRC
[12:03:45] *** mark-use has joined #postfix
[12:13:33] <Knoedel2> i get this error: Recipient address rejected: Access denied
[12:14:01] <Knoedel2> if i allow the internet ip, it will work
[12:14:19] <Knoedel2> but then other users can connect from inet to port 10025 etc.
[12:14:25] <Knoedel2> how can i solve this ?
[12:17:30] <sysmonk> allow only to connect from localhost to 10025
[12:17:38] <sysmonk> or wherever you want it to allow from
[12:18:11] <Knoedel2> i have
[12:18:21] <Knoedel2> mynetworks=127.0.0.1/8
[12:18:30] <Knoedel2> in master.cf
[12:18:48] <Knoedel2> -o mynetworks=127.0.0.0/8
[12:19:25] <sysmonk> so, it should work ( if you have permit_mynetworks)
[12:21:15] <Knoedel2> i have
[12:21:44] <Knoedel2> !permit_mynetworks
[12:21:44] <knoba> Knoedel2: "permit_mynetworks" : Permit the request when the client IP address matches any network or network address listed in $mynetworks. Can be used in smtpd_*_restrictions.
[12:22:29] <Knoedel2> main.cf mynetworks = 127.0.0.0/8, 78.xx.xx.xx, localhost
[12:22:49] <sysmonk> Knoedel2: postconf -n + master.cf + logs && pastebin it
[12:28:16] <sysmonk> where are the logs?
[12:28:19] <sysmonk> i didn't ask for main.cf
[12:29:55] <Knoedel2> sec
[12:31:57] *** FallOnMe has joined #postfix
[12:34:44] *** the_5th_wheel has joined #postfix
[12:35:11] <the_5th_wheel> Hi. How can i set the maximum amount of recipients an email can have in postfix?
[12:37:12] <Knoedel2> do you need something else sysmonk or is this enough ?
[12:37:41] <f3ew> the_5th_wheel smtpd_recipient_limit
[12:39:43] <sysmonk> Knoedel2: add 78.47.80.80 to your mynetworks on :10020
[12:39:58] *** xming has joined #postfix
[12:39:58] <PaSzCzUs> anyone using vacation with postfix?
[12:40:12] <PaSzCzUs> autoresponder?
[12:40:38] <sysmonk> Knoedel2: although i don't get it why it connects with that source ip. what os are you using?
[12:41:09] <the_5th_wheel> f3ew: thanks
[12:41:13] *** Trengo has joined #postfix
[12:41:45] <Knoedel2> sysmonk that work but now i can also access this port over internet
[12:42:13] <sysmonk> Knoedel2: what os is it?
[12:42:18] <Knoedel2> freebsd
[12:42:22] <sysmonk> jails?
[12:42:27] <Knoedel2> vmware
[12:42:47] <sysmonk> but is the thing in a jail or not?
[12:43:48] <Knoedel2> i havent installed the system, so i dont know
[12:44:08] <Knoedel2> it was before in a jail
[12:44:17] <Knoedel2> then he ported it to another vmware
[12:44:25] <sysmonk> Knoedel2: what's the output of sysctl security.jail.jailed
[12:45:27] <Knoedel2> mx1# sysctl security.jail.jailed
[12:45:27] <Knoedel2> security.jail.jailed: 1
[12:45:40] <sysmonk> then yes, you are still in a jail
[12:45:57] <sysmonk> poor thing :) you're on vmware + in a jail
[12:46:07] <Knoedel2> hehe
[12:46:19] <sysmonk> Knoedel2: sadly, but jails don't have their own loopback interface yet
[12:46:30] <sysmonk> Knoedel2: that's why you're connecting with a 'remote' ip
[12:46:37] <Knoedel2> ahhh
[12:47:07] <sysmonk> that's also why you can connect to :10020
[12:47:12] <Knoedel2> can i solve it without new installation ?
[12:47:34] <sysmonk> what do you want to solve? postfix won't accept any emails to :10020 anyway because it has restrictions with permit_mynetworks
[12:47:42] <sysmonk> (except from you)
[12:49:08] <Knoedel2> is there a way to say freebsd you are not in jail and use lo interface
[12:49:16] <sysmonk> hehe
[12:49:19] <sysmonk> you ARE in a jail
[12:50:21] <Knoedel2> prison break ;)
[12:50:27] *** cilly has joined #postfix
[12:50:30] <sysmonk> yeah, go break the jail ! :)))
[12:51:13] *** boink__ has joined #postfix
[12:51:13] <Knoedel2> thanks sysmonk
[12:54:56] *** cpm has joined #postfix
[12:55:42] *** Internat has quit IRC
[12:55:56] *** Internat has joined #postfix
[12:56:40] * cpm starts screaming, really loudly
[12:56:56] <f3ew> Why?
[12:56:59] * sysmonk mutes cpm
[12:57:18] <cpm> f3ew, I can't tell ya, I've been muted by sysmonk.
[12:57:25] <Knoedel2> going home see ya later
[12:57:31] *** Knoedel2 has quit IRC
[12:57:32] <cpm> No reason, just felt like screaming.
[12:57:37] <cpm> G'night
[12:57:55] <sysmonk> heh :)
[13:00:57] *** xming_ has joined #postfix
[13:01:25] *** knoba has quit IRC
[13:01:31] *** knoba` has joined #postfix
[13:03:10] *** knoba` is now known as knoba
[13:03:21] *** raz has quit IRC
[13:03:21] *** boink__ is now known as raz
[13:12:44] *** xming has quit IRC
[13:16:25] <Roobarb> anyone using dkim-milter ?
[13:23:10] <cpm> probably, seeing how someone went to all the trouble to write it and everything.
[13:23:57] *** felix-da-catz_zz has quit IRC
[13:25:27] *** rootsvr has joined #postfix
[13:27:45] *** cyclops has quit IRC
[13:28:40] *** standel has joined #postfix
[13:29:01] <dragonheart> Roobarb: yes . me even.
[13:29:18] <dragonheart> cpm: you could get more scientific and count the sf.net downloads
[13:29:39] <cpm> dragonheart, yeah, but downloads only imply use.
[13:29:59] <dragonheart> yeh true
[13:30:27] *** rootsvr has quit IRC
[13:30:27] <cpm> I stand by my 'probably'. Probability implies that someone is in fact, using it. So the poll answer may be 'Yes'. As 'No' is unlikely.
[13:30:52] <cpm> So, Roobarb now knows everything he needs to. We've done a good deed today. Time to go back to bed.
[13:31:20] <cpm> Or start screaming, maybe.
[13:34:23] <dragonheart> who knows. go to bed :-)
[13:35:07] *** mark-use has quit IRC
[13:38:02] <Roobarb> cpm: you're too kind...
[13:38:10] <Roobarb> anyone *in here* using dkim-milter ?
[13:39:01] <Roobarb> dragonheart: ok, I've just set it up and I'm getting success when sending to some dkim reflector addresses...
[13:39:42] <Roobarb> I'm a little confused as to whats supposed to happen to mail proporting to be from my domain, but that isn't signed
[13:42:33] *** Fallenou has quit IRC
[13:47:50] *** knoba has quit IRC
[13:48:53] *** knoba has joined #postfix
[13:49:00] *** randra has joined #postfix
[13:50:03] *** ming_zym has quit IRC
[13:52:44] *** hark has quit IRC
[13:53:02] *** hark has joined #postfix
[13:59:47] *** rootsvr has joined #postfix
[14:01:40] *** tux_ has joined #postfix
[14:01:42] <tux_> hi
[14:02:15] <tux_> i'm migrating my mail server from one server to another, and while i'm doing that i'd like to bounce all messages back to their senders with a default message
[14:02:17] <tux_> any idea ?
[14:03:20] <dragonheart> tux_: setup the first to relay to the second until the MX ttl expires
[14:04:24] <tux_> dragonheart, any rtfm ?
[14:04:57] <Roobarb> dragonheart: ah, so I need an additional TXT record?
[14:05:06] <dragonheart> Roobarb: yep.
[14:05:11] <Roobarb> ok thanks
[14:05:16] <Roobarb> I'll go read dome more
[14:05:18] <Roobarb> *some
[14:05:32] <dragonheart> happy reading
[14:06:24] *** cyclops has joined #postfix
[14:06:35] <dragonheart> tux_: assuming the first is postfix look at the documents for making the first a backup MX pointing to the primary which is the replacement
[14:06:46] <tux_> both are postfix
[14:07:01] <tux_> :p
[14:07:47] <tux_> dragonheart, as i was saying, any rtfm link ? (have searched already for a quick & dirty solution like that but was not able to found any really usable docs)
[14:08:00] <tux_> (i'm not asking to ask..)
[14:10:04] <tux_> thanks
[14:11:07] <tux_> have fun
[14:11:10] *** tux_ has quit IRC
[14:11:26] * Roobarb wishes RFC's would include some freaking examples insteaf of ABNF forms
[14:11:32] <Roobarb> bastards
[14:11:49] <cpm> yeah, bastards.
[14:12:24] <cpm> Fortunately, the IETF is essentially an open forum. So, all you need do, is participate, and you can make the intarweb world a better place!
[14:12:46] * cpm made that mistake/misjudgement.
[14:12:57] <Roobarb> or, I could beat myself in the scrotum for an hour, and have the same effect
[14:13:08] <dragonheart> _adsp._domainkey.{domain}. 3600 IN TXT "dkim=discardable"
[14:13:22] <cpm> No no, only a single hour of scrot beating can't even touch the joy of endless ietf wrangling.
[14:13:43] <dragonheart> if you want real fun try to join a ietf working group
[14:13:48] *** spion has quit IRC
[14:13:50] <Roobarb> adsp-dkim-tag = %x64.6b.69.6d *WSP "=" *WSP
[14:13:50] <Roobarb> ("unknown" / "all" / "discardable")
[14:13:54] <cpm> dragonheart, did that.
[14:14:14] <Roobarb> who here knows that "%x64.6b.69.6d" actually means "dkim" ?
[14:14:17] <cpm> yeah, it's similar to scrot beating, or nailing your feet to the floor repeatedly.
[14:14:18] * f3ew is on a WG
[14:14:19] <dragonheart> oh. my sympathies
[14:14:25] <cpm> f3ew, which one?
[14:14:34] <f3ew> the antispam
[14:14:37] <f3ew> though that's more of a RG
[14:14:38] <cpm> ah. cool. I was on marid
[14:14:49] <f3ew> Roobarb, man 7 asii
[14:14:53] <f3ew> ascii
[14:14:55] <Roobarb> hush child ;)
[14:16:09] *** spion has joined #postfix
[14:16:24] <Roobarb> dragonheart: do you know any reflectors other than autorespond+dkim at dk dot elandsys.com ?
[14:18:24] *** FallOnMe is now known as jelly-work
[14:19:33] *** hark has quit IRC
[14:19:56] *** hark has joined #postfix
[14:20:34] <Roobarb> meh
[14:21:00] <Roobarb> can't seem to get the adsp record working
[14:22:19] *** jelly-work is now known as jelly
[14:22:22] <dragonheart> its fairly new - i wouldn't be supprised if the autoresponders aren't up to date yet
[14:24:06] *** Nockian has quit IRC
[14:26:41] <Roobarb> you ever seen this error before: SSL error:04067069:rsa routines:RSA_EAY_PUBLIC_DECRYPT:pkcs1 padding too short
[14:30:44] *** Zeit|awy_ has quit IRC
[14:38:45] <dragonheart> yes. can't remember why. there's a trick to decoding ssl error numbers that i can't remember
[14:38:57] <dragonheart> anyway gn. sorry can't stay around
[14:46:57] <cpm> Wow, load averages over 100. I'm impressed with this old box man.
[14:55:22] *** shasta has joined #postfix
[14:59:17] *** rootsvr has quit IRC
[15:01:41] <cpm> that's about the worst spike I've ever seen. And I was still able to work from the shell. saw over 120
[15:01:49] *** Haris1 has joined #postfix
[15:02:45] *** saurabhb has quit IRC
[15:03:27] *** madrescher has quit IRC
[15:04:02] *** cilly has quit IRC
[15:04:10] <sysmonk> cpm: heh, my usual load avg is ~50
[15:04:38] <cpm> yeah ~50 makes me pay attention, as it starts to impact performance.
[15:04:57] <sysmonk> depends what makes the load
[15:04:59] <cpm> but I usually run in the ~20 ~35 range.
[15:05:14] <sysmonk> usual 50+, some hours ~100
[15:05:18] <sysmonk> somtimes get's up to ~200-210
[15:05:58] <cpm> I never see loads like that. What on earth are you doing to get loads of 200?
[15:07:04] <sysmonk> nuffin, watching pron
[15:07:10] <sysmonk> hosting servers ;/
[15:07:50] <cpm> well, pr0n, of course.
[15:08:14] <cpm> that goes without saying. After all, that's the point.
[15:11:39] <jelly> shred webhosting can do that, if a client's web is getting DoSed
[15:13:00] <Haris1> the mail server is getting alot of mails with big attachments?
[15:17:18] *** ElDios has joined #postfix
[15:17:21] <ElDios> hey guys..
[15:17:29] <ElDios> weird problem with PCRE
[15:17:51] <ElDios> if I test it with postap -fq "string" pcre:file it works like a charm
[15:17:56] <ElDios> but then when in use in the main.cf
[15:18:01] <ElDios> it doesn't want to work
[15:18:06] <ElDios> anyone has a clue?
[15:18:42] <f3ew> uh?
[15:18:55] <ElDios> yeah.. strange indeed
[15:22:59] <f3ew> !debug
[15:24:02] <cpm> Haris1, no, recalcitrant users who refuse to manage their inboxes, all of sudden deciding to do a search by content in a 10 gig inbox with 10s of thousands of emails.
[15:28:42] *** st_iron` has joined #postfix
[15:28:44] <st_iron`> hello
[15:29:01] <st_iron`> my messages stick in the mailq
[15:29:06] <st_iron`> what is this? :/
[15:36:23] *** havvg has joined #postfix
[15:37:58] <ElDios> cpm 10gigs inbox? O_o
[15:38:03] <ElDios> wtf
[15:38:22] <cpm> ElDios, yup. Not uncommon with my lUsers
[15:38:46] <ElDios> lots of huge attachments I hope
[15:39:01] <cpm> Prolly.
[15:39:11] <cpm> I don't know. I can't look, it's too painful.
[15:39:15] <ElDios> :)
[15:39:44] <cpm> before enlightenment, chop wood, carry water, after enlightenment, chop wood, carry water.
[15:39:48] <cpm> I'm just the postmaster.
[15:40:02] <cpm> I suggest, they ignore.
[15:40:14] <cpm> 'just keep it running boy'
[15:40:20] <ElDios> ahahah
[15:40:46] <cpm> Ya know, there are basically two professions that refer to their clients as users.
[15:41:12] <ElDios> :P
[15:42:02] <cpm> that's not a coincidence. If this is what it's like, I'd really hate to be a drug dealer. That must really suck. They should start a union or something.
[15:42:22] *** mcblady has quit IRC
[15:42:36] <ElDios> eheh
[15:42:50] <sysmonk> cpm: :)
[15:43:01] <cpm> :)
[15:44:01] *** mcblady has joined #postfix
[15:44:33] *** memetic has joined #postfix
[15:44:49] <ElDios> goddamit
[15:45:00] <ElDios> postmap -q "X-pstn-levels: (S: 0.00052/3.34875 CV:99.9999 R:95.9108 P:49.0672 M:89.2203 C:99.7951 )" pcre:/etc/postfix/headers_check_scan.regexp
[15:45:00] <ElDios> FILTER smtp:127.0.0.1:10099
[15:45:30] <ElDios> but in the main work istance it doesn't work... this is really a WTF
[15:45:50] <ElDios> obviously that header has just been taken from a not matched mail
[15:45:58] <ElDios> that here matches perfectly
[15:46:06] <f3ew> show postconf -n!
[15:46:41] <ElDios> what section of postconf are you interested in f3ew?
[15:46:52] <f3ew> !debug
[15:46:56] <f3ew> See
[15:47:05] <ElDios> header_checks = pcre:/etc/postfix/headers_check_scan.regexp
[15:47:06] <ElDios> done
[15:47:18] <ElDios> and everything looks fine with -v
[15:47:31] <ElDios> ahrd debugging isn't helpful IMHO in this case
[15:47:36] <ElDios> hard* sorry
[15:50:35] <ElDios> full postconf -n
[15:50:47] *** st_iron has joined #postfix
[15:51:03] <ElDios> as you can see header_checks is in place and BTW
[15:51:10] <ElDios> is I use a .* regexp it works
[15:51:13] <ElDios> so..
[15:51:21] <ElDios> headers check is working
[15:51:40] <f3ew> so the specific entry isn't matching
[15:52:10] <ElDios> yes, but the strange part is that it match correctly with the postmap -q command
[15:52:14] <ElDios> :.(
[15:52:17] <ElDios> it's driving me mad
[15:53:15] <ElDios> AFAIK the results should be the same from the postmap -q and the main istance of postfix..
[15:53:19] <f3ew> yes
[15:53:27] <ElDios> well, in this case thery're not
[15:54:19] <f3ew> Is the header actually what is being seen by Postfix?
[15:54:26] <f3ew> Or is it encoded somehow?
[15:54:53] <ElDios> I took it from the messsage source of the recipient user client
[15:55:21] <ElDios> the message arrives with that header already in place to the mail system
[15:55:44] <ElDios> so I think we could say yes.. that is the message seen by postfix
[15:59:25] <f3ew> Not necessarily
[15:59:52] <f3ew> /^X-pstn/ WARN
[16:00:00] <f3ew> should log the header in your logs
[16:00:18] <ElDios> oks
[16:00:25] <ElDios> let's try
[16:03:29] <ElDios> yep
[16:03:30] <ElDios> Aug 29 16:04:19 mail-be-in-ch-srv0 postfix/cleanup[3706]: 43585605A7: warning: header X-pstn-levels: (
[16:03:33] <ElDios> BTW
[16:03:43] <ElDios> i don't think that's the part that make it screw
[16:04:05] <ElDios> I think that ---> :[\s]*([0-9]|[0-7][0-9]|8[0-5])\..*$/
[16:04:17] <ElDios> this is the part that it's confusing it
[16:04:28] <ElDios> sorry
[16:04:40] <ElDios> this is a complete line
[16:04:44] <ElDios> /X-pstn-levels: \(.*C:[\s]*([0-9]|[0-7][0-9]|8[0-5])\..*$/ FILTER smtp:127.0.0.1:10099
[16:05:11] <ElDios> after the [\s]* is making it insane
[16:05:40] <ElDios> the problem is that if that's not working, it should not match with the postmap -q command either
[16:06:43] *** denis_ has quit IRC
[16:06:56] *** st_iron` has quit IRC
[16:08:00] *** CosMiC_Touch has quit IRC
[16:13:19] *** st_iron has quit IRC
[16:16:44] * rob0 hesitates to ask ... what is the real goal of all this?
[16:16:56] <ElDios> :)
[16:17:20] <ElDios> simply speaking
[16:17:26] <ElDios> there's this field
[16:17:31] <ElDios> with a rating
[16:17:39] <ElDios> that can go from 0-85
[16:17:41] <ElDios> sorry
[16:17:43] <ElDios> 0-100
[16:17:49] <ElDios> and must match only from 0-85
[16:18:05] <ElDios> end
[16:18:40] <ElDios> X-pstn-levels: (S: 0.00052/3.34875 CV:99.9999 R:95.9108 P:49.0672 M:89.2203 C:99.7951 )
[16:18:43] <rob0> Seems to me like you should be making the routing decision in the content filter itself.
[16:18:45] <ElDios> this is a tipical line
[16:19:02] *** seekwill has joined #postfix
[16:19:20] <f3ew> and you want to filter based on the last bit
[16:19:46] <ElDios> f3ew no.. on what's between : and .
[16:19:56] <ElDios> like P:49.0672
[16:20:01] <ElDios> the interesting part is 49
[16:20:46] <f3ew> *C: <===
[16:21:21] *** denis_ has joined #postfix
[16:21:47] <ElDios> /(.* is because there could be stuff between the first parentheses and the pattern C:
[16:21:58] <ElDios> for example C: in this case is the last parameter
[16:23:18] <f3ew> /^X-pstn-levels: .+C:([0-7][0-9]|[8[0-5]|[0-9])\..*$/ FILTER 127.0.0.1:10025
[16:23:48] <ElDios> oke.. I'll try
[16:24:29] <rob0> we still haven't been told the real world goal :)
[16:24:41] <rob0> Also, I don't think there are leading zeroes.
[16:26:52] <ElDios> f3ew between : and the first number there could be a space
[16:27:02] <ElDios> that's why the [\s]*
[16:27:26] <ElDios> rob0 the purpose is to match only if the rating is between 0 and 85
[16:27:27] <ElDios> :)
[16:27:34] <ElDios> think of it as a spam service or such
[16:27:37] *** loglibrarian has joined #postfix
[16:33:52] *** saurabhb has joined #postfix
[16:34:01] *** saurabhb has quit IRC
[16:34:15] <loglibrarian> Is there a way to see settings for various services, such as pickup, cleanup, qmgr, etc.. via postconf?
[16:34:40] <Dominian> postconf -n ?
[16:34:49] <Dominian> dunno to tell you the truth..
[16:34:51] <Dominian> man postconf ?
[16:35:16] <f3ew> no
[16:35:59] <loglibrarian> all of that configuration must be done in master.cf and not through postconf?
[16:36:22] <f3ew> mostly
[16:37:12] <Dominian> oh yah
[16:37:16] <Dominian> postconf applies to main.cf
[16:37:17] <Dominian> duh
[16:37:22] <Dominian> my bad
[16:38:41] <rob0> well, a lot of postconf settings can be put in master.cf too.
[16:39:05] *** moimeme has joined #postfix
[16:39:08] <moimeme> Hi
[16:41:09] <moimeme> Anyone can please help me, I'm trying to not forward the ip address in the received field in postfix. Is there any way to hide the originating address and replace it by my smtp relay's address ? I don't want people to know from where I send my emails. thx a lot
[16:41:45] <ElDios> f3ew works now
[16:41:50] <ElDios> but with the [\s]*
[16:42:01] <Dominian> moimeme: no.. not really.. that's part of the "path" it takes..
[16:42:15] <f3ew> See header_checks /IGNORE/
[16:42:15] <ElDios> /^X-pstn-levels: .+C:[\s]*([0-7][0-9]|[8[0-5]|[0-9])\..*$/ FILTER 127.0.0.1:10025
[16:42:38] <moimeme> Dominian : thx, but isn't there a way to make the path initiate at the smtp relay ?
[16:43:08] <f3ew> moimeme see above
[16:44:19] <ElDios> thanks f3ew
[16:45:11] <moimeme> thanks a lot, I didn't see the line ;)
[16:47:42] *** Aron is now known as aron
[16:49:55] <rob0> BTW it can be risky. Make sure your header_checks line for your Received: header ONLY matches what you need to match. Also, take care that you don't get into a mail loop, as Postfix can't stop a mail loop when you're stripping Received: headers.
[16:50:02] <rob0> don't ask how I know this
[16:53:03] *** xpoint has joined #postfix
[16:53:18] *** denis_ has quit IRC
[16:55:00] *** drindt has quit IRC
[16:57:03] <Dominian> rob0: hehe
[16:57:08] * Dominian can venture a guess
[17:03:52] <ElDios> thnka uo
[17:03:53] <ElDios> you guys
[17:03:55] <ElDios> bye
[17:03:57] *** ElDios has left #postfix
[17:06:18] *** livid has joined #postfix
[17:06:56] *** standel has quit IRC
[17:19:38] *** jeev has quit IRC
[17:23:15] <moimeme> thx rob0 !
[17:24:17] *** st_iron has joined #postfix
[17:24:18] <st_iron> re
[17:24:19] <st_iron> sorro
[17:24:21] <st_iron> sorry
[17:24:34] <st_iron> I have the same problem like before
[17:24:43] <st_iron> the messages are stick in the mailq
[17:25:01] <st_iron> and don't be delivered to the maildirs :(
[17:25:14] <st_iron> why is this?
[17:25:28] <lunaphyte_> !logs
[17:25:29] <knoba> lunaphyte_: "logs" : by default, postfix logs to the mail facility of syslog. Something like grep -i `postconf -h syslog_facility` /etc/syslog.conf or grep -rl `postconf -h syslog_name` /var/log/* should tell you where logs are going.
[17:25:32] *** magyar_ has quit IRC
[17:25:44] *** magyar_ has joined #postfix
[17:25:58] *** moimeme has quit IRC
[17:26:03] <st_iron> hi magyar_
[17:26:20] *** stonith has joined #postfix
[17:26:46] <st_iron> lunaphyte: if you sent to me, there are nothing in the logs
[17:27:15] <rob0> In that case, no one can help until you fix your syslogd.
[17:28:33] <st_iron> rob0: I mean I can see the connects, the mailscanner activity and the queue IDs but they won't be delivered to the maildirs
[17:28:40] <stonith> Is it possible to rewrite e-mail addresses in postfix? I think there is and would like to do it in the 'virtual' file. For example, I have a user that receives mail at jbrown at mydomain dot com, jbrown would like to receive e-mails at jbrown+<whatever> at mydomain dot com (ie jbrown+684 at mydomain dot com, jbrown+685 at mydomain dot com, etc.). Do I do this with regular expressions or am I complicating things...
[17:28:41] <stonith> ...further? Thanks!
[17:29:06] *** xdie has joined #postfix
[17:29:12] *** xdie is now known as xDie
[17:29:50] <st_iron> rob0: do you have an idea what is it?
[17:30:43] <lunaphyte_> stonith: ADDRESS_REWRITING_README
[17:30:55] <stonith> lunaphyte_: thanks
[17:30:58] *** Fallenou has joined #postfix
[17:31:07] *** st_iron` has joined #postfix
[17:31:17] <st_iron`> re
[17:31:19] <st_iron`> net split...
[17:32:07] <st_iron`> I update my problem: messages come and go but this progress is so slowly and my mailq is grow and grow
[17:32:59] <rob0> Sure, sounds like a mailscanner problem.
[17:33:06] <lunaphyte_> st_iron: to get help here, you need to prove/demonstrate what you're describing is occurring.
[17:34:16] <rob0> !recipient_delimiter
[17:34:17] <knoba> rob0: "recipient_delimiter" : a configuration parameter in the main.cf: The separator between user names and address extensions (user+foo). See canonical(5), local(8), relocated(5) and virtual(5) for the effects this has on aliases, canonical, virtual, relocated and on .forward file lookups. Basically, the software tries user+foo and .forward+foo before trying user and .forward.
[17:35:11] <rob0> stonith, I would recommend "-" rather than "+" because I have found that a lot of clueless programmers think email usernames should not contain a +.
[17:35:38] <stonith> But RFC allows the use of + sign in e-mail addresses
[17:35:55] <rob0> Eventually I will switch, but the transition will be very painful (have to resubscribe all my mailing lists.)
[17:36:20] <rob0> stonith: being right doesn't mean your address will be accepted.
[17:36:29] <stonith> rob0: I hear you, i'm trying to get the regular expression worked out
[17:36:57] <rob0> why do you need a regular expression?
[17:38:58] <st_iron`> lunaphyte: so, I use postfix on debian etch with amavis (spamassassin and clamav), virtual users. when I get a mail it seems in the queue but it won't be delivered to the maildir for severeal minutes (e.g. 20-40min)
[17:40:33] <st_iron`> sorry for my english but it's not my native language
[17:41:20] <rob0> 15:28 < st_iron> rob0: I mean I can see the connects, the mailscanner activity and the queue IDs ...
[17:41:28] *** Fallenou has quit IRC
[17:41:35] <rob0> which is it, amavisd or mailscanner?
[17:41:45] <rob0> Let's all vote.
[17:41:47] *** Fallenou has joined #postfix
[17:41:56] *** Fallenou has quit IRC
[17:42:03] *** Fallenou has joined #postfix
[17:42:12] <st_iron`> rob0: amavisd
[17:43:14] <lunaphyte_> i vote amailscannerivsd
[17:43:31] *** standel has joined #postfix
[17:43:56] <rob0> 15:40 < mkhelp> you are more friendly in this channel then postfix rob0 ;)
[17:44:08] <rob0> (said in #iptables)
[17:44:29] <st_iron`> is that possible that one message in the queue can't be delivered and all the messages behind that are standing and waiting for ths one?
[17:44:54] <livid> I am thinking about switching away from sendmail to postfix - anyone have any good docs about differences/comparisons?
[17:45:56] <vice-versa> !mtacompare
[17:46:10] <livid> I knew there had to be something ;) thanks
[17:46:29] <st_iron`> livid: postfix users are more friendly ;)
[17:46:41] <livid> I like crusty old curmudgeons
[17:46:43] <livid> ;)
[17:46:46] <st_iron`> am I right rob0? :)
[17:47:07] <livid> hehe
[17:47:19] <st_iron`> my main problem is that I have ~2000 formal emails in the queue
[17:47:28] <st_iron`> and it's growing and growing
[17:48:14] *** st_iron has quit IRC
[17:48:26] *** j_s has joined #postfix
[17:48:47] <st_iron`> please! I trust in these friendly persons technical knowledge ;)
[17:52:25] <lunaphyte_> rob0: still having problems with people hijacking your nick eh? :p
[17:52:48] <lunaphyte_> if you ask me, the rob0 in #bind is the worst of them all.
[17:54:26] <livid> haha
[17:56:11] *** phnord has quit IRC
[17:57:03] *** loglibrarian has quit IRC
[18:00:03] *** jwit has quit IRC
[18:00:04] *** jwit_ has joined #postfix
[18:04:33] <st_iron`> good bye
[18:04:36] *** st_iron` has quit IRC
[18:05:16] *** sophokles has quit IRC
[18:09:43] *** redduck676 has joined #postfix
[18:11:10] *** Haris_ has joined #postfix
[18:11:32] *** Haris has quit IRC
[18:11:36] *** Haris_ is now known as Haris
[18:12:12] *** pitakill has joined #postfix
[18:12:48] *** redduck666 has quit IRC
[18:19:23] *** stonith has quit IRC
[18:23:01] *** gcleric has joined #postfix
[18:23:44] *** tshine has quit IRC
[18:23:45] *** jwit_ has quit IRC
[18:28:33] *** tom_wifi has joined #postfix
[18:29:33] <tom_wifi> Goodday all. II'm looking for rough machine specs I'll need to handle 100,000 emails daily.
[18:31:15] <seekwill> heh
[18:31:36] *** tshine has joined #postfix
[18:31:48] <seekwill> "/day" isn't a good spec since mail doesn't stream evenly over 24 hours
[18:32:14] <seekwill> Is this a corporate email server? is this for sending out mailing lists? etc
[18:32:19] *** jwit has joined #postfix
[18:32:28] <mcp> tom_wifi: we have a dual xeon with 4gb ram and have >300,000 to 500.000 mails / day
[18:32:36] *** jeev has joined #postfix
[18:32:48] <tom_wifi> most emails are during a 8hr period & the mailserver will just forward those on based on the forwarding setup inside of postfix.
[18:33:06] <seekwill> 100k/8 hours...
[18:33:16] <mcp> tom_wifi: no spam? no antivir? then you can still use a 486 ;)
[18:33:21] <seekwill> hehe
[18:33:21] <e_> content scanning?
[18:33:50] <tom_wifi> clam-av + spam assisin
[18:34:40] <e_> forget the 486 then
[18:34:57] <tom_wifi> :(
[18:35:06] *** ribasushi has quit IRC
[18:35:06] <tom_wifi> shame i've got two spare 486s
[18:35:33] <jelly> seekwill: my incoming streams look pretty much random these days, due to spam. No 24 hour cycle visible at all
[18:35:35] *** ribasushi has joined #postfix
[18:36:10] *** havvg has quit IRC
[18:36:15] <jelly> so /day might be as good as anything, as long as his machine is able to process peaks of 3-5 more traffic than usual
[18:36:21] <seekwill> jelly: True. I was more concerend about the 100k of legit mail
[18:36:27] <tom_wifi> I'm planning on using two; 2x dual core xeons + 4gb ram each & 3x 73gb raid5.
[18:36:50] <seekwill> Three drive RAID5?
[18:37:08] <tom_wifi> 3x 73gb hard drives running raid0
[18:37:11] <tom_wifi> 3x 73gb hard drives running raid5*
[18:37:17] <jelly> seekwill: ah yes, legit (well, those not detected as spam) mail graphs still look sensible
[18:37:32] *** hark has quit IRC
[18:38:02] *** hark has joined #postfix
[18:40:44] <thumbs> seekwill: I'm still bitter from last night's defeat
[18:40:54] <jelly> tom_wifi: we're filtering 4M/day using 4-5 machines that look like yours. The filter isn't SA but a perl-based commercial product. I guess 100k mails/day on your input side will be easily chewed by your machines
[18:41:47] <tom_wifi> thats grand. You guys hav a good evening. thanks
[18:42:23] *** tom_wifi has quit IRC
[18:42:26] <vice-versa> 100,000 legit per day = 1,000,000 plus delivery attempts
[18:42:35] <jelly> yep, pretty much.
[18:43:55] <jelly> still, he has twice the processing power that I have. I guess clamav+spamassassin+some quarantine solution can't be written much worse than sophos
[18:44:05] <jelly> oops, I meant $vendor
[18:44:08] <jelly> :-)
[18:46:09] * cpm chuckles
[18:46:57] <livid> heh
[18:49:28] <livid> do many of you run multiple mtas?
[18:50:20] <slurpee> i have a new postfix server running in debian. i am not using the PAMdb/local users for email. I am using virtual users in mysql. I have a virtual user john in mysql and no system user name john.
[18:50:39] <slurpee> when i send email to john, the email is recieved AND i recieved a bounceback saying john is not a valid user.
[18:51:01] <slurpee> when I adduser john to the system, the email goes through without the bounceback.
[18:51:08] <seekwill> thumbs: :)
[18:51:14] <seekwill> thumbs: Just your off day...
[18:51:33] <vice-versa> !unknown_virtual
[18:51:33] <knoba> vice-versa: "unknown_virtual" : \"User unknown in virtual $X table\" means that the recipient domain was found in $virtual_$X_domains but the username@domain was not found in $virtual_$X_maps. ("$X" can be either alias or mailbox .)
[18:51:55] <slurpee> so how can I fix?
[18:51:57] * slurpee is newb
[18:52:06] *** standel has quit IRC
[18:54:05] *** brancaleone has quit IRC
[18:55:12] *** magyar_ has quit IRC
[18:56:12] <vice-versa> hare_krishna
[18:56:50] <slurpee> i am worried the mailboxes might get mixed up with I have both mysql virtual users & virtual system users?
[19:00:09] *** jonez has quit IRC
[19:01:35] <rob0> If I had to run multipls MTA's I would be livid!
[19:03:28] *** hark has quit IRC
[19:06:32] *** Tinozaure is now known as _Tino
[19:06:41] *** _Tino is now known as Tinozaure
[19:08:27] *** stonith has joined #postfix
[19:13:21] <sysmonk> vice-versa: hare_krishna is da best :)
[19:18:35] <cpm> Caribou Gorn
[19:19:57] *** cyclops has quit IRC
[19:21:28] *** hparker has joined #postfix
[19:25:23] *** githogori has quit IRC
[19:26:35] * vice-versa ponders what a warp-capable, bipedal reptilian species has to do with caribou
[19:29:13] *** viyyer_ has joined #postfix
[19:30:37] <viyyer_> I have a authenticated smtp for one mail in domain.com(others are hosted on remote servers) and a mailman list for lists.domain.com. Can they co-exist?
[19:33:31] *** standel has joined #postfix
[19:38:38] *** gcleric has quit IRC
[19:38:48] <Dominian> yep
[19:40:10] *** xsamurai has joined #postfix
[19:41:49] *** xp_gone is now known as xp_prg
[19:45:44] <xsamurai> just checked my mail server after two weeks due to complains about delays in receiving emails, found 300,000k messages in the queue and when I try to flush nothing happens.
[19:48:22] *** cpm has quit IRC
[19:49:08] *** amrit|zzz is now known as amrit|wrk
[19:51:50] <vice-versa> flushing with the hopes everything will just spiral down to the sewer is wishful thinking at best
[19:52:35] <vice-versa> you need to investigate why there is a queue backlog in the first place
[19:53:27] <xsamurai> yeah i tried to grep all warning in the mail logs but its been 2 weeks and the previous logs have been overwritten
[19:54:18] <xsamurai> any other way to get the reason for the backlog ?
[19:55:04] <seekwill> Someone has been a naughty postmaster...
[19:57:15] <vice-versa> !obvious
[19:57:15] <knoba> vice-versa: "obvious" : look for obvious signs of trouble, egrep '(warning|error|fatal|panic):' /some/log/file See: !logs factoid if you're unsure of where your mail logs are located
[19:57:22] <vice-versa> !qshape
[19:57:23] <knoba> vice-versa: "qshape" : qshape(1) - The qshape program helps the administrator understand the Postfix queue message distribution in time and by sender domain or recipient domain.
[19:57:29] *** xsamurai has quit IRC
[19:57:43] <vice-versa> meh
[19:58:14] <vice-versa> think you hurt his feelings seekwill
[19:58:34] <seekwill> :(
[19:58:56] *** havvg has joined #postfix
[20:01:41] *** eye69 has quit IRC
[20:04:02] <thumbs> mean seekwill
[20:04:05] <thumbs> I hate seekwill
[20:04:08] <seekwill> :(
[20:04:18] *** lumpek has joined #postfix
[20:04:55] <seekwill> I'm about to sign up for a colo for a personal server... but then I realized... wtf am I going to do with it, other than hosting my personal email...
[20:05:45] <seekwill> I need something to host...
[20:05:55] <thumbs> I am looking for hosting too.
[20:06:06] <seekwill> We can host a DTD server!
[20:06:10] <thumbs> I'll run ftps, http and imap/smtp
[20:06:14] <thumbs> we can
[20:06:21] <jeev> seekwill, i have 7 servers i dont knw what to do
[20:06:27] <jeev> the other 13 are actually doing tihngs
[20:06:37] <jeev> 2 servers i'm about to number them and drop it off at the datacenter today
[20:06:43] <jeev> absolutely NO idea what i'm gonna do with them.
[20:06:44] <seekwill> thumbs: What is it for?
[20:06:57] <thumbs> personal music sites
[20:07:26] <thumbs> oh yes and some sql server too
[20:07:50] <seekwill> I want to host a mailing list
[20:08:05] <thumbs> we could run it off the same server
[20:08:27] <seekwill> We could. Run ESXI on the box and each have our own vm
[20:08:46] <thumbs> yeah.I want slack.
[20:08:53] <seekwill> I want CentOS
[20:09:16] <seekwill> thumbs: PM me your email. I want to show you the server I'm thinking about getting
[20:09:18] <thumbs> I thought you would.
[20:12:56] <sysmonk> personal what?
[20:13:11] *** madrescher has joined #postfix
[20:13:25] <sysmonk> seekwill: show me too :P
[20:14:19] <seekwill> sysmonk: Just a server I'm thinking about buying for a colo
[20:14:46] <Dominian> What he means when he says "buying for a colo" he means "buying and giving to" me
[20:14:51] <sysmonk> seekwill: show it to me :)
[20:15:12] <seekwill> It's a $1k box... :)
[20:15:15] <seekwill> sysmonk: email?
[20:15:31] <sysmonk> wow, i didn't know you have to have an email to give a link :)
[20:15:49] <seekwill> It's not a link
[20:15:56] <sysmonk> what is it?
[20:16:02] <sysmonk> pdf? txt? png? jpg?:)
[20:16:10] <sysmonk> don't say it's a .DOCX evilness
[20:16:10] <seekwill> html shopping cart
[20:16:19] <sysmonk> ogh :) good then :)
[20:17:14] <sysmonk> oh
[20:17:22] <sysmonk> i bought mine for a bit more than 1k$ usd
[20:17:33] <sysmonk> 1.5k it was
[20:18:23] <seekwill> This is a personal server, so trying to go cheap
[20:18:31] <sysmonk> mine is personal too
[20:18:36] <sysmonk> but it's a nice one
[20:18:43] <seekwill> You make more than me :(
[20:18:45] <sysmonk> although, it's a 2U ;/
[20:18:46] <seekwill> Rich boy
[20:18:50] <seekwill> oh
[20:19:10] <sysmonk> seekwill: nah, it's my devel + projects + lots of other stuff server
[20:19:24] <sysmonk> runing ~15 web sites on it
[20:19:44] <sysmonk> oh, and i'm irc'ing from it! :)
[20:20:05] <seekwill> ah
[20:20:12] <sysmonk> and, with this kind of server i won't be needing one for a few years
[20:20:38] <sysmonk> so, if we'll take that i'll be using this server for i.e. 3 years
[20:20:42] <sysmonk> that's 36 months
[20:20:52] <sysmonk> 1500 usd / 36 months = ~41 usd /month
[20:20:56] <sysmonk> not much, imho, for this beast
[20:21:30] <seekwill> yeah
[20:22:21] <sysmonk> + havin ILO is great
[20:22:41] *** loompek has quit IRC
[20:22:49] <sysmonk> and scsi disks
[20:22:53] <seekwill> ILO = integrated lights out?
[20:22:55] <seekwill> HP?
[20:22:59] <sysmonk> yeah
[20:24:07] *** magyar_ has joined #postfix
[20:24:10] *** githogori has joined #postfix
[20:24:13] <sysmonk> funiest thing is, i bought this server from a guy working in IBM :)
[20:24:27] <seekwill> heh
[20:25:06] *** pitakill has quit IRC
[20:28:50] *** Haris_ has joined #postfix
[20:29:17] *** xsamurai has joined #postfix
[20:29:41] <vice-versa> sysmonk: perhaps that's why he sold it ;)
[20:29:43] <xsamurai> vice-versa: sorry got dc
[20:30:23] <sysmonk> vice-versa: nah, he has some projects in some other organisation, which use[ds] HP's
[20:30:41] <vice-versa> what was the last you seen xsamurai?
[20:30:51] <xsamurai> vice-versa: but i saw your last message of qshape, heres the top 2 lines
[20:31:25] <xsamurai> vice-versa: i did grep for warning panic error but its been over 2 weeks and the logs have been overwritten so I cant find the original error that started the issue
[20:32:25] <vice-versa> sure, worth a shot with that egrep in case you overlooked anything
[20:33:01] <xsamurai> i keep seeing errors about flushing large q's at once, one sec let me recheck
[20:34:36] <vice-versa> xsamurai: are you doing address verification?
[20:35:59] <xsamurai> i'll google the errors
[20:37:56] <vice-versa> looks like possible hardware issues to me
[20:38:52] <xsamurai> thanks vice-versa
[20:40:21] <xsamurai> looks like a kernel issue, sstill checking
[20:44:01] <vice-versa> are you using large maxproc values in master.cf for anything?
[20:46:26] *** Haris has quit IRC
[20:46:27] <xsamurai> vice-versa: nope
[20:46:33] <xsamurai> vice-versa: should i be ?
[20:48:30] <vice-versa> no, was just thinking you may be exhausting kernel resources if some processes were set to something ridicules
[20:48:50] <xsamurai> well had some runaway rsync process awhile back on the server that ate up all the memory
[20:48:59] <xsamurai> but it was cleared about a week a go
[20:49:00] <xsamurai> ago
[20:49:33] <vice-versa> could be related
[20:49:58] *** danbeck has joined #postfix
[20:50:08] <xsamurai> vice-versa: yeah im thinking the same but the q should start clearning up but its not
[20:50:20] <vice-versa> are you seeing any other oddities in other logs or dmesg
[20:51:29] <jeev> is it possible to accept mail to virtual directory (which works) and also forward it to somewhere?
[20:51:53] <xsamurai> vice-versa: the only other complaints from postfix is about flushing the q when its a large size, but i've only flushed once or twice, I dont know why the msg keeps repeating
[20:52:53] <xsamurai> err saw the setting of qmgr_clog_warn_time
[20:52:59] *** Haris_ has quit IRC
[20:55:53] <vice-versa> xsamurai: postcat -q some of the queue ids to see it there are any clues
[20:56:53] *** viyyer_ has quit IRC
[20:57:04] *** tombar has joined #postfix
[20:59:19] <vice-versa> xsamurai: possibly, however they're smtp client delivery attempts so it's not looking good
[20:59:39] <xsamurai> vice-versa: running postcat, its taking awhile
[21:00:23] <vice-versa> running it how?
[21:01:13] <xsamurai> postcat -q like you pasted earlier
[21:01:35] <vice-versa> did you specify a queueid?
[21:01:52] <xsamurai> ...not really
[21:01:58] <vice-versa> lol
[21:02:44] <xsamurai> yeah just got the id from postqueue
[21:07:00] <xsamurai> i see a couple that are fake email addresses
[21:07:22] <vice-versa> no doubt
[21:07:51] <vice-versa> I guess the real question here is how is this shit making it into your queues ... are you using smtp auth? poor mynetworks choices?
[21:08:21] <vice-versa> let's see the output from postconf -n
[21:10:40] *** m1n3s6 has joined #postfix
[21:13:53] *** pitakill has joined #postfix
[21:15:54] <vice-versa> nothing stands out there
[21:17:17] <xsamurai> maybe if flush the queue in blocks of 25's
[21:17:35] <xsamurai> his it possible to do that without flushing the whole queue
[21:18:00] <xsamurai> s/his/is/
[21:18:09] <vice-versa> I'm not sure you want to be flushing anything at this point
[21:18:10] *** jonez has joined #postfix
[21:19:41] <xsamurai> should i write up something to weed out msgs in the Q, for invalid email addresses ?
[21:19:58] *** molnarp has joined #postfix
[21:20:07] *** pitakill has quit IRC
[21:20:41] *** tombar_ has joined #postfix
[21:21:19] <vice-versa> well you haven't yet discerned, or elaborated, exactly what's clogging the queues
[21:21:43] *** GoGi has joined #postfix
[21:21:50] <molnarp> Hi everyone! Could somebody please help me setting up mailman with postfix virtual domains?
[21:22:06] <xsamurai> vice-versa: ...yes
[21:26:06] <Dominian> molnarp: Do what I did.. research it.
[21:28:58] <xpoint> anyone have a php or html code to integrade mailman on own homepage ?
[21:29:59] <molnarp> Dominian: I did, but I couldn't find anything that helped me... I've set up a transport for lists.company.com to "mailman:", I've tried including the lists domain in virtual_domains table, and removing it, setting up relay between the two domains (company.com and lists.company.com), but id didn't work out
[21:30:16] <molnarp> that's why I'm asking you guys
[21:30:17] <Dominian> I'll give you a hint..
[21:30:20] <Dominian> mailman unix - n n - - pipe flags=FR user=mailman argv=/usr/local/mailman/bin/postfix-to-mailman.py
[21:30:46] <rob0> In a nutshell: virtual(5) aliases to redirect to local(8) aliases(5).
[21:30:46] <Dominian> you have to have the mailman service running..
[21:30:57] <rob0> (that's how I do it)
[21:31:12] <Dominian> not only that.. mailman has to be working properly.. when you hand off the lists to the mailman service it will check against its own "internal" alias database.
[21:31:14] <rob0> (Dominian's way works too)
[21:31:41] <Dominian> then liek you.. I have: lists.domain.tld mailman:
[21:31:59] *** stonith has quit IRC
[21:32:01] <Dominian> rob0: Yeah both ways work.. the awy I did it is what I've seen most recommend as it appears to be "easier"
[21:32:09] * Dominian shrugs
[21:32:20] <Dominian> I do like transport though hehe
[21:32:31] <Dominian> transport(5) that is
[21:32:56] <Dominian> got a dovecot transport configured as well for users that want to take advantage of IMAP sieve filters.
[21:33:18] <randra> dovecot is better courier?
[21:33:25] <Dominian> easier
[21:33:29] <Dominian> lots easier imo
[21:34:03] *** pirho has joined #postfix
[21:34:07] <molnarp> If I've got lists.company.com in virtual_domains, I get: Relay access denied, If I haven't, I get: Recipient address rejected: User unknown in virtual mailbox table
[21:34:07] <Dominian> Or in the case when a certain domain was having issues *looks at rob0* I used transport coupled with something else to allow them to still relay email through me till they got it fixed.
[21:34:17] <Dominian> eh
[21:34:28] <molnarp> I'd just like to simply pipe all mails for lists.company.com to mailman
[21:34:43] <rob0> !unknown_virtual
[21:34:45] <knoba> rob0: "unknown_virtual" : \"User unknown in virtual $X table\" means that the recipient domain was found in $virtual_$X_domains but the username@domain was not found in $virtual_$X_maps. ("$X" can be either alias or mailbox .)
[21:34:51] <Dominian> I have my lists.domain.tld in relay_domains and in relay_recipient_maps
[21:35:27] <Dominian> and in relay_recipient_maps for lists.domain.tld.. I have:
[21:35:39] <Dominian> list-announce-admin at lists dot domain.tld etc. etc. etc
[21:36:08] <Dominian> That way, I don't have some jack off trying to relay "junk at lists dot domain.tld" through my mailman service
[21:36:39] <Dominian> and to get even mroe confusing..
[21:36:42] <Dominian> for domain.tld
[21:37:09] *** Haris1 has quit IRC
[21:37:11] <Dominian> I have aliases for each of the @lists.domain.tld addresses.. so list-announce-admin at domain dot tld -> list-announce-admin at lists dot domain.tld ;0
[21:37:20] *** danbeck has quit IRC
[21:37:26] <Dominian> Just to make things easier.
[21:37:30] *** tombar has quit IRC
[21:37:30] * Dominian shaddups no
[21:37:31] <Dominian> w
[21:38:18] <xsamurai> vice-versa: so i checked and most of the invalid email msgs are being sent by the web app on the internal network
[21:39:51] <Dominian> molnarp: I don't think you can pipe * at lists dot domain.tld to mailman without a relay map
[21:39:52] *** Broken|work__ has quit IRC
[21:40:05] <Dominian> not sure if a relay map will take a catchall...
[21:40:05] *** Broken|work__ has joined #postfix
[21:40:19] *** jwit has quit IRC
[21:40:26] <rob0> sure it will
[21:40:27] *** jwit has joined #postfix
[21:40:40] <Dominian> @lists.domain.tld mailman:
[21:40:46] <Dominian> er.. actually ..
[21:41:02] <vice-versa> xsamurai: you might want consider excluding that host from mynetworks until you get this sorted out
[21:41:18] <Dominian> yeah..
[21:41:26] <Dominian> rob0: that looks right.. the @lists.domain.tld mailman:
[21:41:31] <Dominian> then postmap the transport file.. that should do it
[21:41:39] <Dominian> er.. relay map
[21:41:55] <xsamurai> vice-versa: seems like a signup script to flood the system
[21:42:13] <Dominian> xsamurai: spammers nailing a web form maybe?
[21:42:28] <vice-versa> I guess, regardless it's obviously an issue
[21:42:45] <xsamurai> Dominian: yep
[21:44:38] <xsamurai> so before i write a bash script to clean out all the fake email addresses related messages in the q , is there something already written for this purpose
[21:45:30] <vice-versa> xsamurai: something like, mynetworks = !10.1.0.222, 10.1.0.0/16, 127.0.0.0/8
[21:45:36] *** jwit_ has joined #postfix
[21:45:47] <vice-versa> replacing 10.1.0.222 with the ip of the host in question in case it wan't obvious ;)
[21:46:05] *** jwit has quit IRC
[21:47:35] <vice-versa> I've got a delete from queue by string script
[21:48:29] <xsamurai> thanks vice-versa, can you paste it on somewhere
[21:49:15] <vice-versa> it's basically this, find /var/spool/postfix -type f | xargs fgrep -l 'SOME_STRING' | sed 's,.*/,,' | postsuper -d -
[21:49:54] <vice-versa> try it first with out the | postsuper -d - to see what you're about to delete, that is if you care
[21:54:53] *** m1n3s6 has quit IRC
[21:54:58] *** m1n3s6 has joined #postfix
[22:01:36] *** Fallenou has quit IRC
[22:02:58] *** randra has quit IRC
[22:03:09] <xsamurai> btw what does postfix use to resolve domain names
[22:03:59] *** Zeit|awy has joined #postfix
[22:07:40] <Dominian> DNS
[22:10:18] <xsamurai> right, what tool
[22:10:20] <xsamurai> as in nslookup
[22:10:21] <xsamurai> dig
[22:10:23] <xsamurai> etc
[22:11:01] <shasta> your resolver
[22:11:18] <shasta> as in: appropriate system calls
[22:11:34] <shasta> man 3 gethostbyname
[22:12:07] <sysmonk> ;))
[22:12:08] <xsamurai> shasta: thanks
[22:12:45] <shasta> (I was *so* tempted to say "yo mamma" :P)
[22:13:09] <sysmonk> shasta: doesn't gethostbyname syscall execute rdp to a windows box and runs nslookup in it and returns the output to you ?!
[22:13:59] <jeev> is it possible to accept mail to virtual directory (which works) and also forward it to somewhere?
[22:14:35] <shasta> rdp? no. it exploits some common vulnerabilities to get access to the windows kernel and then uses TheUserWantsToKnowWhatIPaddressThisThingHas();
[22:15:26] <sysmonk> ah, nice
[22:16:57] *** f3ew has quit IRC
[22:18:19] *** f3ew has joined #postfix
[22:18:53] <rob0> shasta: wrong, not gethostbyname
[22:19:06] <rob0> DNS is used directly
[22:19:08] <xsamurai> shasta: if you used the "yo mamma" reply i'd retaliate with naked pictures of myself
[22:20:12] <rob0> I believe Postfix includes its own DNS client, but it will look in resolv.conf for nameservers to query.
[22:21:12] <xsamurai> rob0: thanks for the correction
[22:21:38] <shasta> is it?
[22:22:07] <shasta> $ ldd /usr/libexec/postfix/smtpd | grep resolv
[22:22:07] <shasta> libresolv.so.2 => /lib/tls/libresolv.so.2 (0xb7b85000)
[22:23:07] <shasta> oh well, I'm not that into UNIX, won't argue
[22:23:45] <rob0> ah, okay, but still, I know it queries DNS
[22:23:55] <rob0> !disable_dns_lookups
[22:23:56] <knoba> rob0: "disable_dns_lookups" : a configuration parameter in the main.cf: Disable DNS lookups in the Postfix SMTP and LMTP clients. When disabled, hosts are looked up with the gethostbyname() system library routine which normally also looks in /etc/hosts.
[22:24:26] <rob0> but apparently you're right about it using the system resolver libraries.
[22:24:39] *** danbeck has joined #postfix
[22:27:03] <vice-versa> isn't it getaddrinfo() if disable_dns_lookups is enabled?
[22:32:24] *** internat1 has joined #postfix
[22:32:24] *** Internat has quit IRC
[22:35:12] <xsamurai> alright guys thanks for all the help
[22:35:15] <xsamurai> im out
[22:35:28] *** xsamurai has quit IRC
[22:35:45] *** m1n3s6 has quit IRC
[22:35:48] *** m1n3s6 has joined #postfix
[22:44:53] *** molnarp has quit IRC
[22:45:05] *** standel has quit IRC
[22:53:51] *** xDie has quit IRC
[22:56:22] *** Nockian has joined #postfix
[23:02:47] *** F6F has quit IRC
[23:02:56] *** F6F has joined #postfix
[23:06:02] *** m1n3s6 has quit IRC
[23:06:14] *** m1n3s6 has joined #postfix
[23:11:32] *** rootsvr has joined #postfix
[23:23:26] *** rootsvr has quit IRC
[23:29:06] *** F6F has quit IRC
[23:38:50] *** Zblakany has quit IRC
[23:44:38] *** syllogism has quit IRC
[23:44:55] *** brad_ has joined #postfix
[23:58:19] *** j_s has quit IRC