[00:00:23] *** danige_ has joined #postfix
[00:04:13] *** evilstrike has quit IRC
[00:09:45] *** GMFlash has quit IRC
[00:09:52] *** GMFlash has joined #postfix
[00:10:29] *** meandtheshell has quit IRC
[00:10:52] *** caravena_ has quit IRC
[00:10:56] *** is_null has joined #postfix
[00:14:40] *** chatran has joined #postfix
[00:18:46] *** danige has quit IRC
[00:24:03] *** rmayorga has quit IRC
[00:24:39] *** Ryushin has joined #postfix
[00:25:03] *** rmayorga has joined #postfix
[00:30:36] *** pickcoder has joined #postfix
[00:44:17] *** master_of_master has joined #postfix
[00:47:03] *** skopii has quit IRC
[00:47:08] *** skopii has joined #postfix
[00:47:09] *** master_o1_master has quit IRC
[00:48:11] *** hparker has quit IRC
[00:49:50] *** war has quit IRC
[00:51:30] *** HKhan has quit IRC
[00:52:24] *** HKhan has joined #postfix
[00:52:31] *** porkpie has joined #postfix
[00:52:38] <porkpie> guy's help
[00:52:49] <porkpie> May  3 23:51:08 server1 postfix/qmgr[4549]: 8BB462441AF: to=<martin.jones at diginetltd dot co.uk>, relay=none, delay=2183, delays=2182/0.3/0/0, dsn=4.4.1, status=deferred (delivery temporarily suspended: connect to 127.0.0.1[127.0.0.1]: Connection refused)
[00:53:10] *** hiper has quit IRC
[00:53:11] <porkpie> I dunno why it's happeing
[00:54:04] *** pirho has quit IRC
[00:54:07] *** hacim has joined #postfix
[00:54:40] *** Taube is now known as taube
[00:54:42] <hacim> i'm getting hundreds of these a day, and am trying to figure out how to debug it:
[00:54:46] <hacim> May  3 12:06:58 mx1 postfix/smtpd[13673]: warning: mysql query failed: MySQL server has gone away
[00:54:50] <xpoint> porkpie, temp disk space ?
[00:55:04] <hacim> so far the only useful things I've found has been this mysql page: http://dev.mysql.com/doc/refman/5.0/en/gone-away.html -- but it hasn't helped any
[00:55:48] <porkpie> xpoint:I have loads of space
[00:56:18] <hacim> approximately 300 times today alone
[00:56:21] <porkpie> why am I getting the connection refused
[00:59:24] *** [dmp] has quit IRC
[01:01:31] <higuita> porkpie: what you have (or should have running) in localhost, port 25?!
[01:01:33] <higuita> check the master.cf
[01:01:57] <higuita> check postconf -n |grep 127.0.0.1
[01:02:05] <higuita> or grep policy
[01:02:13] <is_null> hello everybody, what is the best way to restrict a postfix server to local users (sending/receivend/forwarding if alias) with several domains (dynamic, file or db driven) please ? from what i have understood, there are several ways to acheive that
[01:02:31] <higuita> you havent a service running that postfix is configured to use
[01:03:53] <higuita> is_null: try to comment the smtpd line in master.cf
[01:03:56] <higuita> it should work
[01:04:33] <higuita> you will not receive external emails, but emails sent via sendmail comand or dropped in the mail queue will be processed
[01:06:14] *** pickcoder has quit IRC
[01:06:30] <higuita> if it fails, setup the inet_interfaces = localhost (with the smtpd above not commented), this way only local process will be able to access the smtp port
[01:08:43] *** cilly has quit IRC
[01:12:18] <is_null> higuita: ah! of course, thank you
[01:14:53] *** the-erm has left #postfix
[01:16:03] <is_null> higuita: oh, what if i want postfix to work for authenticated users ? won't restricting inet_interfaces to localhost prevent it ?
[01:18:38] *** porkpie has quit IRC
[01:19:21] <higuita> yes... you only asked for local users, not both local and auth 8)
[01:20:07] <is_null> higuita: my bad ...
[01:22:15] *** hparker has joined #postfix
[01:22:34] <higuita> for auth, put the permit_sasl_authenticated and the mynetworks in the smtpd_*_restrictions, then all your other rules...
[01:22:49] <higuita> and of course, configure the sasl 8)
[01:23:38] <telmich> hmmm.strange, what did I use to debug postfix's postgresql connections?
[01:24:06] <telmich> it looks like postfix does not open a connection and debug_peer_list does not give enought information
[01:28:08] <is_null> higuita: it still accepts unauthentified mail sending with : smtpd_recipient_restrictions = permit_sasl_authenticated, permit_mynetworks, reject_unauth_destination ... is that normal ?
[01:32:32] <higuita> is_null: what have you in mynetworks?
[01:34:48] <telmich> hmm..it also looks like postfix does not even try to do a lookup on the virtual domain table
[01:37:07] <higuita> telmich: i have little experience with postfix with databases, but check the postconf  -n output and check the permissions and location of the files refered
[01:37:51] <telmich> higuita: good idea
[01:40:48] <telmich> hmm..should work, but does not work
[01:41:40] <is_null> higuita: mynetworks=192.168.0.0/24, 127.0.0.0/8 (sorry for the delay, i had to re-read some documentation to understand what i understood whan i had put this)
[01:43:12] <higuita> is_null: so you are telling that all machines in 192.168.0.0/24 and the localhost can send emails without auth, all other must auth
[01:43:41] <higuita> is this what you want? you arent testing from a ip in the 192.168.0.0/24 network, arent you 8)
[01:44:10] <is_null> higuita: no, from a 82.x.x.x, internet
[01:44:56] <is_null> i'll set it to localhost only
[01:48:38] <higuita> is_null: yes, try that, maybe the way your firewall works, it maps the external ip to a lan ip...
[01:48:57] <higuita> and try also to replace the reject_unauth_destination to reject
[01:49:14] *** nemo_work has joined #postfix
[01:49:32] <nemo_work> I'd like to take this opportunity to sing postfix's praises.  especially the header manipulation
[01:50:27] <is_null> it works now, thanks higuita, this should keep postfix from getting me blacklisted right ?
[01:51:14] <is_null> *not saying that postfix itself get the server blacklisted of course
[01:52:22] <higuita> yes, this way only auth users and localhost can send emails...
[01:53:07] <is_null> higuita: ok thank you (just being another newbie making sure)
[01:53:30] <higuita> if you were a open relay, probably the problem was the firewall mapping the internet IPs to the lan IPs, and with the mynetworks open to all lan... a open relay 8)
[01:54:05] <higuita> is_null: everyone was a newbie sometime, dont worry, with time people learn more and more
[01:59:24] <is_null> oh i don't worry, thank you for your help. my workstation and the server aren't supposed to be open relays, however the server is provided by my ISP, and for example the newsgroup authentification is made by IP ... maybe this has something to do ?
[01:59:47] *** magyar has joined #postfix
[02:00:06] <is_null> *not that the newsgroup would have something to do, but the ISP/server provider network architecture could
[02:00:56] *** HKhan has quit IRC
[02:02:34] *** th3073ch has quit IRC
[02:05:46] *** chatran has quit IRC
[02:18:26] *** olinux has joined #postfix
[02:19:25] *** HKhan has joined #postfix
[02:22:35] *** nemo_work is now known as nemo_home
[02:23:21] *** nemo_home has left #postfix
[02:43:00] *** Mazon is now known as mazon
[02:46:43] *** phl4kx has joined #postfix
[02:51:03] *** phl4kx has quit IRC
[02:55:16] *** chiwawa_42 has joined #postfix
[02:58:20] <chiwawa_42> i'm looking for some hints setting up a mailling server. It's intended to render mails (from template) and send it to up to 10k mail adresses. Most of these adresses are hosted by 5 MX systems (ISP). Afaik, some MTA are able to maintain a connexion to a distant MX server to deliver severall mail during a single session. First, is postfix able to do so ? If so, must I sort mails to send by theyr destination domain or may postfix
[02:58:20] <chiwawa_42> handle it ?
[02:59:59] <chiwawa_42> if not, what may I use to make the mailing list sending faster?
[03:03:52] <chiwawa_42> the mail rendering is done with python scripts with a python web framework's template system to generate html mails. One rendered, a mail is directly sent with the basic mail() function. I'm wondering if storing the rendered mails in a DB and then asking for threaded functions to pull em out, selecting form their RCPT TO field, and actually send em will make the process faster, or if I'll have to use a SMTP client to do the job.
[03:03:53] <chiwawa_42> ..
[03:04:35] *** olinux has quit IRC
[03:05:23] *** danige__ has joined #postfix
[03:05:25] *** danige__ is now known as danige
[03:06:04] *** Tachy_ has joined #postfix
[03:07:08] *** caravena has joined #postfix
[03:08:10] <higuita> chiwawa_42: yes, postfix is able to reuse existent connections to deliver more emails... postfix choose what emails may reuse the connections
[03:08:25] *** _snd has joined #postfix
[03:09:21] <_snd> quick question: is there any obvious reason i should not be able to use ${domain} as a macro to pass on the domain to procmail in master.cf when configured tobe a transport so i can use procmail on my virtual delivery?
[03:09:33] <chiwawa_42> higuita: so it can pick mails from its queue if they have to be delivered to the same host, right?
[03:09:36] <higuita> check the connections_cache and connections_reuse
[03:09:56] *** Mez has joined #postfix
[03:10:19] <higuita> you may also sent one message with many emails in bcc (NOT to: nor CC!!)
[03:10:40] <higuita> this one, one email to a ISP will be delivered to many users
[03:11:04] <higuita> i use 50 emails for message, usually the limit is about 100 emails per message
[03:11:11] <chiwawa_42> mails are customized on a per user basis
[03:11:29] <chiwawa_42> so i'd have to send it once per target mail adress
[03:11:30] *** Tachy has quit IRC
[03:12:42] <_snd> no takers on virtual transport and procmail? :)
[03:14:02] <chiwawa_42> higuita: google doesn't know about connections_cache and connections_reuse :/ any URL?
[03:14:33] <higuita> chiwawa_42: yes,  postfix will pick the emails that are delivered to a server and reuse a connections if that already exist
[03:14:56] <higuita> _snd: sorry, no 8)
[03:15:04] <chiwawa_42> what is a safe lenght for postfix queue?
[03:15:21] <higuita> _snd:  IIRC, procmail couldt be used to deliver to virtual domains, but not really sure
[03:15:35] *** etaylor has joined #postfix
[03:15:48] <chiwawa_42> i mean, it probably won't like to have 10k mails in its queue, mostly because most of the servers I have to deal with uses greylisting
[03:15:52] <_snd> higuita: just looks wierd, because the docs say that should be available and i usually trust postfix doc :)
[03:15:58] <higuita> chiwawa_42: how much ram, HD you have ?  :)
[03:16:08] <_snd> higuita: googling around there are many people saying procmail and virtual is hard, but not impossible
[03:16:14] <chiwawa_42> 1Gig ram, 160G HD, 100Mb bandwidth
[03:16:33] <_snd> higuita: the other approach was to pass it ${recipient] and do some procmail-foo on it to pick user/domain from it
[03:16:34] <higuita> the queue can be HUGE, it will eat lot of HD space and mailq will take ages to run, but postfix will deliver then all
[03:17:05] <chiwawa_42> is HUGE like 10k or like 10M ?
[03:17:06] <chiwawa_42> :D
[03:17:08] <higuita> _snd: as i said, i dont know for sure, so i'm probably wront 8)
[03:17:19] <_snd> higuita: thanks :)
[03:18:14] <higuita> chiwawa_42: i already had about 200K in my newsletter server (a P700, with 512Mb ram, and my worst problem, a 10Gb hd )
[03:18:26] <chiwawa_42> higuita: anyway thanks a lot, using postfix will probably save me lot of time
[03:18:37] <chiwawa_42> ouch
[03:19:04] <higuita> i had in the past run out of space on that server, but then i put many emails on each message for the newsletters that allow it, and it saved me lot of HD space
[03:20:26] *** Zeit|idle has joined #postfix
[03:20:29] <chiwawa_42> so what if I want to save on performance ? Is there a way to let my python scripts that the queue is full enough to not be feeded to a point of inneficiency?
[03:20:31] <higuita> chiwawa_42: i fine tune my server, some ISPs only allow 1-2 connections, other have many MX and allow 20 connections, so i created several smtp like lines in master.cf
[03:21:05] <higuita> to each main ISP, like hotmail, google, my country biggest ISPs, etc
[03:21:41] <chiwawa_42> i'd like to get your parameters for those big international providers :)
[03:22:32] <higuita> yes, you can put a sleep on each loop, so that the delivery to the postfix is slower, allowing the each incoming message to be sent before the next one arrives
[03:22:57] *** danige_ has quit IRC
[03:23:06] <chiwawa_42> ok but that would need to fine tune it with an heavy benchmarking course
[03:23:23] <chiwawa_42> i'm rather thinking of letting the scripts know the queue size
[03:23:34] *** _snd has left #postfix
[03:24:21] <chiwawa_42> and then "while queue_size > 1k mails: sleep 10" in each thread
[03:24:31] <higuita> chiwawa_42: actually isnt nothing special, the postfix defaults are good
[03:25:03] <chiwawa_42> as always, some will probably say...
[03:25:31] <higuita> you must see if a remote server is taking too long to reply or if it deliver the message fast after the connections
[03:26:13] <chiwawa_42> does mailing lists managers take care of these performance tuning?
[03:26:37] <higuita> waiting for connections indicates that the remote server doesnt allow many connections, so its better to lower the default for that server so the server dont have to try and test what the limit is
[03:27:07] <chiwawa_42> that will lead to a long "try&error" process
[03:27:21] <higuita> the script to see the size of the mailqueue will eat IO, that will be important for a newsletter delivery, test it the less you can 8)
[03:27:34] <chiwawa_42> :)
[03:27:49] <chiwawa_42> can't postfix provide this information ?
[03:27:58] <higuita> not really (mailing lists managers)
[03:28:38] <higuita> yes and no... see the logs to see the speed of delivery to a domain
[03:28:59] <higuita> and see the output of mailq, of the active emails
[03:29:19] <chiwawa_42> ok
[03:30:12] <higuita> but a report saying the status... i dont know any, but maybe it exists (never really care about that fine tunning, i have about 5 fine tuned servers only,  and all of then to lower the postfix defaults
[03:30:48] *** Ryushin has quit IRC
[03:33:11] <chiwawa_42> well, performance may not be areal issue after all. The fat is, I'm having to fix an undocumented mass mailing system that broke up last week. It has to be back online next week. As there's no doc about what has been done yet, and as the performance were bad (less than 20k mails delivered per hour) where my python prototype gets up to 200k / hour on a default postfix configuration, I was looking for any hint on what has to be do
[03:33:12] <chiwawa_42> ne...
[03:34:16] <higuita> dont forget about the ISPs... dont abuse then, they have the power to blacklist your IP...
[03:34:18] <chiwawa_42> if my understanding is right, the "fine tuning" done on this mailing system was conter productive...
[03:34:55] <higuita> if you try to deliver too many emails to then, you will increase the load of their servers, and even worst, the storage
[03:35:33] <higuita> then receiving emails, and user reading the new emails will take a performance hit, and the ISP may choose to block you
[03:35:51] <higuita> if they are nice, they will contact you or limit your rate
[03:35:53] <chiwawa_42> well, it's far from spamming ! afaik my users have only opted-in subscription files of 3 to 20k adresses
[03:37:21] <higuita> i'm not saying its spamming, but one small ISP with a smtp server receiving 200K/hour emails from you, will probably notice that someone is "abusing" their server, that the load is strangly high
[03:37:49] <higuita> always think as the sender and as the recipient...
[03:38:11] <higuita> think that you are sending the newsletter to your system
[03:38:43] <higuita> dont forget that the servers that are receiving will scan for virus, classify for spam, storage the message
[03:38:52] <chiwawa_42> you're right, but I don't think it may happen. I'm far more concerned about misconfigured mail "security" gateways
[03:39:17] <chiwawa_42> as some mailing lists are beeing sent to corporate mail systems...
[03:39:40] <higuita> so things like sending 1000 message in 100 connections with each message with 500 emails will create a high load on the other side that they would for sure not like
[03:40:32] <chiwawa_42> got it
[03:40:51] <chiwawa_42> i will care of that :)
[03:41:02] <higuita> speed is good, being nice is also good 8)
[03:41:14] <chiwawa_42> higuita: as you seem quite knowledgeable, may I bother you with some other topics?
[03:41:22] <chiwawa_42> like archiveopteryx
[03:41:59] <higuita> most of the time, will not matter if the newsletters arrives now or in about 5 hours, so take care about the presure to increase the delivery rate
[03:42:23] <chiwawa_42> it's a mail storage system written in C++ using a PostgreSQL database for mail storage wich can be feeded via LMTP
[03:42:31] <higuita> and sending at late hours during the night is also good 8)
[03:43:49] <chiwawa_42> it stores the mails in an exploded form : one entry per mime attachment to avoid redudancy in attachement storage
[03:44:01] <higuita> i dont know about archiveopteryx, first time i know about it (i will test is someday, thanks 8)
[03:44:13] <chiwawa_42> ok :)
[03:44:22] <chiwawa_42> i'm looking for some feedback about it
[03:44:44] *** dj-fu has quit IRC
[03:44:59] *** dj-fu has joined #postfix
[03:45:03] <chiwawa_42> i've set it up on my personnal mail system (3 mailboxes, 500 ham / day max) but I'm willing to know more about it before using it on a production server
[03:45:24] <higuita> but this is for "normal" everyday mail server, of more for resending from the normal server to this one to archive older emails?
[03:46:21] <chiwawa_42> I think it has been designed for archival, but I can feed it with up to 4k mails / min via LMTP, wich looks sufficient for most of my customers
[03:46:53] <chiwawa_42> (tested on a 1,5GHz machine with 512Mb ram and 2 ATA drives in RAID1)
[03:47:55] <chiwawa_42> it's IMAP implementation seems good and implements a server side virtual folders and search functions extension to the protocol
[03:48:29] <chiwawa_42> they are planing to release a webmail frontend, didn't test the POP connector yet
[03:49:47] <higuita> yep... must test this :)
[03:49:52] <chiwawa_42> as an archive system, i've feed one with a 4 years history from the company I work for, it saved 70% of disk space by redundency checks
[03:49:55] <higuita> its usefull for work :)
[03:51:49] <chiwawa_42> well, i'll probably hang around in the next weeks, so please let me know how you tested it ;)
[03:53:23] <chiwawa_42> one of my developper is currently porting a nice ajax webmail to connect to archiveopteryx DB without using IMAP, would be nice to see how it works on your heavilly loaded mail systems ;)
[03:54:58] <higuita> i must go to sleep now
[03:55:03] <higuita> cya, good luck
[03:55:03] <chiwawa_42> so do I
[03:55:06] <chiwawa_42> 4 AM :(
[03:55:15] <chiwawa_42> cya ;)
[03:55:16] <higuita> here is 3H...
[03:55:21] <higuita> portugal here
[03:55:28] <chiwawa_42> france
[03:55:32] <higuita> 8)
[03:55:39] <higuita> close enough 8)
[03:55:50] <chiwawa_42> yup :)
[03:55:57] <higuita> cya then, sleep well
[03:56:03] <chiwawa_42> same
[03:58:52] *** rmayorga has quit IRC
[04:00:03] *** prebur has quit IRC
[04:09:05] *** dj-fu has quit IRC
[04:19:16] *** doomas has joined #postfix
[04:23:01] *** dj-fu has joined #postfix
[04:29:54] *** danige__ has joined #postfix
[04:33:01] *** doomas_ has quit IRC
[04:44:24] *** amrit|wrk is now known as amrit|afk
[04:48:21] *** danige has quit IRC
[04:54:43] *** cobra427 has joined #postfix
[05:00:09] *** joeyea323 has joined #postfix
[05:04:22] <joeyea323> hey
[05:05:55] *** cobra427 has quit IRC
[05:08:37] <joeyea323> can someone help me with setting up email forwarding rules in postfix?
[05:08:42] *** Zand3r has joined #postfix
[05:11:02] <Zand3r> Hi all... I am looking for examples of MySQL lookups for "check_client_access" and "check_recipient_access" for use in "smtpd_recipient_restrictions", any suggestions?
[05:15:03] *** rmayorga has joined #postfix
[05:21:45] *** rmayorga has quit IRC
[05:22:16] *** rmayorga has joined #postfix
[05:31:29] *** etaylor has quit IRC
[05:39:06] *** prebur has joined #postfix
[05:44:34] *** TheOutlander has joined #postfix
[05:48:12] *** joeyea323 has quit IRC
[05:51:34] *** Zand3r has quit IRC
[06:01:00] *** Zand3r has joined #postfix
[06:01:32] *** Tino is now known as Tinozaure
[06:02:12] *** caravena has quit IRC
[06:03:52] <Zand3r> I have the line "virtual_transport = dovecot" in main.cf - If I want to override this for particular domains and/or recipients then should I be using "transport_maps=" or "virtual_transport_maps=" - I have seen examples using both although only the former seems to be mentioned in the official documentation.
[06:10:09] <rob0> !virtual_transport
[06:10:10] <knoba> rob0: 'virtual_transport' : a configuration parameter in the main.cf: The default mail delivery transport for domains that match the $virtual_mailbox_domains parameter value. This information can be overruled with the transport(5) table.
[06:11:03] <rob0> postconf: warning: virtual_transport_maps: unknown parameter
[06:11:19] *** roe has quit IRC
[06:17:13] *** Motoko-chan has joined #postfix
[06:21:27] *** prebur has quit IRC
[06:23:21] *** lkthomas has quit IRC
[06:23:45] *** lkthomas_ has joined #postfix
[06:24:04] *** memic has quit IRC
[06:24:29] *** memic has joined #postfix
[06:28:26] *** TheOutlander has quit IRC
[06:33:25] *** rmayorga has quit IRC
[06:37:49] *** magyar has quit IRC
[06:41:00] *** prebur has joined #postfix
[06:48:11] <Zand3r> rob0: Thanks for the guidance. I have seen several examples online using  virtual_transport_maps  which is what was confusing me. Initial testing suggests that transport_maps is working fine in conjunction with my setup.
[06:49:19] *** Zand3r has quit IRC
[07:07:06] *** kjdavies has joined #postfix
[07:07:11] <kjdavies> how can I tell why postfix isn't starting?
[07:07:22] <kjdavies> I see nothing in /var/log that tells me anything useful
[07:07:37] <kjdavies> "/etc/init.d/postfix start" just tells me "[!!]"
[07:07:43] <kjdavies> not the most useful error message in the world
[07:09:40] *** Fr0zen_ has joined #postfix
[07:09:53] <Fr0zen_> whats the default postfix message size limit?
[07:09:56] <Fr0zen_> 9.7mb?
[07:13:18] <f3ew> 10MB
[07:13:26] <f3ew> postconf -d message_size_limit
[07:13:39] <f3ew> kjdavies, what *is* in your logs
[07:13:48] <f3ew> what happens if you type /usr/sbin/postfix start
[07:13:49] <kjdavies> which one?
[07:13:54] <f3ew> the maillog
[07:14:30] <kjdavies> May  3 21:59:25 cthulhu authdaemond: modules="authpam authshadow", daemons=5
[07:14:31] <kjdavies> May  3 21:59:27 cthulhu authdaemond: Installing libauthpam
[07:14:31] <kjdavies> May  3 21:59:27 cthulhu authdaemond: Installation complete: authpam
[07:14:31] <kjdavies> May  3 21:59:27 cthulhu authdaemond: Installing libauthshadow
[07:14:31] <kjdavies> May  3 21:59:27 cthulhu authdaemond: Installation complete: authshadow
[07:14:34] <kjdavies> that's it
[07:14:40] <kjdavies> (I cycled the logs earlier today)
[07:14:41] <f3ew> tha't courier
[07:14:54] <kjdavies> nothing from postfix, then
[07:15:08] <f3ew> what happens if you type /usr/sbin/postfix start <==
[07:15:24] <kjdavies> ah.
[07:15:35] <kjdavies> 'libpq.so.4: cannot shared object file'
[07:15:41] <kjdavies> postgresql updated
[07:15:46] <kjdavies> so.  that's that one
[07:15:56] <kjdavies> starting revdep-rebuild
[07:16:02] <f3ew> heh
[07:16:22] <kjdavies> ... and same for dspam
[07:16:25] <kjdavies> so.
[07:16:30] <kjdavies> thanks. I didn't know that would tell me
[07:18:21] <kjdavies> yep.  dspam's borked too
[07:18:28] <kjdavies> I wish I'd realized earlier
[07:26:12] <kjdavies> at least I remember to do etc-update now
[07:26:17] <kjdavies> *that* was an unpleasant surprise
[07:35:10] <ayeuu> hi there,can someone helpme please about quota on postfix ? here is my prob, I would like to use quota for specific virtual users, but I wont patch postfix. So maildrop seems to be the solution,but I won't use mysql neither, but I don't find any good howto about it.
[07:36:07] <ayeuu> does someone can tell me how to do it,or have a good url for ?
[07:37:04] *** dj-fu has quit IRC
[07:37:10] <ayeuu> (sorry for my english, I'm a bit tired..)
[07:37:50] *** kjdavies has left #postfix
[07:49:19] <Fr0zen_> lol
[08:00:14] <Yaroon> What's the best way to see whether postfix is running?
[08:00:20] <Yaroon> There's no postfix process or anything.
[08:01:04] <Signum> Yaroon: you should find a "master" process running
[08:01:13] <Signum> Yaroon: or try "telnet localhost 25"
[08:01:24] <Yaroon> Ahah, there we go.
[08:01:28] <Yaroon> Yes thank you :)
[08:03:53] <ayeuu> netstat too
[08:11:57] *** hparker has quit IRC
[08:49:39] *** frennkie has joined #postfix
[08:50:00] *** RT_852 is now known as Faq
[08:56:11] *** Faq is now known as RT_852
[09:03:00] <Fr0zen_> virtual_mailbox_limit
[09:03:12] <Fr0zen_> is that the maximum amount of space a users mailbox can hold?
[09:03:34] <Fr0zen_> i have a few users with no quota, does that mean they cant exceed 20mb of email? (the default)
[09:03:35] *** UQlev has joined #postfix
[09:04:32] <Fr0zen_> in my virtual users db, I have quota 0
[09:04:38] <Fr0zen_> so im guessing that limit won't apply to them.
[09:06:38] *** meandtheshell has joined #postfix
[09:08:25] *** Motoko-chan has quit IRC
[09:09:38] *** [miles] has joined #postfix
[09:17:37] *** danige__ is now known as danige
[09:21:35] *** mastachand has joined #postfix
[09:25:39] *** cutmasta has joined #postfix
[09:27:17] *** pmjdebruijn has joined #postfix
[09:27:50] *** rdallarmi has joined #postfix
[09:28:29] <rdallarmi> hi, is there anybody in here with experience of sasl auth using authdeamond?
[09:28:38] *** redondos_ has joined #postfix
[09:34:48] *** mazon is now known as Mazon
[09:36:56] *** amrit|afk is now known as amrit|zzz
[09:42:57] *** Tinozaure is now known as Tino
[09:43:46] *** redondos has quit IRC
[09:43:47] *** redondos_ is now known as redondos
[09:48:58] <danige> so, I have done this check with relaytest (#telnet relay-test.mail-abuse.org)
[09:49:38] *** eltech has quit IRC
[09:50:03] <danige> and there are four numbered tests, which are checking for open relay, the first is called Quote Test and four are Numbered, is the first one a test as well?
[09:53:00] <R1ck> danige: yes
[09:53:12] <danige> how do I analyse the result?
[09:54:06] <R1ck> if you only get 'Recipient address rejected: Access denied', its OK
[09:54:13] <danige> the system appeared to accept 1 relay attempt, but both teh Quote and Test #4 seem to have been accepted
[09:54:41] <R1ck> and the last message should be 'System appeared to reject relay attempts
[09:54:42] <R1ck> '
[09:54:43] <danige> I would like to understand what test 4 does and what the quote test means, and how to fix the hole
[09:55:12] <danige> I have defined the sender_access and recipient_access
[09:55:19] <danige> seem to be okay.
[09:55:41] <R1ck> danige: looks like your server accepts mail to any address as long as the client identifies itself as someone@localhost in the MAIL FROM
[09:55:55] <R1ck> >>> mail from: <spamtest@localhost>
[09:55:56] <R1ck> <<< 250 2.1.0 Ok
[09:55:56] <R1ck> >>> rcpt to: <nobody at mail-abuse dot org>
[09:55:56] <R1ck> <<< 504 5.5.2 <spamtest@localhost>: Sender address rejected: need fully-qualified address
[09:55:59] <R1ck> >>> rset
[09:56:02] <R1ck> that is how it should look
[09:56:26] <danige> yes, I got that but this is something that should not be on my server, where would the hole be?
[09:56:48] *** bostik has joined #postfix
[09:57:18] <R1ck> danige: whats your main.cf?
[09:57:29] <R1ck> (not postconf)
[09:57:59] <sysmonk> R1ck: why not postconf -n ?
[09:58:40] <R1ck> right, thats ok too
[09:59:18] *** taube is now known as Taube
[09:59:33] <R1ck> probably better even, without all the comments
[10:01:02] *** mastachand has quit IRC
[10:01:15] <danige> I have no comments on this one: http://mentalitea.com/?page=main.cf
[10:01:25] <rdallarmi> hi guys I should need a hand with sasl authorisation using authdeamond. I am a newbie I had it working but somehow I screw it up. authdeamond is working for imap and pop3 access but it is not working for authorizing mail sending from the server. it shoud not take more than 5 minutes for you but it is has been a showstopper for me for a few days now
[10:01:56] *** war has joined #postfix
[10:02:12] <danige> I just saw a thing called relayhost with an empty value... is that open
[10:02:31] <danige> ?
[10:03:59] <R1ck> danige: no, its also empty per default
[10:04:51] <danige> It is quite bad to have this kind of server running but I have to receive my mail, too... I have tried to filter the spam out, but so far no luck...
[10:08:32] *** mastachand has joined #postfix
[10:08:54] <R1ck> danige: can you paste the output of the telnet ?
[10:14:53] *** is_null_ has joined #postfix
[10:17:59] *** is_null has quit IRC
[10:18:13] *** WALoeIII has joined #postfix
[10:19:07] <danige> http://mentalitea.com/?page=telnet
[10:23:08] *** WALoeIII has quit IRC
[10:24:42] <danige> so, although it seemed to have accepted the Quote test and test #4 the result is only 1 attempt accepted...
[10:28:31] <R1ck> no, it stopped after the 4th test
[10:28:34] <R1ck> there are 20 total
[10:31:56] <R1ck> danige: try removing 'localhost' and 'localhost.localdomain' from mydestination
[10:34:36] <danige> ok, seems to be the same problem
[10:34:48] <rob0> rdallarmi: simply change the permissions on the authdaemond socket (and any directories leading thereto) such that postfix (the user account or group) has read privilege (and execute on directories.)
[10:35:44] <rob0> If you have a working authdaemond, that plus an appropriate smtpd.conf file is all it takes
[10:37:46] <danige> What I want is that I can receive mail from any client, but have it undergo rigorous SPAM testing and then only accept it for addresses on my server listed in recipient_access
[10:37:49] <rob0> Google "smtpd.conf authdaemond" for samples. I think it's mainly the pwcheck_method line you need.
[10:38:20] <rdallarmi> rob0 I did that already
[10:38:33] <danige> next I want to be able to send mail to every recipient from my own server.
[10:39:01] <danige> what happens, is that someone can fool my little server and say he is me...
[10:41:27] <danige> bastards!
[10:43:42] <danige> I hate CIALIS softtabs!
[10:45:42] <rob0> What's in those sender_ and recipient_access files?
[10:45:47] <rob0> !cheatsheet
[10:45:48] <knoba> rob0: 'cheatsheet' : http://jimsun.linxnet.com/misc/postfix-anti-UCE.txt : A HOWTO for pre-DATA spam control.
[10:45:52] <rdallarmi> rob0: could you have a look at this? I think I went trough all the steps by myself before asking for help. : http://www.pastebin.ca/470437
[10:46:08] <rob0> ^^ good starter HOWTO for smtpd_recipient_restrictions
[10:46:16] <rdallarmi> rob0 there you'll find main.cfm ,aster.cf and smtpd.conf
[10:47:18] <danige> rob0: http://pastebin.ca/469263
[10:47:36] <danige> http://pastebin.ca/469259
[10:48:19] <rob0> rdallarmi: So? What are permissions on /var/run/courier/authdaemon/socket and /var/run/courier/authdaemon and perhaps /var/run/courier ? No logs in that post. And "postconf -n" is preferred over main.cf contents.
[10:49:26] <rdallarmi> rob0: as I said permission are set as drwxrwxrwx for the whole path
[10:49:53] <rob0> and the socket? and the logs say?
[10:49:56] <rdallarmi> rob0 and in relation to postconf -n you should tell me which setting you want me to query
[10:51:13] <rob0> danige: those are definitely not safe as check_sender_access. As you seem to have discovered, any spammer can pretend to be you and you say "OK".
[10:52:19] <rdallarmi> rob0 the log says arning: sysdevel[192.168.2.3]: SASL PLAIN authentication failed: authentication failure but I can veryfy correctly with authtest the same username/password
[10:52:22] <danige> okay, but how can I change that, do I have to explicitly say: myname at mydomain dot tld OK or is there a different way?
[10:52:25] <rob0> danige: What specifically happened is that the test used an unqualified sender address and your Postfix appended $myorigin.
[10:52:59] <rob0> I simply would not use a check_sender_access lookup like that. What is the goal here?
[10:53:28] <rob0> You must not understand how SMTP works. Anyone can forge a sender address ... ANY sender address.
[10:53:42] <danige> so, how would I do that? The goal is, that only people from my server can send mail away from it.
[10:54:11] <danige> but I must understand how smtp works...
[10:54:34] <danige> I don`t
[10:54:39] <danige> yet...
[10:55:19] <rob0> !basic
[10:55:20] <knoba> rob0: 'basic' : http://www.postfix.org/BASIC_CONFIGURATION_README.html : a good starting place for Postfix beginners, many common questions are answered here.
[10:55:22] <rdallarmi> rob0: here you have the full log, with few  additional information  http://www.pastebin.ca/470453
[10:55:50] <rob0> Usually relaying is safely controlled by $mynetworks and the default smtpd_recipient_restrictions setting.
[10:58:12] <rob0> rdallarmi: looks like your correct smtpd.conf is not the one being used. It wants to auth against /etc/sasldb2 ? Find the other smtpd.conf (a Cyrus SASL compile-time issue I bet) and fix it?
[10:59:02] <f3ew> Hmmm, it's Friday
[11:05:59] *** UQlev has quit IRC
[11:06:42] *** milestone has joined #postfix
[11:06:46] <milestone> hi all
[11:06:56] <milestone> i am using postfix as our backup mx mail relay
[11:07:03] <milestone> works like a charm
[11:07:15] <danige> but?
[11:07:32] <milestone> now one of our customers asked me a weird question. He wants to do some maintenance on his mail server
[11:07:50] <milestone> our mail server would collect the mail and deliver once it is back up
[11:08:15] <milestone> but he wants one email adress forwarded to a different address while his maintenance
[11:08:18] <milestone> is this possible?
[11:08:33] <milestone> i am doing mx lookups to determine if i accept mail or not
[11:08:43] <milestone> so the domain is not really locally configured
[11:09:17] <[miles]> morning guys
[11:10:21] <[miles]> I want to pass every mail coming into the server via a script. All mail goes to one mbox, would i be ok doing it in "local"
[11:11:03] <[miles]> or better to do it smtpd
[11:13:28] <rdallarmi> rob0 in auth.long I find: _sasl_plugin_load failed on sasl_auxprop_plug_init for plugin: sql can it have anything to do? I am using a postgres database for storing passwords althought I am using authdeamond
[11:13:55] <rdallarmi> AND ALSO May  4 11:10:09 server postfix/smtpd[6042]: sql_select option missinG
[11:14:26] <milestone> is my setup possible? Or is my customer a complete weirdo
[11:22:49] *** birmaan has joined #postfix
[11:22:54] *** lkthomas_ has quit IRC
[11:23:51] <danige> rob0: so my smtp_recipient_restrictions are not restrictive enough?
[11:24:05] <danige> smtpd_recipient_restrictions = permit_mynetworks, permit_sasl_authenticated, check_sender_access hash:/etc/postfix/sender_access, check_recipient_access hash:/etc/postfix/recipient_access, reject
[11:26:15] *** nescius has joined #postfix
[11:27:40] <danige> besides, I have my domain names in two places and would like to simplify, I have them once in the sql database and then in the files sender_access and recipient_access. I would like to throw out the files and do it only with database... can I do that? Where do I find documentation for that?
[11:28:15] <rob0> rdallarmi: again, YOU think you're using authdaemond, but Cyrus SASL disagrees. Find and fix the smtpd.conf it's using.
[11:29:04] <rob0> danige: again, a check_sender_access lookup which returns OK in smtp_recipient_restrictions is not safe. NEVER safe.
[11:29:10] <rdallarmi> I made a search, there is no other smtpd.conf in the system
[11:29:45] <rob0> hmmm, Cyrus SASL can be a mysterious beast.
[11:30:00] <danige> so I have to throw out the check_sender_access from the restrictions altogether?
[11:31:01] <rob0> danige, absolutely. I'd start over from !basic if I were you.
[11:32:14] <rob0> f3ew: TGIF (F stands for ... what? "f3ew"?)
[11:32:47] <danige> I threw it out and all 20 Tests where rejected. If this wasn`t a stupid computer who does what you tell it, I would say you are a wizard...
[11:32:52] *** _ac3_ has joined #postfix
[11:33:01] <danige> now I just have to try if I can still accept mail...
[11:33:02] <_ac3_> hi
[11:34:09] 
[11:35:13] *** yess has joined #postfix
[11:35:17] <yess> Hello.
[11:38:16] *** UQlev has joined #postfix
[11:38:30] <danige> hi yess
[11:38:43] <Signum> _ac3_: http://en.wikipedia.org/wiki/Explicit_Congestion_Notification perhaps?
[11:40:19] <yess> Ha. Channel topic helped me solve my problem before I asked the question. Cool.
[11:41:03] * rob0 hands a medal to yess. Yay!!
[11:41:21] <rob0> Topic Reader First Class
[11:41:26] <_ac3_> found it: disabling /proc/sys/net/ipv4/tcp_window_scaling solved the problem
[11:41:37] <[miles]> mmm ok, how can I make an incoming message be sent to it's normal transport, and also an additional one please?
[11:42:17] <Signum> yess: if the topic could store longer strings we might as well dump google :)
[11:42:39] <rob0> [miles] might want always_bcc or *_bcc_maps, or maybe consider using virtual_alias_maps too.
[11:43:15] <[miles]> mmm
[11:43:26] <[miles]> no bcc stuff
[11:43:48] <[miles]> sorry rob0, I don't think either of those methods are the way to go
[11:46:50] <f3ew> Fridat
[11:48:32] <[miles]> f3ew: hi
[11:49:04] <[miles]> f3ew: do you have any idea how I can get postfix to always pass a message to an extra transport please?
[11:57:19] *** bisoc_ has quit IRC
[11:57:22] <milestone> rob0: how do you mean that?
[11:57:41] <milestone> so it is possible to create a virtual bcc map`?
[11:58:25] <milestone> rob0: never mind you did not mean me ;)
[12:00:12] <rob0> 1.6kmstone :)
[12:00:13] *** war_ has joined #postfix
[12:01:26] <rob0> milestone: actually, upon scrolling up, I think you might want to use a virtual(5) alias there.
[12:03:24] <milestone> rob0 i will read into it. But again, the domain is not locally configured, all postfix is doing, is checking if it is a valid mx for a given domain, and queues the mail then and delivers it to the primary mail server. So event without being a locally configured domain, this would be possible to route only certain e-mail adresses differently? That would be a real cool feature...
[12:07:37] *** pmjdebruijn has quit IRC
[12:08:23] *** nfi|ermes has joined #postfix
[12:09:48] *** GMFlash has quit IRC
[12:09:54] *** GMFlash has joined #postfix
[12:11:43] <milestone> rob0: from the man page i read:
[12:12:02] <milestone> With a virtual alias domain, the Postfix SMTP server accepts  mail  for  known-user at virtual-alias dot domain, and rejects mail for unknown-user at vir-tual-alias dot domain as undeliverable.
[12:12:12] <milestone> which is not really what i want
[12:13:20] <milestone> i want to queue mail for domain.com while the primary mail server is unavailable and route only one address user.a at domain dot com to usera at anotherdomain dot com
[12:14:57] <rob0> nono not virtual_alias_domains, just a virtual_alias_maps listing for the affected user.
[12:30:31] <milestone> rob0: is it enough to just put virtual_alias_maps ... into main.cf, or do i have to reference to cinfig entry somewhere
[12:31:23] <ayeuu> hi/re : " postfix/sendmail[22546]: fatal: no debugger_command variable set up" does somone know why i've got that please ?
[12:34:26] <ayeuu> ok, i've found, thx
[12:34:30] <rob0> virtual_alias_maps has to point to some kind of lookup, like a text file you would postmap(1), or a mysql query.
[12:34:41] <rob0> !virtual
[12:34:42] <knoba> rob0: 'virtual' : a way to configure additional domains and user accounts (that do not need to exist in your /etc/passwd). See: http://www.postfix.org/VIRTUAL_README.html
[12:35:04] <rob0> The format of the lookup is described in virtual.5.html .
[12:36:27] *** _ac3_ has quit IRC
[12:36:44] <milestone> rob0: of course i have already done so
[12:36:51] *** _ac3_ has joined #postfix
[12:36:54] <milestone> hash:/etc/postfix/virtual
[12:37:02] <milestone> or to be more complete
[12:37:15] <milestone> virtual_alias_maps = hash:/etc/postfix/virtual
[12:37:48] *** eltech has joined #postfix
[12:38:34] <milestone> all i am asking is is this enough? Or do I have to reference virtual_alias_maps within some other configuration section?
[12:40:04] <ayeuu> hum,still this error :/
[12:42:37] <rob0> That is enough.
[12:44:32] <danige> hooray, everything works...
[12:44:55] *** danige has left #postfix
[13:05:55] <[miles]> mmm...
[13:06:00] <[miles]> this is becoming a pain in the a**
[13:09:52] *** noetik has joined #postfix
[13:13:09] *** knoba has quit IRC
[13:13:18] *** knoba has joined #postfix
[13:16:49] *** UQlev has quit IRC
[13:17:02] *** hooch has quit IRC
[13:17:08] *** hooch has joined #postfix
[13:20:04] <[miles]> can anyone recommend the best way to get postfix to pass a mail to an additonal transport... when recieving mail... I need all mail to be passed to pyzor report
[13:21:45] *** bisoc has joined #postfix
[13:24:00] *** HKhan has quit IRC
[13:25:45] *** eltech has quit IRC
[13:28:08] *** dj-fu has joined #postfix
[13:36:12] *** rdallarmi has quit IRC
[13:37:33] *** HKhan has joined #postfix
[13:42:28] *** HKhan has quit IRC
[13:43:00] <telmich> if I try to send a mail to nico at test dot schottelius.org, i get relay access denied. viewing the logs it does not look like postfix tries to resolve the mailaddr via virtual table
[13:45:24] *** noetik has quit IRC
[14:03:00] *** HKhan has joined #postfix
[14:06:00] *** HKhan has quit IRC
[14:13:33] *** dj-fu has quit IRC
[14:14:34] *** HKhan has joined #postfix
[14:26:11] <Signum> telmich: what is your "virtual_mailbox_domains" set to? is the domain listed there?
[14:30:38] *** TheOutlander has joined #postfix
[14:33:36] *** noetik has joined #postfix
[14:33:49] *** swampfox has quit IRC
[14:41:04] *** sep has quit IRC
[14:43:54] <telmich> Signum: I just played around a bit, using a hash-table it works. now I am retrying with pgsql, but get the following error:
[14:44:06] <telmich> May  4 14:41:56 [postfix/trivial-rewrite] warning: connect to pgsql server localhost: SSL SYSCALL
[14:44:10] <telmich> error: No such file or directory?
[14:46:39] *** roe has joined #postfix
[14:46:52] *** Mavvie has quit IRC
[14:46:58] *** Mavvie has joined #postfix
[14:54:52] *** ashd has joined #postfix
[14:57:25] *** Ryushin has joined #postfix
[14:58:01] *** eltech has joined #postfix
[14:59:41] *** eltech has quit IRC
[15:07:35] <telmich> ssl must be off on debian/unstable+chroot
[15:16:34] *** ziro has joined #postfix
[15:21:05] *** vlt|home has joined #postfix
[15:21:31] <vlt|home> Hello. Is there an clean and easy way to tell postfix not to try to send mail via the defined relayhost but to collect emails sent to it? The relayhost is configured wrong and refuses all emails at the moment ...
[15:23:17] *** _ac3_ has quit IRC
[15:25:02] <vlt|home> For now I changed "relayhost" in postfix's main.cf to an invalid ip address ... but I think there must be a better way to set the mail queue "on hold". Any idea?
[15:25:18] <f3ew> vlt|home use the defer in transport_maps
[15:25:29] <f3ew> or just  /./ HOLD
[15:25:49] <f3ew> in an access map
[15:26:51] <vlt|home> f3ew: Where can I find an "access map"?
[15:29:34] <vlt|home> f3ew: I have neither a config file called transport_maps nor access* ... Can you help?
[15:29:50] <sysmonk> defer_transports = virtual
[15:29:51] <sysmonk> or local
[15:29:55] *** sep has joined #postfix
[15:30:04] *** milestone has quit IRC
[15:30:59] <sysmonk> like, postconf -e defer_transports=virtual
[15:31:44] <vlt|home> sysmonk: Thank you. I think I need a little more detailed help. Where to put "defer_transports=virtual"? main.cf? What does it exactly mean?
[15:31:51] *** Tenshi has joined #postfix
[15:32:27] <Tenshi> Could someone explain to me the difference between the "smtpd_*" directives and the "smtp_*" directives in the config file?
[15:34:32] <Tenshi> which one deals with sending mail directly to/through the smtp server?
[15:35:21] <R1ck> they both do
[15:35:24] *** Ryushin has quit IRC
[15:36:26] <Tenshi> not receiving local mail who's final destination is the server
[15:36:34] *** UQlev has joined #postfix
[15:37:26] *** sep is now known as sepski
[15:40:06] *** RiEgEl has joined #postfix
[15:42:19] <RiEgEl> i got a problem with my postfix +sasl2.. when I try to send an email to an adress which is not on my server WITHOUT authentification i got an error: relay access denied.. when i try WITH smtp-auth my mail-client asks me lots of times for the password. What's going wrong?
[15:43:49] <f3ew> !defer_transport
[15:43:50] <knoba> f3ew: Error: "defer_transport" is not a valid command.
[15:43:54] <f3ew> !transport_maps
[15:43:55] <knoba> f3ew: 'transport_maps' : a configuration parameter in the main.cf: Optional lookup tables with mappings from recipient address to (message delivery transport, next-hop destination). See transport(5) for details.
[15:44:01] <Tenshi> RiEgEl, join the club :(
[15:44:48] *** ^Timo^ has quit IRC
[15:47:49] <RiEgEl> hmm
[15:49:25] <Signum> !learn defer_transport as the names of message delivery transports that should not deliver mail unless someone issues "sendmail -q" or equivalent. Specify zero or more names of mail delivery transports names that appear in the first field of master.cf
[15:57:26] <f3ew> !access
[15:57:27] <knoba> f3ew: Error: "access" is not a valid command.
[15:57:43] *** vlt|home has quit IRC
[16:00:45] *** _RiEgEl has joined #postfix
[16:11:17] *** freeemind has joined #postfix
[16:11:47] <freeemind> hello mates, I'd setup postfix with dspam(lmtp) - after dspam chain finished, postfix said: authentication required - how to fix this?
[16:12:17] <freeemind> It seems that postfix isnt notice that dspam is localhost connection
[16:16:19] *** RiEgEl has quit IRC
[16:18:59] *** noetik has quit IRC
[16:23:30] *** noetik has joined #postfix
[16:24:51] *** higuita has quit IRC
[16:34:39] *** af_ has joined #postfix
[16:38:25] *** pickcoder has joined #postfix
[16:45:07] *** sparkleytone has quit IRC
[16:45:30] *** sparkleytone has joined #postfix
[16:49:13] *** GutterPunk has quit IRC
[16:53:36] *** cutmasta has quit IRC
[16:53:54] *** Mez has quit IRC
[16:59:26] *** Mez has joined #postfix
[17:03:05] *** tchmnkyz has quit IRC
[17:04:59] *** hparker has joined #postfix
[17:05:45] *** Lap_64 has joined #postfix
[17:06:17] *** xpoint has quit IRC
[17:08:15] *** _RiEgEl has quit IRC
[17:11:25] *** bisoc has quit IRC
[17:11:26] *** doomas has quit IRC
[17:11:27] *** master_of_master has quit IRC
[17:11:27] *** Fr0zen_ has quit IRC
[17:11:31] *** master_of_master has joined #postfix
[17:16:25] *** bisoc has joined #postfix
[17:16:34] *** doomas has joined #postfix
[17:18:09] <ayeuu> does a "virtual_mailbox_domains =  file.not.hash" for <1000 domains listed is fast to use please ? (not putting ram/cpu down)
[17:20:48] *** noetik has quit IRC
[17:25:01] *** knoba has quit IRC
[17:25:07] *** Mez_ has joined #postfix
[17:29:13] *** knoba has joined #postfix
[17:31:57] *** rmayorga has joined #postfix
[17:32:39] *** freeemind has left #postfix
[17:34:08] *** UQlev has quit IRC
[17:38:40] *** Lap_64 has quit IRC
[17:39:08] *** Mez has quit IRC
[17:44:46] *** Taube is now known as taube
[17:51:04] <pickcoder> how do I tell the virtual transport to write a maildir file with a specific permission setting?
[17:51:09] <pickcoder> I can't find anything in the docs
[17:51:24] *** Fr0zen_ has joined #postfix
[17:51:26] <Fr0zen_> May  4 10:55:38 mail postfix/local[30777]: fatal: main.cf configuration error: mailbox_size_limit is smaller than message_size_limit
[17:51:39] <Fr0zen_> why am i seeing that? The mailbox size limit is bigger... in main.cf
[17:51:41] <pickcoder> I need group rw, but it only will set user rw
[17:52:26] <pickcoder> Fr0zen_: postconf mailbox_size_limit
[17:52:48] <Fr0zen_> ah
[17:52:53] <Fr0zen_> mailbox_size_limit = 51200000
[17:52:56] <Fr0zen_> message_size_limit = 52428800
[17:53:00] <pickcoder> that's default
[17:53:11] <pickcoder> 1024 is default for message
[17:53:33] *** ashd has quit IRC
[17:53:34] <Fr0zen_> gotcha, so i need to change this in default.cnf or just add it to main.cf
[17:53:49] <pickcoder> whereever it is defined
[17:53:55] <pickcoder> or increase mailbox
[17:54:00] <pickcoder> in the main config
[17:56:15] *** Mc_Fly has quit IRC
[17:56:49] *** devdas has joined #postfix
[17:59:39] *** xai has joined #postfix
[18:00:16] *** backz has joined #postfix
[18:00:24] <xai> I'm rsyncing some Maildir folders and wanted to know if I can omit anything that isn't needed. Can I omit any data like the ~user/Maildir/tmp/* ?
[18:01:52] <Fr0zen_> it's defined in main.cf.default
[18:01:55] <Fr0zen_> not in main.cf
[18:04:06] *** mastachand has quit IRC
[18:04:42] <roe> Fr0zen_, you need to define it in main.cf
[18:06:07] <devdas> xai: that's mail being delivered
[18:07:51] <Fr0zen_> thanks roe
[18:10:42] *** [miles] has quit IRC
[18:11:09] *** taube is now known as Taube
[18:13:39] *** JoKoT3 has joined #postfix
[18:16:02] *** nescius has quit IRC
[18:18:07] *** higuita has joined #postfix
[18:20:08] *** bostik has quit IRC
[18:26:41] *** Fullmoon has joined #postfix
[18:28:30] *** stellina_ has quit IRC
[18:28:44] <xai> devdas: bummer..
[18:34:53] *** nfi|ermes has quit IRC
[18:37:03] *** hacim has left #postfix
[18:37:36] *** piedoggie has joined #postfix
[18:38:53] *** hiper has joined #postfix
[18:39:32] *** af_ has quit IRC
[18:40:53] *** JoKoT3 has quit IRC
[18:42:15] <magikman> hello. I am running postfix in a chroot environment and am having trouble with time stamps on emails received. the system time is set correctly which i ma guessing wont matter due to the chroot. so, what i have done is copied over the zoneinfo directory and linked localtime to the correct zone file.
[18:42:35] <piedoggie> had a mail loops back to self problem.  DNS points to the mail server but I'm trying to direct mail to a different mail server for one domain.  I put an entry in transport specifying the destination machine with square brackets around an IP address.  But still, postfix is finding the DNS entry and thinks its loopback to self.  I've encountered this problem before but I can't remember or find...
[18:42:37] <piedoggie> ... the information I need
[18:42:43] <magikman> the time stamps on emails is still incorrect.
[18:43:18] <piedoggie> chroot doesn't affect system time
[18:43:37] <piedoggie> is the difference between correct time and message time an exact number of hours?
[18:44:16] <magikman> yes. it is 5 off of EST
[18:45:26] <piedoggie> so the timestamp and the messages are UTC?
[18:45:35] <magikman> yep.
[18:45:50] <piedoggie> What is your time zone setting in chroot?
[18:46:13] <magikman> localtime -> /var/spool/postfix/usr/share/zoneinfo/EST
[18:46:29] <magikman> i see the problem!
[18:46:30] <magikman> haha
[18:46:41] <magikman> that shouldn't be using that path, right?
[18:46:44] <piedoggie> y
[18:46:47] <magikman> haha
[18:46:48] <magikman> duh
[18:46:56] <piedoggie> it should be a relative path within the context of your chrooted environment
[18:47:11] <magikman> yeah...
[18:47:26] <piedoggie>  I personally think chroot is more trouble than it's worth most of the time
[18:47:45] <piedoggie> virtual machines on the other hand are really quite nice
[18:48:24] *** xpoint has joined #postfix
[18:48:30] <magikman> i prefer using freebsd jails but unfortunately the company i use for hosting doesn't offer it.
[18:49:01] <Tenshi> my sasl authentication always fails, and that's all that I'm seeing in the logs.  How can I diagnose it further?
[18:52:28] *** tminos has quit IRC
[18:52:35] *** tminos has joined #postfix
[18:55:47] *** klauwhamer has quit IRC
[18:56:13] *** klauwhamer has joined #postfix
[19:01:13] *** hiper has quit IRC
[19:01:36] *** olinux has joined #postfix
[19:03:11] *** magikman has left #postfix
[19:14:05] *** gAri- has quit IRC
[19:14:39] *** gAri- has joined #postfix
[19:14:53] <Tenshi> I also don't get the "AUTH" or "AUTH=" advertisement when I telnet into my smtp server.  Any ideas?
[19:17:40] *** Fullmoon has quit IRC
[19:19:37] *** cruxeternus has left #postfix
[19:21:04] *** th3073ch has joined #postfix
[19:21:12] <th3073ch> Hi, I'm getting logfile msgs "mail for mail.XXX.org loops back to myself".  Mailman seems to be working correctly, but I believe that these are coming from bounces from mailman list recipients.  What do I need to modify to fix this?
[19:22:37] <sepski> th3073ch, add mail.XXX.org to your mydestination
[19:23:02] <sepski> ie:your mailserver should accept mail for them
[19:23:53] <piedoggie> I have a similar problem except I'm trying to route mail for the domain to another machine.  It keeps telling me to mail loops back self (which it does if you ignore the transport entry)
[19:53:41] *** Zaw has left #postfix
[20:05:31] *** hemry has joined #postfix
[20:12:59] *** devdas has quit IRC
[20:14:13] *** th3073ch has left #postfix
[20:15:26] *** Tenshi has quit IRC
[20:18:03] *** amrit|zzz is now known as amrit|wrk
[20:26:50] *** Leonidas has joined #postfix
[20:27:57] *** LinuxLoaf has joined #postfix
[20:43:33] *** Taube is now known as taube
[21:14:07] *** taube is now known as Taube
[21:17:20] *** ziro has quit IRC
[21:18:05] *** frennkie has quit IRC
[21:21:52] *** LinuxLoaf has quit IRC
[21:27:29] *** birmaan has quit IRC
[21:30:47] *** stony has joined #postfix
[21:31:39] *** nfi|ermes has joined #postfix
[21:42:19] *** nfi|ermes has quit IRC
[21:47:17] *** Mavvie has quit IRC
[21:48:51] *** knoppix has joined #postfix
[21:53:49] *** Zeit|awy has joined #postfix
[21:56:17] *** UQlev has joined #postfix
[21:57:00] *** knoppix has quit IRC
[21:57:28] *** stony has quit IRC
[21:59:35] *** Zeit|idle has quit IRC
[22:09:07] *** war has quit IRC
[22:23:16] *** Bronsky has quit IRC
[22:23:33] *** Bronsky has joined #postfix
[22:31:36] *** f3ew has quit IRC
[22:32:07] *** f3ew has joined #postfix
[22:34:36] *** ziro has joined #postfix
[22:36:49] *** gAri- is now known as _gAri-
[22:39:29] <Leonidas> is it true that postfix-tls 2.1.5 does not support smtpd_tls_auth_only properly? I tried setting it to 'yes', but it still accepts unencrypted authentification.
[22:48:16] *** nach0s has joined #postfix
[22:48:41] *** sepski has quit IRC
[22:49:08] <nach0s> Hi all i have a problem with a mailserver, i have postfix+ldap+cyrus on a opensuse 10.2 and i have this error when i try to receive email....  status=bounced (Command died with signal 11: "/usr/lib/cyrus/bin/deliver")
[22:56:25] *** nach0s has quit IRC
[23:02:26] *** UQlev has left #postfix
[23:04:56] *** hemry has quit IRC
[23:08:02] *** Bronsky has quit IRC
[23:08:03] *** backz has quit IRC
[23:11:59] *** eltech has joined #postfix
[23:20:31] *** TheOutlander has quit IRC
[23:21:46] *** hachiya has quit IRC
[23:30:04] *** backz has joined #postfix
[23:30:25] *** Bronsky has joined #postfix
[23:38:54] *** mcepl has joined #postfix
[23:41:08] <mcepl> Hi, is there anybody capable enough of internals of postfix to have some clue about https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=215722 -- why in the world postfix goes to /boot?
[23:43:17] <xpoint> mcepl, redhat
[23:43:33] <mcepl> xpoint: yes, and?
[23:43:46] <xpoint> postfix is not doing this
[23:44:44] <mcepl> xpoint: OK, let me be sure about what you are saying -- the fact that postfix tries to open /boot, /home, /tmp is in your opinion result of RedHat-specific patch?
[23:44:59] <xpoint> yes
[23:45:13] <xpoint> could be selinux ?
[23:45:32] *** war has joined #postfix
[23:45:37] <mcepl> no, SELinux *detects* this behavior, it doesn't do it!
[23:45:55] <mcepl> there is no way, how SELinux will make any program to go to /boot
[23:46:25] <mcepl> sorry, for exclamation mark -- that's probably unnecessary
[23:47:29] <xpoint> point is that its redhat, postfix sources dont do anything in boot
[23:47:37] *** hparker has quit IRC
[23:47:42] *** _gAri- is now known as gAri-
[23:49:57] <mcepl> xpoint: are you sure? you reject as a possible explanation https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=215722#c9 ?
[23:50:15] <mcepl> (or that this could be in the upstream postfix?)
[23:53:02] <sysmonk> mcepl: it does a getattr /boot ?
[23:53:11] <sysmonk> or tries to OPEN the /boot ?
[23:54:25] <sysmonk> mcepl: do you have /boot as a partition, or it's just a subfolder of / ?
[23:54:46] <mcepl> I got in https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=145732 only getattr / but some other user reported getattr on /boot, /home, and /tmp
[23:54:52] <mcepl> sysmonk: partition
[23:55:04] <sysmonk> mcepl: and /home and /tmp are partitions too, right?
[23:55:11] <mcepl> only /home
[23:55:57] * sysmonk ponders
[23:57:24] <mcepl> sysmonk: take a look at https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=141261 (I think it should be grokable even for non-RH guy, grep on getattr)
[23:57:29] <sysmonk> ok, just a guess, postfix doesn't queue mails if the partition where the queue dir ir has less than SOME_CONFIG_PARAM
[23:57:37] <mcepl> sysmonk: and of course, I have no clue about partitions of that other guy
[23:57:38] <sysmonk> so that it wouldn't fill up the space
[23:58:11] *** gAri- has quit IRC
[23:58:11] *** tminos has quit IRC
[23:58:11] *** xpoint has quit IRC
[23:58:12] *** piedoggie has quit IRC
[23:58:12] *** Leonidas has quit IRC
[23:58:12] *** Mez_ has quit IRC
[23:58:13] *** hooch has quit IRC
[23:58:15] *** Tachy_ has quit IRC
[23:58:17] *** bugz_ has quit IRC
[23:58:17] *** seekwill has quit IRC
[23:58:17] *** chiwawa_42 has quit IRC
[23:58:18] *** redondos has quit IRC
[23:58:18] *** Tachy has joined #postfix
[23:58:18] *** hparker has joined #postfix
[23:58:18] <sysmonk> so, just a guess, that it could take the info about all the partitions ...
[23:58:19] <mcepl> sysmonk: could you switch off this test (even with possibly fatal results for user)?
[23:58:22] *** hooch has joined #postfix
[23:58:30] *** tminos has joined #postfix
[23:58:31] *** gAri- has joined #postfix
[23:58:34] *** xpoint has joined #postfix
[23:58:45] *** chiwawa_42 has joined #postfix
[23:58:46] *** piedoggie has joined #postfix
[23:58:53] *** seekwill has joined #postfix
[23:59:28] <mcepl> sysmonk: curl 'https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=141261' | grep getattr | less
[23:59:32] *** pirho has joined #postfix
[23:59:36] *** bugz_ has joined #postfix
[23:59:54] *** Leonidas has joined #postfix





top