[00:02:04] *** jtaji has quit IRC
[00:04:26] *** sepski has joined #postfix
[00:09:56] <adaptr> !learn thumbs as an insignicifant snippet of C#
[00:10:33] <adaptr> !thumbs
[00:10:33] <knoba> adaptr: "thumbs" : an insignicifant snippet of C#
[00:10:48] <adaptr> speling gratuit
[00:12:05] *** sepski has quit IRC
[00:13:01] <thumbs> adaptr: I hate you.
[00:13:09] *** jeffspeff has joined #postfix
[00:14:09] <adaptr> so, what's new ?
[00:18:32] *** jeev_ is now known as jeev
[00:21:19] *** XPCFan has quit IRC
[00:21:55] *** XPCFan has joined #postfix
[00:22:23] <jeffspeff> i'm getting an input/output error in my maillog and am unable to login... can somebody help?   http://pastebin.ca/1291080
[00:22:50] *** Aleons has quit IRC
[00:26:03] <roe_> I have a more high level question rather than specific.  I have multiple servers: a.fake.domain.com b.fake.domain.com c.fake.domain.com, etc... They are behind a nat.  what is the "suggested" way to gather all of those emails to a single address: user at real dot address.net
[00:26:35] <roe_> "those emails" = system messages, usually addressed to root@machine
[00:28:45] *** jtaji has joined #postfix
[00:29:45] *** diazepam has quit IRC
[00:30:06] <adaptr> roe_: a fake domain will never, ever receive mail
[00:30:31] <roe_> a b and c represent servers on an internal domain
[00:30:36] <adaptr> but collecting them to a single mailbox is simple
[00:30:48] <adaptr> do you use those domains for any real mail ?
[00:30:56] <roe_> no
[00:31:07] <adaptr> so.. you don't actually need those domains
[00:31:31] <roe_> not for mail
[00:31:33] <adaptr> just set the relay to your postfix box on each server and accept mail. done
[00:31:48] <adaptr> ssmtp is well-suited for this
[00:32:27] <roe_> but I need to configure a mapping for root at [a,b,c] dot fake.domain.com to user at real dot address.net
[00:32:51] <adaptr> not really
[00:33:04] <jeffspeff> roe_, why don't you just use fetchmail?
[00:33:08] <adaptr> !parent_domain_matches_subdomains
[00:33:09] <knoba> adaptr: "parent_domain_matches_subdomains" : a configuration parameter in main.cf: What Postfix features match subdomains of domain.tld automatically, instead of requiring an explicit .domain.tld pattern. This is planned backwards compatibility: eventually, all Postfix features are expected to require explicit .domain.tld style patterns when you really want to match subdomains.
[00:33:30] <adaptr> it's set automatically to on for mydestination
[00:33:45] <roe_> jeffspeff, fetchmail is a hack
[00:33:49] <adaptr> so just have one alias where root goes to your user
[00:33:49] <roe_> at least in my mind
[00:34:03] <adaptr> roe_: ssmtp is simple and elegant
[00:34:11] <jeffspeff> roe_  it's good at what it does. :)
[00:34:13] <roe_> secure smtp?
[00:34:19] <adaptr> simple
[00:34:26] <adaptr> you can set a relayhost
[00:34:30] <adaptr> that's pretty much it
[00:34:36] <roe_> ala nullmailer
[00:34:51] <adaptr> yes, except that ssmtp is an actual MTA
[00:35:06] <adaptr> albeit only from localhost
[00:35:19] <roe_> does it expire messages properly?
[00:35:24] <roe_> nullmailer does not
[00:35:36] <adaptr> set each machine's relayhost to the postfix machine, make sure postfix accepts all subdomains, and alias root to whomever
[00:35:47] <adaptr> roe_: SMTP does not, it is an unkown concept
[00:36:23] <adaptr> if you mean, when sending, no - an MTA is assumed to always be available, why else run one ?
[00:36:53] <roe_> so collect them all to one machine on the network and from that machine send it out to the final destination
[00:37:13] <adaptr> if that machine is not the final destination, yes
[00:37:32] <roe_> nullmailer never creates DSNs
[00:37:36] <adaptr> OR set up th ealias file on each machine to send out to an external address
[00:37:46] <adaptr> roe_: why would it, ever
[00:38:23] <roe_> if something changes or breaks, months of mail CRON email gets backed up
[00:38:36] <adaptr> roe_: local process creates a mail to root. local mailer hands it off. there is no chance of ever getting a DSN, ever
[00:39:00] <adaptr> to WHAT ? you're not running an MTA locally, no MTA means no DSNs
[00:39:17] <roe_> right, but all of that is being redirected over the network, hence, now it is possible to get a DSN
[00:39:29] <adaptr> how ? how is this possible ?
[00:39:43] <roe_> I can show you the queues
[00:39:44] <adaptr> all the mail is created by you, controlled by you - and you presumably know where you're sending it
[00:40:15] <adaptr> you're making a very simple matter incredibly complicated
[00:40:42] <roe_> I realize that, and I am trying not to, give me a sec to read the parent_domain...
[00:42:28] <roe_> this is what that parameter is currently set to:
[00:42:32] <roe_> parent_domain_matches_subdomains = debug_peer_list,fast_flush_domains,mynetworks,permit_mx_backup_networks,qmqpd_authorized_clients,smtpd_access_maps
[00:43:19] <roe_> which I guess is the default.  So I just add .fake.domain.com and it will accept mail addressed to root at [a,b,c] dot fake.domain.com
[00:43:49] <adaptr> no, the list is obviously a list of SETTINGS where this is true
[00:43:54] <adaptr> there are no values in the list
[00:44:13] <adaptr> add mydestination
[00:44:31] <adaptr> and make sure you put fake.domain.com in mydestination
[00:44:42] <adaptr> all subdomains thereof will now be handled by postfix
[00:45:02] * adaptr thought this was on by default
[00:45:44] <adaptr> !masquerade_domains
[00:45:44] <knoba> adaptr: "masquerade_domains" : a configuration parameter in the main.cf: Optional list of domains whose subdomain structure will be stripped off in email addresses.
[00:46:02] <adaptr> hmm no you don't want that, or you'll lose the server machine info
[00:48:28] *** Juspion has joined #postfix
[00:53:15] <jeffspeff> i'm getting an input/output error in my maillog and am unable to login... can somebody help?   http://pastebin.ca/1291080     i can't tell if it's a postfix auth problem or a courier auth problem
[00:55:08] <adaptr> there is no "postfix auth"
[00:59:07] *** war9407 has quit IRC
[01:01:37] <adaptr> rob0: are you still awake ?
[01:01:42] *** pirho has quit IRC
[01:10:24] <roe_> the destination mail server is rejecting it because the sender domain is a non fqdn, do I need to whitelist domain.com or [a,b,c].domain.com
[01:10:32] *** pulsar has joined #postfix
[01:15:02] <adaptr> yes, I just ran into that myself, which is odd, let me check the manual
[01:15:54] <roe_> it looks like this is the right spot:  check_sender_access hash:/etc/postfix/sender_checks
[01:16:45] <roe_> but I'd like to whitelist .domain.com meaning domain.com and all subdomains, not just user at a dot domain.com
[01:17:29] <roe_> or should I just add that network's IP address to my networks
[01:19:42] <adaptr> I don't understand what it's doing, I made a small test where I added mydestination to parent_etc, then sent mail from a submachine, and it is accepted but then it tries to relay it
[01:20:10] <adaptr> and my relay_domains is empty, it should never relay
[01:20:16] <adaptr> !mydestination
[01:20:17] <knoba> adaptr: "mydestination" : a configuration parameter in the main.cf: The list of domains that Postfix delivers via the $local_transport mail delivery transport. By default, mail is given to the Postfix local(8) delivery agent that looks up all recipients in /etc/passwd and /etc/aliases, or their equivalents.
[01:20:41] <roe_> $mydestinaion?
[01:21:20] <adaptr> roe_: if postfix is rejecting them because of a non-fqdn address, that's an issue on the client side
[01:22:03] <roe_> there is no "client". the destination postfix server is rejecting mail, because cron at a dot fake.domain.com is non-fqdn
[01:22:19] <adaptr> of course it's not
[01:22:29] <roe_> correct
[01:22:30] <adaptr> and there is always a client
[01:22:54] <adaptr> if a at b dot com is non-fqdn, then which parts are missing ?
[01:23:05] <adaptr> it's as fqdn as you're ever going to get
[01:24:25] <roe_> right but the receiving mail server can't resolve it, thus I need to put an exception in on the receiving mail server to allow *.b.com (to use your example).
[01:24:53] <roe_> actually to be more accurate, * at * dot b.com
[01:25:14] <roe_> I could probably limit that to root@.
[01:26:32] <adaptr> that's not what is happening at all
[01:27:24] <adaptr> it tries to *send* the mail to root at a dot domain.com instead of root at domain dot com, and attempts to forward (relay) to a.domain.com via SMTP - which fails, for obvious reasons
[01:31:21] <roe_> ah, yes, you are correct
[01:35:00] <adaptr> I'm experimenting a bit, but not understanding what I see...
[01:35:25] <adaptr> I tried adding *.domain.com to mydestination, which it accepts, but it doesn't do what I wanted it to
[01:35:47] <roe_> the docs say .domain.com' is how to include subdomains
[01:35:54] <rob0> !parent_domain_matches_subdomains
[01:35:55] <knoba> rob0: "parent_domain_matches_subdomains" : a configuration parameter in main.cf: What Postfix features match subdomains of domain.tld automatically, instead of requiring an explicit .domain.tld pattern. This is planned backwards compatibility: eventually, all Postfix features are expected to require explicit .domain.tld style patterns when you really want to match subdomains.
[01:36:22] <adaptr> rob0: it's making zero sense, it's not working at all
[01:36:23] <roe_> yea, that's what is says :)
[01:36:27] *** BuenGenio has joined #postfix
[01:36:52] <rob0> if you UNSET that, you get the leading-dot syntax
[01:37:21] <rob0> postconf -e "parent_domain_matches_subdomains ="
[01:37:53] <adaptr> ..but if I SET it, it should just...match
[01:37:55] <adaptr> and it doesn't
[01:38:23] <adaptr> common sense tells me adding mydestination to parent_etc will accept mail for all subdomains of the domains in mydestination
[01:38:34] <adaptr> or, rather, that's what the docs say
[01:38:38] <adaptr> and it doesn't work
[01:39:23] <adaptr> it keeps trying to relay back to the originating machinem even though relay_domains is empty
[01:40:44] <roe_> not so easy is it :)
[01:41:09] <adaptr> I think rob0 hates me for unsticking his glue...
[01:42:02] <adaptr> I'm off to bed... not thinking straight anymore
[01:42:44] <rob0> match "example.com" if it IS set
[01:42:59] <rob0> match ".example.com" if it is UNSET
[01:43:06] <roe_> I have it unset
[01:43:13] <roe_> I *think* I do at least
[01:43:17] <rob0> me too, it's a future default
[01:43:36] <adaptr> rob0: it should match *.example.com IFF I have mydestination in parent_domain_
[01:43:40] <adaptr> that's what it's FOR
[01:43:57] <rob0> but what is "postconf mydestination"?
[01:44:09] <adaptr> domain.com .domain.com
[01:44:19] <adaptr> for NOW
[01:44:42] <adaptr> but when I had mydestination in parent_bla I had only domain.com in mydestination
[01:44:56] <adaptr> which SHOULD match all its subdomains
[01:45:04] <rob0> is this working as you expected?
[01:45:07] <adaptr> and it doesn't
[01:45:13] *** XPCFan has quit IRC
[01:45:38] *** Juspion has quit IRC
[01:45:40] <adaptr> I'll do one more test
[01:45:48] *** XPCFan has joined #postfix
[01:45:52] <rob0> and then off to bed!
[01:47:08] <adaptr> doesn't work
[01:47:12] <adaptr> parent_ is unset
[01:47:20] <adaptr> mydestination contains .domain.com
[01:47:30] <adaptr> and it tries to relay it back to the sending machine
[01:47:35] <adaptr> with an EMPTY relay_domains
[01:47:38] <adaptr> go figure
[01:47:41] <roe_> I have parent_... UNset.  I telnet from a.domain.com to b.domain.com, with sender:root at a dot domain.com recipient: root at a dot domain.com.  b.domain.com tries to send ti back to a
[01:47:48] <adaptr> ditto
[01:47:54] <adaptr> checked and doublechecked
[01:48:02] <roe_> glad I'm not the only one
[01:49:11] <adaptr> now I understand, sender and recipient are identical, but who cares ? postfix SHOULD accept mail to * at * dot domain.com and NOT try to relay for domains that are nowhere IN relay_domains
[01:49:59] <roe_> I'm assuming listing each subdomain in mydestination will work?
[01:50:28] <adaptr> I am not sure at this point, .domain.com SHOULD be identical to a.domain.com, b.domain.com, etc.
[01:50:37] <roe_> but that is a pain in the tuchas
[01:50:39] <adaptr> apparently postfix thinks otherwise
[01:50:45] <adaptr> well, you could hack it ;)
[01:51:32] <adaptr> set up a virtual_alias_domains map that lists *.domain.com, and a virtual_alias_maps that lists "root   you at yourdomain dot com"
[01:51:43] <adaptr> but it should not be necessaryto do this
[01:51:48] <roe_> I thought about that
[01:52:11] <roe_> I was all set to try that, then you mention this subdomain matching and seemed more appropriate
[01:52:18] <adaptr> yeah, right :)
[01:53:17] <adaptr> well, I give up
[01:53:24] <adaptr> good luck, the virtual hack should work
[01:53:35] <roe_> yea
[01:53:44] <roe_> should I submit a bug?
[01:54:27] <adaptr> thinking about it
[01:54:47] <adaptr> best to submit this to the postfix-users mailing list first, it's probably incredibly obvious but we're missing it ;)
[01:54:52] <rob0> llllloooooooggggggggggsssssssssss
[01:55:09] <adaptr> rob0: connecting back to $original_machine: refused.
[01:55:12] <rob0> pppppppooooooooossssssssstttttttcccccccooooonnnnnfffffffff
[01:55:13] <adaptr> as I have been sayyyingggg
[01:55:29] <adaptr> rob0: assume a WORKING system
[01:55:35] <adaptr> add .domain.com to mydestination
[01:55:39] <adaptr> clear parent_
[01:55:43] <rob0> yeah but adaptr is running it!!
[01:55:47] <rob0> ;)
[01:55:50] <adaptr> send mail from machine a.domain.com to mailer
[01:56:07] <adaptr> mailer accepts it, but wants to relay it back to a.domain.com
[01:56:40] <adaptr> tell me what more information you require that explains this!
[01:59:33] *** BuenGenio has quit IRC
[01:59:51] *** BuenGenio has joined #postfix
[02:00:41] <roe_> I think you scared him off
[02:01:13] <adaptr> he's not scared, he's just laughing at me
[02:04:45] *** ming_zym has joined #postfix
[02:04:52] *** BuenGenio has quit IRC
[02:05:11] *** BuenGenio has joined #postfix
[02:07:32] *** _val_ has joined #postfix
[02:10:08] <_val_> Hi, I'm running my postfix server on 1 serv. Now I've set another server with postifx. What's the best way to link those servers in 1 domain email at mydomain dot com . Is there something other than load balancing?
[02:11:15] <adaptr> such as ?
[02:12:04] <roe_> what does "link" mean
[02:12:28] <_val_> like  linking irc servers?
[02:12:33] <roe_> do you want both of them to receive mail for mydomain.com?
[02:12:45] <roe_> where is your mailstore going to live?
[02:12:53] <_val_> roe_: yes I want to use the same domain
[02:13:00] <_val_> roe_: 1 in holland 1 in Sweden
[02:13:27] <roe_> and the second question...
[02:13:53] <_val_> mailstore?  in holland
[02:14:09] *** BuenGenio_ has joined #postfix
[02:14:32] <roe_> ok so you want your Sweden server as a secondary mailserver that will accept mail from the world and deliver it to holland
[02:15:03] *** BuenGenio has quit IRC
[02:15:06] <_val_> roe_: yes, I also want to use the sweden server as a store server , so I could create users there
[02:15:40] <roe_> so you just increased complexity by 2 orders of magnitude
[02:15:47] <roe_> imho
[02:16:16] <adaptr> _val_: you can't simply "create users" there, sinec both servers must know which users belong where
[02:16:28] <adaptr> it's one order of magnitude, but still
[02:16:45] <_val_> adaptr: imagine if the store server goes down?
[02:17:09] <adaptr> _val_: you don't have one store server, and if one goes down, the other HOPEFULLY queues it
[02:17:29] *** Knoedel2 has quit IRC
[02:18:28] <_val_> adaptr: I'have linked a lot of irc servers I'm probably confusing the theoretical communication of how postfix communicates
[02:18:42] <adaptr> not just probably
[02:19:00] <roe_> you also have to worry about your pop/imap server
[02:19:23] <bigtone> _val_: IMO the complexity is in syncronising the users and domains.  You could have replicated LDAP, or slave MYSQL, or just plain 'ol "SSH in, dump list of users and domains, munge, create hash, erload postfix"
[02:19:38] <_val_> I see, but if my storeserver would be unavailiable.. nobody will have access to it's own mailbox anymore
[02:19:55] <roe_> right, so you need to replicate mailstores realtime
[02:19:57] <bigtone> I did the latter, because LDAP scares me, and MySQL was overkill for the size of my domains
[02:20:25] <_val_> bigtone: that would be a great idea yes.
[02:20:50] <_val_> I already have my webserver synchronised, but I didn't know if I could do the same with postfix
[02:22:00] <_val_> however it will effect the HD space
[02:22:14] <roe_> postfix isn't the challenge, its config isn't highly mutable
[02:22:25] <roe_> the mailstores on the other hand are
[02:23:23] <jangell> I think there is a fundamental problem with the understanding of e-mail to do the above.
[02:24:08] <adaptr> you do ? why ?
[02:24:20] <jangell> Well..What exactly is he wanting?
[02:24:31] <_val_> I installed the postfix everything is working. What I needed was just how to get some fail-proof
[02:24:35] <jangell> Two servers to just magically be in sync with both the authentication and mailstore and constantly update each other?
[02:25:10] <adaptr> I hope not, that's just silly
[02:25:21] <bigtone> _val_: do you want a highly available mailstore for your clients to use, or just highly available SMTP service so yuo can always *receive* emails?
[02:25:24] *** makerc has joined #postfix
[02:25:26] *** F6F has quit IRC
[02:25:36] *** sektor1952 has joined #postfix
[02:25:40] <sektor1952> evening
[02:25:52] <bigtone> adaptr: I din't think it's silly.  It's not trivial, though
[02:25:52] <jangell> _val_: you must first define what level of fail-proof you want.  If one server goes down--do you want the other server to just cache incoming e-mails or do you want it to continue to work perfectly for all your users?
[02:25:54] <_val_> bigtone: just to receive email.. it doens't have to be a lot of space
[02:26:30] <sektor1952> I installed postfix from source following a guide I found online are there any init.d scripts for centos 5.2?
[02:26:50] <_val_> jangell: it won't work for all clients since the clients will be devided between 2 servers
[02:26:54] <jangell> sektor1952: no idea.  It takes like 5 minutes to write an init script from scratch.  I'd suggest you just teach yourself how.
[02:27:37] <jangell> _val_: well what do you want to have happen when you lose one of the two servers?
[02:27:58] <jangell> _val_: Are you just wanting to have 50% of your users on one server and 50% on the other..and if you lose a server you lose 50% of your users?
[02:28:09] <sektor1952> Understood jangell I usually learn by following examples
[02:28:41] <_val_> jangell: I'm trying to find a solution to save some disk space.
[02:29:00] <jangell> _val_: ok..so you want two servers..and each server is responsible for X users?
[02:29:02] <_val_> but I think I will have to use syncronisation of the users on both servers.
[02:29:11] <roe_> _val_, that doesn't answer his questino
[02:29:27] <_val_> jangell: 2 servers , same users
[02:29:35] <roe_> so if one fails...
[02:29:44] <roe_> the users don't notice anything?
[02:29:45] <_val_> roe_: yep that's it
[02:30:01] <sektor1952> load balancing/fail over
[02:30:41] <jangell> _val_: That would be sort of a bitch.  First off you'd need a central location where the clients make their SMTP / IMAP / whatever connections..and that location then sends it to one of the two servers. (this needs to be redundant as well) and then you would need to keep all your authentication and your mail store in sync..and keeping the mail store in sync would be a major challenge.
[02:31:13] <jangell> _val_: Let me describe a traditional clustered setup...mine for example
[02:31:15] <roe_> I'm sure there is a rdms mailstore of some kind
[02:31:29] <roe_> that can be kept replicated
[02:31:46] <_val_> jangell: I'm listening!
[02:32:30] <_val_> roe_: I did the same with my rootfolders of the web.. but it is much easier than this, at least that's how I experience it
[02:32:59] <jangell> _val_: I have load balancers which are clustered and redundant.  All incoming connections hit the load balancer.  The load balancer then sends it to one of my servers.  I have two SMTP servers, two IMAP servers, two POP3 servers.  If one of those servers fails the load balancer detects that and won't send any traffic to those anymore.  I also have an OpenLDAP cluster which is my central location for authentication.  All the e-mail is stored on a Netapp SA
[02:33:28] <adaptr> jangell: he has to geo locations
[02:33:31] <adaptr> *two
[02:33:41] <jangell> adaptr: I'm aware.  That doesn't really change anything though
[02:33:50] <adaptr> no centralized storage
[02:33:50] <jangell> adaptr: It just makes the mail store even harder.
[02:34:05] <jangell> The fact of the matter is..keeping a Maildir mail store in sync..wouldn't be simple.....
[02:34:48] <jangell> basically
[02:34:53] <jangell> you could use a network file system
[02:35:01] <jangell> but..you better have a really fast pipe inbetween
[02:35:03] <_val_> jangell: ok thanks for the explainatino.. I think sync would be the solution, create same users on both servers, and also try to prevent the fail of both servers at once
[02:35:20] <jangell> _val_: sure.  but..how do you plan on keeping your mail store in sync?
[02:35:38] <_val_> jangell: an bashscript?
[02:35:43] <roe_> ha
[02:35:44] <jangell> _val_: that does what exactly?
[02:36:05] <jangell> _val_: the mail store sync problem..isn't exactly an easy one to solve :)
[02:36:10] <_val_> jangell: at least did it with webserver.. it will copy the content of each mailbox.. to the other server
[02:36:28] <jangell> okay
[02:36:33] <jangell> it basically needs to be instant
[02:36:39] <jangell> and it needs to deal with changes happening on either end
[02:36:47] <roe_> do you webserver files change a 100times a minute?
[02:36:58] <jangell> it's not like you can just kick off rsync every .0001 seconds
[02:37:14] *** stas has quit IRC
[02:37:23] <jangell> It would take rsync minutes before it could even decide what to do to sync htem..and by that time..it's way too late.. rsync wouldn't work
[02:37:28] <_val_> if user at domain dot com get's mail in holland than  user at domain dot com will be synced in sweden. Make a script. put in the cronjob  and look for every minute if there are made changes
[02:37:56] <jangell> _val_: good luck with that.
[02:38:04] <jangell> _val_: that would be a huge mess
[02:38:08] <jangell> _val_: your syncs will take forever
[02:38:18] <jangell> _val_: your clients will get e-mails twice
[02:38:27] <jangell> _val_: client connects to server 1..downloads email...server 1 deletes email
[02:38:37] <jangell> client connects again..hits server 2..wow the email is still on server two..they get it again
[02:38:51] <jangell> there is no way you could get a rsync to be fast enough
[02:39:12] <jangell> you'd need some sort of distributed mailstore..not sure what that would be...you might be able to use a distributed file system... It would be a project.
[02:39:56] <_val_> jangell: I than probably have to figure out something else. U are right it still won't have the effect I want.
[02:40:04] <jangell> keep in mind that a busy Maildir system is going to have millions of files that are changing hundreds of times a second.
[02:40:17] <jangell> How many domains/users do you have?
[02:40:19] <_val_> than I probably will use 1 server as the mailstore the otheron .. just redirecting the mail and stuff
[02:40:29] <_val_> jangell: 100 by now
[02:40:55] <jangell> jangell: personally...I'd just have the 2nd server work as a cache basically with a 2nd MX record
[02:41:06] <jangell> that way if the first server goes down..the second one will cache incoming e-mails until you get the first one back up again
[02:41:16] *** Motoko-chan has joined #postfix
[02:41:22] <jangell> 100 users...is not worth the complexity that this would take....unless they're paying you like $1000 a month per user :)
[02:42:12] <_val_> jangell: I wish it was true. :)  anyways, thanks for your advice. And thank you everyone who helped
[02:42:46] <_val_> I know everything, but postfix :p . lol
[02:42:51] <_val_> j/k
[02:43:03] <jangell> you'd hit this problem no matter which mail server you chose
[02:43:33] <jangell> If I have to do geo seperation..I'd probably go with an Active / Passive setup
[02:43:42] <jangell> replicate the file system to the standby location
[02:44:00] <jangell> and then have some sort of near-instant rapid fail over system incase the primary site went down
[02:44:12] <_val_> jangell: I configuring postfix in it's own isn't a big deal. but getting redundacy and this what I mentioned is pretty hard.  How is hotmail doing it? :)
[02:44:44] <jangell> With million dollar budgets.
[02:45:09] <jangell> multi million dollar
[02:45:45] <roe_> and large large pipes
[02:46:05] <_val_> uuhm, I thought I could keep up with them, dream gone
[02:46:18] *** nonsequitir has quit IRC
[02:46:23] <jangell> _val_: It is funny..e-mail is so simple..but it can be so tricky sometimes
[02:46:27] <jangell> It is rather demanding
[02:46:34] *** jeffspeff is now known as jeffspeff[A]
[02:46:49] <jangell> My company's primary focus is web...but we provide e-mail hosting to our clients..and the e-mail I/O / hardware demand is greater than the web by many times.
[02:47:39] <_val_> jangell: true. And I'm always curious how things work behind the nice webbased applications. For the end user it's not that importand. What they want is "stability"  so they can check their mail any place any time
[02:48:21] <jangell> _val_: you can offer some pretty powerful user experiences with free open source software.
[02:48:34] <jangell> _val_: there are some really nice web based mail front ends out there...like roundcube
[02:48:44] <jangell> _val_: if you're looking for groupware..Zimbra is pretty impressive
[02:49:07] <_val_> jangell: roundcube is what I'm running at the moment. Beta thogh
[02:49:15] <roe_> roundcube is pretty, though slow in development
[02:49:30] <jangell> roundcube is no longer beta
[02:49:30] <tomocha6> i think os..
[02:49:32] <jangell> and hasn't been..for a long time
[02:49:33] <tomocha6> so
[02:49:35] <_val_> jangell: I personally send mails via my cli ^^ as I'm a cli geek ;p
[02:49:46] <jangell> I'm in the process right now of writing a PHP front end for managing domains, users,  alias domains,  and forwards, reporting,  distributed storage,  etc for my company right now.
[02:49:47] <bigtone> _val_: IMO it'sd doable - imapsync is the key. (this is syncronising the users mailbox contents)
[02:50:11] <jangell> might be able to release it at some point under the BSD license... we'll see... but it is more powerful than the JAMM/Phamm/ postfix admin stuff I've seen so far
[02:50:45] <roe_> jangell, something like phamm or postfixadmin?
[02:50:53] <roe_> ha, nm
[02:50:56] <roe_> didn't finish reading
[02:51:02] <jangell> roe_: Like those..but better looking..and more of the features i wanted.
[02:51:07] <_val_> bigtone: ? imapsync.. how does it work exactly?
[02:51:08] <roe_> what backend
[02:51:22] <jangell> roe_: openldap
[02:51:28] <roe_> nice
[02:51:46] <jangell> the big thing I needed
[02:51:47] <jangell> was alias domains
[02:51:48] <roe_> do you address per user sieve/maildrop/procmail filters
[02:52:09] <_val_> jangell: I just don't trust the webbased applications, especially php. They are pretty sensitive to exploits
[02:52:15] <jangell> basically..underneath a primary domain..you can add an alias domain..and instantly everything@primarydomain works as everything@aliasdomain and all of that forwards to the primary
[02:52:43] <jangell> _val_: they're only as good as the guy that wrote them.  PHP in itself is a great language for web work
[02:53:23] <jangell> roe_: Not right now.  I'm focusing on our business needs only.
[02:53:36] <jangell> roe_: I'm also writing web services so that our other products can integrate.
[02:53:46] <_val_> jangell: true, I do script myself in php, but 2 dutch leadin php members , resigned from supporting php anymore, as it keeps getting exploited
[02:54:23] <jangell> _val_: my company uses PHP exclusively and has some complex applications
[02:54:26] <_val_> I'm not saying other languages aren't exploitable, just don't know
[02:54:35] <jangell> _val_: It isn't hte language that gets exploited.
[02:54:58] <jangell> _val_: It is the application developed on the language that gets exploited and it happens because of bad business logic.
[02:55:32] <jangell> _val_: PHP is easy to start programming on.  The problem with that is beginner programmers tend to develop very insecure applications.  the *APPLICATION* is insecure,  not the langauge.
[02:55:43] <_val_> jangell: You know what I mean. A language can't be exploited indeed, it's the application that gets exploited that's made in that particular programming lang.
[02:56:01] <jangell> _val_: If you through those same programmers in C they would write worse crap..and wouldn't actually get anything done..ebcause they'd have no idea how..and if they did..it would have massive security exploits and major memory leaks
[02:56:13] <jangell> threw*
[02:56:27] <_val_> jangell: true, I quit programming in php, did for a while.
[02:57:12] *** growltiger has joined #postfix
[02:57:20] *** growltiger_ has quit IRC
[02:57:50] <_val_> jangell: yes, it would have much more influnce if a C application get exploited. php isn't able to manage memory , stack etc.. whereas  C and asm does this
[02:59:20] <_val_> jangell: I think we aren't at the toppic. but anyways, it was nice having this conversation with you.
[02:59:48] <jangell> roe_: http://jesseangell.com/forums/app.png
[02:59:58] <jangell> roe_: There is a screen shot of one portion of the app.
[02:59:59] <_val_> and Thanks again for the advise . I'll take all in consideration and decide what will be the best thing to use
[03:00:03] <jangell> roe_: with some of my company info painted over..
[03:03:05] <_val_> jangell: looks great. background isn't really my taste. As I often choose for black.
[03:04:00] <jangell> _val_: Oh..sure..corporate colors..I didn't pick them :)
[03:04:41] <jangell> back to working on the car in the garage....
[03:04:52] <_val_> jangell: yeah, I thoguht so, as I did php projects, we had to use corporate colors too
[03:05:23] <_val_> ok take care
[03:05:25] *** _val_ has left #postfix
[03:06:02] *** growltiger has quit IRC
[03:06:13] *** growltiger has joined #postfix
[03:07:29] *** makerc has quit IRC
[03:08:50] *** stas has joined #postfix
[03:09:11] *** sektor1952 has quit IRC
[03:09:29] *** sektor1952 has joined #postfix
[03:11:12] *** luke-jr has joined #postfix
[03:11:27] <luke-jr> well, I don't know how, but Postfix has decided to start rejectign SASL with 0 changes :/
[03:12:28] <roe_> jangell, looks nice
[03:12:58] <roe_> what is nicer is the leopard machine you took that screenshot from
[03:16:09] <luke-jr> wtf, it won't relay no-SASL anymore either >:O
[03:17:29] <luke-jr> whoa, wait a minute
[03:17:32] <luke-jr> this isn't even relaying
[03:17:34] <luke-jr> it's local delivery
[03:17:40] <luke-jr> Dec 22 02:16:35 [postfix/smtpd] NOQUEUE: reject: RCPT from unknown[2002:46bb:1a76:0:20e:a6ff:fec4:4e5d]: 554 5.7.1 <unknown[2002:46bb:1a76:0:20e:a6ff:fec4:4e5d]>: Client host rejected: Access denied; from?luke at dashjr dot org> to?corinne-jr at dashjr dot org> proto=ESMTP helo?tsuruki.localnet>
[03:18:15] * luke-jr drops packets to port 25 to minimize damage
[03:18:19] <sahil> ew, ipv6, so ugly.
[03:18:54] <rob0> postconf smtpd_client_restrictions
[03:19:07] 
[03:19:07] * rob0 bets there's a "reject" in it
[03:19:23] <rob0> hmm
[03:19:35] *** growltiger has quit IRC
[03:19:37] <luke-jr> reminder this was all working 10 mins ago
[03:19:41] <luke-jr> nothign has changed since
[03:19:41] *** growltiger has joined #postfix
[03:19:53] <thumbs> then everything is fine
[03:19:56] <sahil> lol
[03:20:01] <rob0> you should send in for a refund!
[03:20:05] <sahil> yes!
[03:20:16] <sahil> or, at the very least, postfix store credit.
[03:20:38] <rob0> I wonder if they still sell the mice with the mailbags?
[03:22:08] <rob0> that log line looks odd, anyone else notice that?
[03:22:32] <roe_> the helo?
[03:22:36] <rob0> from<= ... to<= ... helo<=
[03:22:46] <roe_> yea
[03:23:03] <jangell> looks broken.
[03:23:20] <jangell> in my,  unprofessional,  I've only used Postfix for two weeks,  opinion.
[03:23:28] * sahil smells trollation.
[03:23:33] <luke-jr> nothing unusual in the log preceding the failures
[03:23:59] <thumbs> luke-jr: postfix does not break for no reason.
[03:24:00] <rob0> sahil, but Luke has been around here before
[03:24:23] <luke-jr> rob0: ah, = < is replaced with =< for me
[03:24:26] <luke-jr> err
[03:24:35] <luke-jr> <= maybe?
[03:24:46] <jangell> luke-jr: so this suddenly just started happening for local delivery,  or everything?
[03:24:49] <luke-jr> anyhow, it's not really relevant
[03:24:51] <rob0> Destruction of municipal property? Huh?
[03:24:56] <luke-jr> jangell: I was trying to send a reply
[03:25:08] <luke-jr> first Postfix began rejecting my auth
[03:25:15] <sahil> perhaps you've been owned?
[03:25:18] <luke-jr> checked my pass etc
[03:25:22] <sahil> hax0red, if you will.
[03:25:24] <rob0> how long have you been using ipv6 here?
[03:25:28] <luke-jr> then said screw it and disabled auth, sicne I'm on the local network
[03:25:29] <rob0> there
[03:25:32] <luke-jr> rob0: year+
[03:25:50] <roe_> luke-jr, nothing you know about
[03:26:00] <jangell> luke-jr: does it happen locally as well... telnet localhost 25 on the mail server?
[03:26:22] <rob0> should be using submission, not 25
[03:26:35] <luke-jr> localhost works
[03:26:55] <luke-jr> rob0: using port 465 for sending mail
[03:27:02] <rob0> okay, read /topic and comply
[03:28:35] <luke-jr> http://rafb.net/p/ZzMKrn65.html
[03:28:38] <luke-jr> that's postconf -n
[03:30:10] *** sektor1952 has left #postfix
[03:30:16] <luke-jr> http://rafb.net/p/QNTzWJ36.html -- last successful send at 01:58
[03:30:21] <rob0> now the smtps line in master.cf
[03:30:51] <luke-jr> http://rafb.net/p/aOe4Kz97.html -- SASL reject
[03:31:22] <luke-jr> http://rafb.net/p/7vBuzv94.html -- master.cf snippet
[03:31:39] <luke-jr> the SASL reject above is 02:02
[03:31:51] <rob0> It said password auth failed
[03:32:00] <luke-jr> rob0: I noticed.
[03:32:09] <luke-jr> that was why I first verified my password is right
[03:32:16] <luke-jr> even though it was stored in my email program
[03:32:36] <rob0> something in the SASL backend
[03:32:46] <luke-jr> that was my guess next
[03:33:01] <luke-jr> oh, SMTPS is disallowing non-authenticated even if local
[03:33:14] <luke-jr> so what does the SASL stuff? :/
[03:37:40] *** spiderbatdad has joined #postfix
[03:37:57] <spiderbatdad> o/
[03:38:19] <spiderbatdad> anyone helping noobs?
[03:38:52] <luke-jr> no ?
[03:39:02] <luke-jr> spiderbatdad: my SASL pooped and I don't know where to go to troubleshoot now
[03:40:10] <spiderbatdad> luke-jr, what errors?
[03:41:08] *** growltiger_ has joined #postfix
[03:41:36] <spiderbatdad> here's my pain: postfix/smtpd[9903]: fatal: bad string length 0 < 1: etc/postfix/mysql_mailbox.cf_dbname =
[03:41:47] *** growltiger has quit IRC
[03:42:07] <luke-jr> spiderbatdad: password failures
[03:43:26] <spiderbatdad> luke-jr, you have a authdaemonrc and authmysqlrc?
[03:44:42] <luke-jr> authdaemonrc, as part of Courier
[03:45:06] <spiderbatdad> i was having password failures. after enough repetition now squirrel just tells me: dropped by imap server
[03:45:17] <spiderbatdad> but that isnt the same as password failure
[03:47:13] <luke-jr> hmm
[03:47:25] <luke-jr> with this in mind, could my mail client be broken
[03:47:47] <spiderbatdad> same permissions
[03:48:08] <spiderbatdad> lol
[03:48:11] *** gutocarvalho has quit IRC
[03:50:27] *** itrebal has joined #postfix
[03:50:35] *** itrebal has left #postfix
[03:51:09] <spiderbatdad> imapd-ssl: authentication error: Input/output error
[03:51:50] <spiderbatdad> ah
[03:53:10] <spiderbatdad> going to have to get apache using ssl for the squirrel mail page i think
[03:53:56] *** Zeit|awy_ has joined #postfix
[03:54:50] *** growltiger has joined #postfix
[03:55:16] <spiderbatdad> luke-jr, it seems your problem is in postfix/main.cf postfix/master.cf
[03:58:25] <spiderbatdad> luke-jr, or more likely in /sasl/smtpd.conf
[03:59:32] <luke-jr> spiderbatdad: not a one has changed in a long time
[03:59:59] *** Zeit|awy has quit IRC
[04:00:13] <spiderbatdad> luke-jr, ah I meant to ask if it was previously working :P
[04:01:41] <roe_> can you put sasl in more verbose logging mode?
[04:02:40] <luke-jr> roe_: how?
[04:03:35] <roe_> but man saslauthd to be sure
[04:05:30] <roe_> yes, the cli option
[04:05:57] <roe_> so something like saslauthd -c - m /var/run/saslauthd -r -d
[04:06:06] <luke-jr> I use init scripts
[04:06:13] <luke-jr> authtest says my password is good
[04:06:16] <luke-jr> so it's postfix or the client
[04:06:32] <roe_> what distro?
[04:07:12] <roe_> enabling debug on all the moving parts while they are moving is going to be the best way to find out why one of them isn't
[04:07:16] <luke-jr> Gentoo
[04:07:25] *** ekneuss has joined #postfix
[04:07:30] <luke-jr> anyhow, I'm blaming my client
[04:07:40] <luke-jr> and turning off auth until I leave the LAN
[04:07:47] <luke-jr> probably never, for this computer
[04:07:56] <ekneuss> Hi, I'd like to migrate from qmail to postfix, is there a guide or something that would help me do that ?
[04:08:05] <ekneuss> I'm worried about the mail formats and settings
[04:08:13] <roe_> wait... luke-jr it is only happening to your one machine?
[04:10:07] <roe_> ekneuss, what "mail formats" and what "settings"
[04:10:14] *** growltiger_ has quit IRC
[04:10:31] *** growltiger_ has joined #postfix
[04:10:53] *** growltiger has quit IRC
[04:12:45] <ekneuss> roe_, I mean, to handle mailboxes and so on
[04:13:24] <roe_> like mbox vs maildir?
[04:13:35] <roe_> postfix can deliver to either
[04:15:27] <ekneuss> ok
[04:16:32] <roe_> as I have never used qmail, I can't guide you in a high level manner, if you have specific questions or concerns I can probably answer them
[04:17:30] <Motoko-chan> There is always #qmr, but it's usually quiet.
[04:17:48] <Motoko-chan> Anyway, moving from qmail to postfix isn't too difficult.
[04:17:53] <Motoko-chan> Are you using virtual users?
[04:17:57] <ekneuss> Sadly I'm a bit taking over an existing configuration that's crippled
[04:18:22] * Motoko-chan knows how that it
[04:18:23] <Motoko-chan> is
[04:18:40] <Motoko-chan> Still waiting for the time to do the migration where I work.
[04:18:43] <roe_> I find that most other mta irc channels are usually dead
[04:18:54] <roe_> never ventured into #exim though
[04:19:03] <ekneuss> I probably need to fully understand how it was previously configured in order to move it...
[04:19:04] <Motoko-chan> Have a ton of users, and slowly removing special things from it (like pop-before-smtp)
[04:19:10] <Motoko-chan> Yeah
[04:19:13] <Motoko-chan> Using vpopmail?
[04:19:18] <ekneuss> but following a qmail configuration is... hard
[04:19:19] <ekneuss> yes
[04:19:23] <Motoko-chan> Okay.
[04:19:27] <Motoko-chan> csb-based, or db?
[04:19:32] <Motoko-chan> Um, sdb
[04:19:33] <Motoko-chan> cdb
[04:19:37] * Motoko-chan can't type
[04:19:48] <ekneuss> no db afaik
[04:20:05] <ekneuss> I'd like to move to a DB though
[04:20:08] <Motoko-chan> You'll likely want to move to db-based first.
[04:20:38] <Motoko-chan> I suggest getting the current install stable.
[04:20:44] <Motoko-chan> Then planning the migration
[04:21:34] <ekneuss> yeah
[04:21:52] <Motoko-chan> The first thing is in understanding how qmail works and how the pieces work.
[04:22:03] <Motoko-chan> Also, in mapping out what services you are currently offering.
[04:22:11] <luke-jr> roe_: I am the only SASL user at the moment.
[04:22:53] <Motoko-chan> Do some test configurations on a spare install, then plan the re-build.
[04:23:05] <Motoko-chan> Once that is stable, plan the move to db-based accounts.
[04:23:20] <roe_> SASL +TLS/SSL?
[04:23:23] <Motoko-chan> That'll let you use Postfix virtual users with the existing info you now have in the database.
[04:24:08] <ekneuss> It's not a really big mail server though, only ~80 domains representing 500 accounts, most of them half empty, but that still is clients data :) So I might as well do that carefully
[04:24:15] <roe_> luke-jr, is it possible what changed was your location? (lan vs wan)
[04:24:22] <Motoko-chan> Yeah, best to do that.
[04:24:35] <Motoko-chan> I manage over 1k domains, and maybe 5k users or more.
[04:24:42] <luke-jr> roe_: no
[04:24:45] *** mavrick61 has quit IRC
[04:24:51] <luke-jr> roe_: this is a desktop machine in the same buidling
[04:24:53] <Motoko-chan> So I have to be very careful. Plus all my other tasks taking priority.
[04:24:59] <luke-jr> SASL+SSL
[04:25:00] <ekneuss> that's another dimension indeed :)
[04:25:03] <roe_> oh well, thought I was on to something
[04:25:38] <Motoko-chan> ekneuss, Courier for IMAP?
[04:25:41] <Motoko-chan> Or no IMAP?
[04:25:42] <roe_> are you sure SSL is working?
[04:25:51] *** mavrick61 has joined #postfix
[04:25:55] <luke-jr> roe_: I'm getting the error message..
[04:26:01] <Motoko-chan> Actually, pop in #qmr so we don't stink up this channel with qmail talk
[04:26:11] <Motoko-chan> It's dead there now.
[04:26:47] <ekneuss> Motoko-chan, courier yes
[04:28:04] <Motoko-chan> ekneuss, #qmr for further discussion since it's kinda off-topic
[04:28:40] <ekneuss> Motoko-chan, ok, thanks
[04:31:15] *** xpoint has joined #postfix
[04:33:41] *** yajith has joined #postfix
[04:40:09] *** jimpop has joined #postfix
[04:56:18] *** Spec has joined #postfix
[04:56:53] *** spiderbatdad has quit IRC
[05:02:59] *** XPCFan has quit IRC
[05:14:29] *** jimpop has quit IRC
[05:20:16] *** _Driver_ has quit IRC
[05:20:42] *** _Driver_ has joined #postfix
[05:25:25] *** amrit|zzz is now known as amrit
[05:53:11] *** Jense has quit IRC
[05:53:41] *** Jense has joined #postfix
[05:57:48] *** Mr_Bee has joined #postfix
[05:59:27] *** f3ew has joined #postfix
[06:13:45] *** Haris1 has joined #postfix
[06:25:49] *** Mr_Bee has quit IRC
[06:25:54] *** bhagat has joined #postfix
[06:29:13] *** XPCFan has joined #postfix
[06:41:42] *** jeffspeff[A] has quit IRC
[07:00:33] *** saurabhb has joined #postfix
[07:10:03] *** ashWIn has joined #postfix
[07:10:36] <ashWIn> can anybody let me know a doc on howto postfix +LDAP
[07:11:01] <roe_> !ldap
[07:11:02] <knoba> roe_: "ldap" : a lookup method that can be used by Postfix. An introduction can be found in the LDAP_README also found at http://www.postfix.org/LDAP_README.html. A worthy project dealing with LDAP and Postfix can be found at: http://jamm.sourceforge.net/howto/html/
[07:11:20] <ashWIn> knoba: thanks
[07:11:27] <roe_> knoba is a bot
[07:11:44] <roe_> !knoba
[07:11:45] <knoba> roe_: "knoba" : an informational bot in this channel (see http://workaround.org/f=postfix)
[07:21:40] *** Verilium has left #postfix
[07:21:48] *** emcepe has joined #postfix
[07:21:52] *** Verilium has joined #postfix
[07:22:01] *** mcp has quit IRC
[07:22:13] *** Verilium has left #postfix
[07:22:16] *** Verilium has joined #postfix
[07:22:32] *** emcepe is now known as mcp
[07:28:35] *** XPCFan has quit IRC
[07:28:57] *** ashWIn has quit IRC
[07:29:10] *** XPCFan has joined #postfix
[07:45:02] *** BuenGenio_ has quit IRC
[07:45:12] *** ashWIN has joined #postfix
[07:50:52] *** Zumu has quit IRC
[07:55:47] <ashWIN> any good howto on postfix with ldap
[08:00:03] <Motoko-chan> Check postfix website
[08:03:08] *** pulsar has quit IRC
[08:03:21] *** pulsar has joined #postfix
[08:04:38] *** pulsar has quit IRC
[08:08:34] *** pulsars has joined #postfix
[08:09:01] *** k4z has quit IRC
[08:10:07] *** hednod- has joined #postfix
[08:16:43] *** hparker has quit IRC
[08:22:08] *** XPCFan has quit IRC
[08:22:42] *** XPCFan has joined #postfix
[08:23:09] *** sophokles has joined #postfix
[08:23:37] *** hednod has quit IRC
[08:23:37] *** hednod- is now known as hednod
[08:27:26] *** xpoint has quit IRC
[08:34:28] *** Jacolyte has joined #postfix
[08:41:29] *** Aleons has joined #postfix
[08:44:05] *** BuenGenio has joined #postfix
[08:47:29] *** Zblakany has joined #postfix
[09:12:42] *** denis has joined #postfix
[09:15:36] *** XPCFan has quit IRC
[09:28:24] *** r3r3 has joined #postfix
[09:30:14] *** jeffspeff has joined #postfix
[09:34:24] *** denis has quit IRC
[09:36:24] *** kiliko has quit IRC
[09:39:28] *** Motoko-chan has quit IRC
[09:39:34] <jeffspeff> how can i view in real time incoming/outgoing mail and when a user logs in to check mail and any errors trying to check the mail
[09:39:53] <R1ck> mailq, logfiles
[09:40:03] <R1ck> qshape
[09:40:06] <yajith> tail -f /var/log/maillog ?
[09:40:20] <jeffspeff> yajith not showing enough detail
[09:40:33] <jeffspeff> is there a way to make it more verbose?
[09:40:40] <yajith> think so..
[09:40:44] <yajith> but not sure how to do it..
[09:40:51] <yajith> someone else should know..
[09:40:53] <yajith> :D
[09:41:18] <ronr> is it possible to configure postfix to only accept mails that are bounces from mails it has sent itself?
[09:41:50] <yajith> someone in here* :D
[09:41:55] *** my007ms has left #postfix
[09:43:30] <yajith> jeffspeff: http://www.postfix.org/DEBUG_README.html#debug_peer
[09:45:08] *** r33 has quit IRC
[09:45:31] <f3ew> ronr google for BATV
[09:45:50] <ronr> f3ew: thx
[09:46:44] *** higuita has joined #postfix
[09:51:57] *** mavrick61 has quit IRC
[09:54:10] *** amrit is now known as amrit|zzz
[09:59:10] *** war9407 has joined #postfix
[10:10:34] *** jeffspeff is now known as jeffspeff[A]
[10:10:34] * jeffspeff[A] is now away - Reason : Auto-Away after 30 minutes
[10:14:52] *** ashWIN has quit IRC
[10:22:08] *** diazepam has joined #postfix
[10:35:45] *** wdp has joined #postfix
[10:37:02] <jangell> What do you guys do about the address_verify settings?  It seems to me like I have to somehow expire that cache whenever an account or domain is added or created...which is very often...and it seems like the only way to do that is to reload postfix..am I missing something?
[10:47:20] <sysmonk> reload postfix to do what?
[10:47:42] <sysmonk> also, you can make the address verify database disk-based, whereas default is memory based
[10:47:49] <sysmonk> that way even after restart you won't loose the database
[10:48:06] <sysmonk> but anyway, you don't need to reload postfix to add accounts/domains
[10:53:24] *** sypher has joined #postfix
[10:55:44] *** jeffspeff[A] is now known as jeffspeff
[10:55:44] * jeffspeff is no longer away : Gone for 1 hour 15 minutes 10 seconds
[11:01:09] *** diazepam has left #postfix
[11:08:07] *** Haris_ has joined #postfix
[11:08:08] *** Haris________ has quit IRC
[11:08:42] <jeffspeff> what is the correct channel for courier?
[11:18:46] <R1ck> #courier?
[11:30:47] *** Pazzo has joined #postfix
[11:38:10] *** jangell has quit IRC
[11:38:42] *** jeffspeff is now known as jeffspeff[A]
[11:38:42] * jeffspeff[A] is now away - Reason : Auto-Away after 30 minutes
[11:38:44] *** Juspion has joined #postfix
[11:39:03] <shasta> argh
[11:39:20] <shasta> jeffspeff[A], http://sackheads.org/~bnaylor/spew/away_msgs.html
[11:39:27] <shasta> ie. you suck
[11:59:40] *** bhagat has quit IRC
[12:02:07] *** bhagat_ has joined #postfix
[12:03:31] *** QuantumTim has joined #postfix
[12:04:58] *** wdp_ has joined #postfix
[12:12:44] *** yajith has left #postfix
[12:21:28] *** wdp has quit IRC
[12:27:00] *** Juspion has quit IRC
[12:35:21] *** ming_zym has quit IRC
[12:43:52] *** jonez has joined #postfix
[12:43:53] *** Jacolyte has quit IRC
[12:44:54] <jonez> greetings
[12:46:13] <jonez> I need to requeue a lot of messages that are currently set to be delivered to a machine that is no longer on my network. I tried 'postsuper -r ALL' but all that does is requeue with the same destination address. I have done this in the past, but I did not document the process. help? :)
[12:48:58] <jonez> the queue on the mail server is not draining after I run the "postsuper -r"
[12:49:27] *** pirho has joined #postfix
[12:55:08] *** kiliko has joined #postfix
[13:00:16] *** XPCFan has joined #postfix
[13:03:32] *** XPCFan has quit IRC
[13:04:07] *** XPCFan has joined #postfix
[13:07:12] *** Jacolyte has joined #postfix
[13:08:09] *** war9407 has quit IRC
[13:09:41] *** cpm has joined #postfix
[13:10:37] <f3ew> jonez postfix flush?
[13:11:08] <jonez> hmm.. is it that simple? figures ;)
[13:11:38] <jonez> still not draining the queue
[13:11:44] <jonez> maybe I should be more patient
[13:12:04] *** war9407 has joined #postfix
[13:15:18] <f3ew> postsuper -r ALL will requeue as if it was ne mail
[13:15:20] <f3ew> new
[13:17:55] *** abbe has joined #postfix
[13:18:01] *** pulsars has quit IRC
[13:18:01] <abbe> hi all
[13:18:04] <jonez> so it will look at the .forward file again?
[13:18:28] <abbe> I'm having a issue with postfix and spamassassin.
[13:18:40] <jonez> what I am trying to do is requeue mail destined for jam@dionysus and send it to jam@cyclops instead
[13:19:34] <f3ew> how?
[13:19:39] <f3ew> Did the MX change?
[13:19:42] <f3ew> IP?
[13:21:22] <jonez> dionysus died
[13:21:26] *** Jacolyte has quit IRC
[13:21:57] *** Captain has quit IRC
[13:22:38] *** Captain has joined #postfix
[13:22:44] <jonez> I saw a post on a forum somewhere that suggested changing the ip of dionysus, and then configuring the new machine to handle the 2nd hostname.
[13:23:34] <abbe> Whenever I send mail to external world (with more than 1 recipients) via my localhost:25, if my MUA sends more than 1 'RCPT TO:' string, the spamassassin never hands over those mails to sendmail. if there is only one recipient, my mail gets sent. even though it classifies all of those mails as HAM.
[13:23:58] <jonez> you are checking your outbound mail for spam?
[13:24:38] <abbe> jonez: yes, http://wiki.apache.org/spamassassin/IntegratedSpamdInPostfix
[13:24:50] <abbe> jonez: followed above HOWTO ^^^^
[13:25:17] <jonez> uhmm.. do you *need* to check your outbound mail for spam?
[13:25:24] <jonez> wouldn't it *always* be "ham"?
[13:26:46] <abbe> jonez: okay, but the problem here is not spam/ham classification, but spamassassin's behaviour. I'm wondering if it is spamassassin's problem or my postfix's.
[13:27:32] <jonez> my point is that if you will always mark your outbound messages as ham, what is the point of having spamassassin in the mix?
[13:28:27] <abbe> jonez: what if someone tries to misuse my internal mail server to relay mails. I'm not using anykind of authentication in mail-server as of now, and it is only listening on local interfaces.
[13:28:32] <jonez> personally I installed procmail and then configured it to call spamc
[13:28:47] <jonez> right.. so nobody should be able to abuse the service
[13:29:02] <jonez> unless you have some botnet nodes behind the firewall
[13:30:15] <abbe> yes, jonez . any ideas about the issue ?
[13:30:39] <jonez> turn off spamassassin for outbound email.
[13:31:30] <abbe> jonez: okay, any ideas why it allows relaying mail, if there is only one recipient.
[13:32:02] <jonez> hold on... "allows relaying mail"?
[13:32:31] <jonez> no, I do not have an idea as to what exactly is wrong, but if your goal is to stop the pain, just turn off sa for now.
[13:37:03] <jonez> "Do not forget also to set spamassassin_destination_recipient_limit = 1 in main.cf . spamc doesn't expect to parse multiple recipients at once."
[13:37:11] <jonez> abbe, did you catch that line?
[13:37:28] *** sophokles has quit IRC
[13:37:29] <abbe> jonez: oops, sorry I didn't look at that line, thanks :)
[13:37:59] <jonez> if that fixes your issue, please get that document fixed so it is more obvious
[13:38:27] <abbe> jonez: BtW, if my SMTP transaction goes like this, EHLO box, MAIL FROM:<...>, RCPT TO:<recipient1@...>, RCPT TO:<recipient2@...>, DATA, QUIT. Is this okay to send mail to recipient1 and recipient2 ?
[13:38:48] <jonez> uhmm... what are you trying to do?
[13:39:21] <jonez> it *looks* right to me, but I have not hand coded smtp in a loooong time
[13:39:23] <abbe> jonez: Gnus (my MUA) does SMTP transaction in above way, when I send mail to 2 recipients.
[13:41:04] *** sophokles has joined #postfix
[13:42:38] <jonez> abbe, test it
[13:42:43] <jonez> after you make that change to main.cf
[13:42:50] <jonez> if it still fails to work, we can look into it
[13:43:14] *** Zerial has joined #postfix
[13:43:42] <Zerial> hi
[13:43:47] <BuenGenio> hello
[13:44:02] <BuenGenio> does anyone have any suggestions for getting mail through to Yahoo mail?
[13:44:12] <BuenGenio> it consistently gets defered
[13:45:00] <Zerial> I try to  configure "mail per second" for some domains using transport. I have this lines on main.cf: default_initial_destination_concurrency = 2, default_destination_recipient_limit = 10, default_destination_rate_delay = 5s
[13:45:39] <Zerial> changing "default" for my transport name
[13:45:50] <abbe> jonez: that works, thanks :)
[13:48:40] <Zerial> With this config postfix will deliver "10 messages per second" ?
[13:49:20] <Zerial> 10 mails each 5 seconds
[13:49:46] <Zerial> ?
[13:50:03] <abbe> jonez: my postfix is not directly on the internet, it relays all mails via a smarthost, and receives all incoming mails via fetchmail
[13:51:22] <abbe> so I think I won't be able to turn off filtering for outgoing mails if my filter is invoked by 'smtp' (from master.cf).
[13:55:18] <jonez> abbe, ok... so that "fix" worked.. please contact the authors of that howto you followed and ask them to make that line a bit more prominent, since it is critical to correct operation.
[13:56:51] <abbe> jonez: okay
[13:59:25] *** a_run has joined #postfix
[13:59:26] <jonez> ty
[13:59:58] <a_run> how do i rewrite the name part of a recepient address
[14:00:24] <a_run> now its 'user1 <user at domain dot com>'
[14:00:35] <a_run> i want rewite the user1 to user2
[14:02:45] <a_run> s/recepient/sender/
[14:03:39] <abbe> a_run: checkout generic(5)
[14:04:06] <abbe> a_run: and smtp_generic_maps
[14:07:20] *** cpm has quit IRC
[14:11:39] *** cpm has joined #postfix
[14:17:43] *** Xjs has quit IRC
[14:24:48] <a_run> abbe: that didn't help
[14:25:03] <adaptr> !smtp_generic_maps
[14:25:04] <knoba> adaptr: Error: "smtp_generic_maps" is not a valid command.
[14:25:11] <adaptr> !generic
[14:25:11] <knoba> adaptr: "generic" : generic(5) table specifies an address mapping that applies when mail is delivered. This is the opposite of canonical(5) mapping, which applies when mail is received. See http://www.postfix.org/generic.5.html
[14:26:37] <a_run> i have mail rewriting working fine
[14:26:43] *** saurabhb has quit IRC
[14:26:53] <jonez> ok.. I think I know how to describe what I want to do with all this queued email.. I need to somehow tell postfix to re-evaluate the .forward file
[14:26:58] <a_run> i just want to rename the unix login
[14:27:49] <jonez> a_run, maybe make an alias?
[14:27:52] <adaptr> jonez: once mail is queued, no changes will be made to the delivery route
[14:28:10] <jonez> I've had to do this before, though.
[14:28:16] <jonez> and it worked.
[14:28:22] <adaptr> then do it again
[14:28:26] <Dominian> jonez: postsuper -r ALL
[14:28:26] <jonez> but I don't remember the steps
[14:28:43] <adaptr> also, .forward has nothing to do with postfix - it is evaluated by the delivery agent
[14:28:44] <Dominian> that will requeue anything in the queue and update "routing" information etc... or at least it should
[14:29:09] <Dominian> but as adaptr pointed out.. the .forward file won't matter to postfix.
[14:29:33] <adaptr> if the mail is still queued, it hasn't reached .forward yet
[14:29:55] <adaptr> but if it SHOULD reach .forward, then it wouldn't be queued - it would be delivered
[14:30:06] <Dominian> aye
[14:30:08] <adaptr> re-queueing mail can not alter the delivery address, for obvious reasons
[14:30:15] <adaptr> it can only alter the route
[14:30:26] <jonez> well, it's in the queue because there is "no route to host"
[14:30:40] <adaptr> then your setup is a bit more involved
[14:30:54] <Dominian> jonez: what's the domain?.. does the domain being delivered to have an MX entry?
[14:30:57] <adaptr> create a new transport that rewrites your addresses to what you want, and then requeue
[14:31:09] <jonez> zoidtechnologies
[14:31:10] <jonez> .com
[14:31:40] <Dominian> er..
[14:31:40] <jonez> yes, there is an mx, but it points to a slab from dataslab.com, which then does a lookup in a db for the delivery address
[14:32:15] <jonez> what I am doing, I suppose, is "intranet" type stuff, and a lot of forwarding
[14:32:33] <jonez> hmm.. I could simplify this set up a lot if I used openvpn
[14:32:34] <Dominian> Well, the MX servers for that domain respond
[14:32:38] <Dominian> So something is wrong with your setup
[14:33:01] <jonez> it was working fine until I had to decom dionysus and change the .forward to cyclops.
[14:33:23] <jonez> but if I set up a vpn, then the slab could get to my workstation "directly" instead of forwarding things around.
[14:33:53] <adaptr> you do realise that we have no idea what you're talking about ?
[14:34:23] <jonez> just thinking out loud about a way to simplify my email routing quite a bit
[14:34:36] <jonez> that does not solve the immediate problem, though.
[14:35:01] *** flart has quit IRC
[14:35:11] <jonez> the queue is flushing! w00t!
[14:35:21] <adaptr> quick! man the toilets!
[14:35:22] *** flart has joined #postfix
[14:37:02] *** dmn_ has joined #postfix
[14:38:50] <jonez> er...
[14:38:56] <jonez> I spoke too soon
[14:40:57] *** stas has quit IRC
[14:42:09] *** marekpe has joined #postfix
[14:42:42] *** a_run has quit IRC
[14:43:17] *** sypher has quit IRC
[14:44:25] *** TheInfinity has joined #postfix
[14:47:32] *** gutocarvalho has joined #postfix
[14:51:14] *** sypher has joined #postfix
[14:51:15] *** cpm has quit IRC
[14:57:15] *** stas has joined #postfix
[15:02:00] *** jangell has joined #postfix
[15:02:35] *** XPCFan has quit IRC
[15:08:34] *** bhagat_ has quit IRC
[15:10:00] <TheInfinity> hmm. am i also allowed to ask questions about mysterious problems with mails sent from postfix which lets spamassassign at target host give back RCVD_ILLEGAL_IP error?
[15:10:20] *** zokidimovski has joined #postfix
[15:10:39] <f3ew> #spamassassin
[15:10:46] <f3ew> what does the rule check for?
[15:11:21] <TheInfinity> http://wiki.apache.org/spamassassin/Rules/RCVD_ILLEGAL_IP
[15:12:21] *** Jense has quit IRC
[15:12:51] <TheInfinity> http://paste.pocoo.org/show/96385/ - heres a mail header (first is the one with the error, 2nd the one from my other mta)
[15:14:23] <zokidimovski> hi everybody
[15:14:42] <zokidimovski> I'm trying to setup postfix with openxchange
[15:15:06] <zokidimovski> I'm following a tutorial but the tutorial is with vmail
[15:15:15] <zokidimovski> and I'm quite new with the mail stuff
[15:16:15] <zokidimovski> I'm stuck with /etc/postfix/mysql_virtual*
[15:16:35] <zokidimovski> can anybody tell me more how actually these files work
[15:16:36] <zokidimovski> ??
[15:17:00] *** cpm has joined #postfix
[15:17:25] <f3ew> zokidimovski, postfix basically connects to the DB, runs the query specified with appropriate string substitution as shown in mysql_table(5)
[15:19:55] <zokidimovski> yes
[15:20:53] *** Zerberus has quit IRC
[15:21:09] *** Zerberus has joined #postfix
[15:21:46] <zokidimovski> for example
[15:21:49] <zokidimovski> in mysql_virtual_alias_maps.cf
[15:21:52] <zokidimovski> i have this
[15:22:38] <zokidimovski> user = postfix password = passoword hosts = localhost dbname = postfix table = alias select_field = goto where_field = address
[15:22:45] <zokidimovski> how would the output query look like
[15:22:47] <zokidimovski> ?
[15:23:06] *** XPCFan has joined #postfix
[15:23:37] *** deface has quit IRC
[15:24:41] <zokidimovski> SELECT goto FROM alias WHERE address equals what?
[15:24:57] <zokidimovski> that is what I'm confused about
[15:26:21] *** dmn_ has left #postfix
[15:26:53] <f3ew> the recipient address
[15:29:58] *** deface has joined #postfix
[15:46:49] *** Xjs has joined #postfix
[15:52:38] *** Knoedel2 has joined #postfix
[16:03:02] *** SARGuy1 has left #postfix
[16:10:32] *** SARGuy1 has joined #postfix
[16:19:28] *** XPCFan has quit IRC
[16:19:31] *** zokidimovski has quit IRC
[16:19:45] *** marekpe has quit IRC
[16:39:45] *** XPCFan has joined #postfix
[16:41:38] *** sepski has joined #postfix
[16:46:49] *** xpoint has joined #postfix
[16:51:35] *** jense has joined #postfix
[16:55:00] *** UQlev has joined #postfix
[16:56:39] *** ckoehler has left #postfix
[17:12:39] *** abbe has quit IRC
[17:15:11] *** BuenGenio has quit IRC
[17:21:02] *** karrotx has joined #postfix
[17:21:32] *** madrescher has joined #postfix
[17:28:54] *** madrescher has quit IRC
[17:29:15] *** jangell has quit IRC
[17:35:42] *** dragonheart has quit IRC
[17:38:59] *** Zblakany has quit IRC
[17:40:22] *** deadpigeon has joined #postfix
[17:40:40] *** TheInfinity has quit IRC
[17:41:06] *** spiderbatdad has joined #postfix
[17:43:12] <spiderbatdad> having a couple of problems: postfix smtpd fatal: bad string length 0 < 1: etc/postfix/mysql_mailbox.cf_dbname = ; and imapd-ssl drops connection immediately after successful authentication
[17:45:46] *** sophokles has quit IRC
[17:51:06] <f3ew> use the full path to mysql_mailbox.cf in main.cf, not a relative path
[17:51:12] <f3ew> You forgot the / before etc
[17:52:05] <jelly> bad string length sounds like an empty variable setting somewhere, "foo ="
[17:52:42] <jelly> and postfix being ungraceful about it, as it was in '04
[17:53:15] <f3ew> no
[17:53:27] <f3ew> See etc/postfix/mysql_mailbox.cf <=== missing /
[17:54:06] <spiderbatdad> what authrc tells mysql where to look for /etc/postfix/mysql_mailbox.cf?
[17:54:33] <spiderbatdad> the fields in the file are complete
[17:54:37] <f3ew> that's in main.cf
[17:54:46] <f3ew> You are missing a /
[17:54:55] <f3ew> Search for mysql_mailbox
[17:57:02] <spiderbatdad> kk
[17:59:39] <spiderbatdad> virtual_mailbox_maps = mysql:etc/postfix/mysql_mailbox.cf
[17:59:42] <spiderbatdad> :)
[17:59:55] <spiderbatdad> f3ew ty
[18:00:10] <f3ew> That's a legacy config issue
[18:00:38] *** kiliko has quit IRC
[18:00:59] *** jangell has joined #postfix
[18:01:47] <spiderbatdad> handy though...it forces new users like myself to go through the config files over and over...
[18:08:23] *** sypher has quit IRC
[18:16:31] *** jimpop has joined #postfix
[18:24:40] <siamba> anyone running ppolicy?
[18:33:29] *** F6F has joined #postfix
[18:35:45] *** UQlev has quit IRC
[18:42:19] <spiderbatdad> my other problems seems to be : SQL query: SELECT id, crypt, "", uid, gid, home, concat(var/spool/mail/virtual,'/',maildir),  causes fatal error, but other restarts produce: SQL query: SELECT id, crypt, "", uid, gid, home, concat('home,'/',maildir),
[18:45:23] <spiderbatdad> from the authmysqlrc: MYSQL_MAILDIR_FIELD concat(home,'/',maildir)
[18:46:18] *** brancaleone has joined #postfix
[18:52:37] *** sepski has quit IRC
[18:56:01] *** SARGuy1 has quit IRC
[18:56:21] *** SARGuy1 has joined #postfix
[18:57:31] <roe_> f3ew, adaptr and I were talking last night about a possible bug in sub-domain matching.  I thought maybe you could weigh in on
[18:57:43] *** Spec has quit IRC
[19:00:21] <roe_> I have parent_domain_matches_subdomains... UNset.  I telnet from a.domain.com to b.domain.com, with sender:root at a dot domain.com recipient: root at a dot domain.com.  b.domain.com tries to send ti back to a
[19:04:00] <roe_> the desired behavior is for b to collect all mail destined for *.domain.com
[19:04:05] *** sepski has joined #postfix
[19:10:20] <rob0> I'd need to see proof that it wasn't a PEBKAC. Then, verbose logs. Then post it to the list to be ruthlessly shot down by Wietse and Viktor.
[19:10:25] *** havvg has joined #postfix
[19:10:59] <roe_> sounds like fun
[19:11:18] <roe_> no body cavity search in there just for un?
[19:11:21] <roe_> *fun?
[19:11:30] <rob0> oh sure ... but that's just for fun
[19:11:52] <roe_> of course
[19:12:23] <roe_> ok, well I will start with logs and config for you
[19:17:22] <rob0> you could just post it to the list if you like
[19:19:09] *** githogori has quit IRC
[19:19:14] *** amrit|zzz is now known as amrit|wrk
[19:20:53] <roe_> can I edit out the amavis logs?
[19:21:08] <roe_> they make it very long
[19:21:35] *** jtaji has quit IRC
[19:22:59] <rob0> as long as you don't omit something important
[19:23:07] <rob0> (please do)
[19:23:49] *** bluethundr has joined #postfix
[19:30:20] <roe_> I *think* that is everything http://www.pastebin.ca/1291612
[19:31:59] <QuantumTim> hi all
[19:32:13] <QuantumTim> in the man page (and on the website) for the virtual alias file
[19:32:18] <QuantumTim> it says:
[19:32:19] <QuantumTim> Redirect mail for user@site to address when site is equal to $myorigin, when site is listed in $mydestination, or when it is listed in  $inet_interfaces or $proxy_interfaces.
[19:32:56] <QuantumTim> surely the $mydestination bit should be $virtual_mailbox_domains or something
[19:33:18] <rob0> why, Tim?
[19:33:25] <roe_> not surely at all
[19:33:26] <rob0> roe, looking
[19:33:39] <QuantumTim> for example, my virtual alias file has mappings like: alias user1, user2
[19:33:51] <QuantumTim> my host accepts mail for a.example.com and b.example.com
[19:34:07] <QuantumTim> but a.example.com is $myorigin
[19:34:23] <QuantumTim> and so alias at b dot example.com isn't aliased
[19:34:31] *** SARGuy1 has quit IRC
[19:34:39] <rob0> Best practice: don't EVER use unqualified "user" addresses (without domain).
[19:35:15] <rob0> then you always know where your mail will be routed (assuming you have a clue, that is ;) )
[19:35:20] <QuantumTim> but the two domains are just aliases for each other
[19:35:41] <QuantumTim> it's not like there are different users on each
[19:35:52] <rob0> If you do NOT want domain namespace separation, mydestination is for you.
[19:36:11] <rob0> How about this ... you step back and describe the problem and the goal?
[19:36:16] <QuantumTim> but I need virtual, not local delivery
[19:36:27] <QuantumTim> mail users don't have unix accounts
[19:37:46] <roe_> virtual to virtual domain aliasing is best done by a perl script to alias user at a-virtual dot com to user at b-virtual dot com.  At least that is my opinion
[19:38:07] <roe_> so you dont' have to manually create the file
[19:38:09] <QuantumTim> I have the whole thing working, I just noticed that my aliases don't work when I send to the second domain
[19:38:10] <rob0> yup, duplicate all the maps for wach domain.
[19:38:17] <rob0> *each
[19:38:31] <rob0> and: 18:34 < rob0> Best practice: don't EVER use unqualified "user" addresses (without domain).
[19:38:35] <QuantumTim> it isn't really much of a problem, as I doubt anyone uses the second domain for sending
[19:38:36] *** SARGuy1 has joined #postfix
[19:38:48] *** SARGuy1 has left #postfix
[19:38:54] <QuantumTim> it just seemed odd that the virtual alias file had anything to do with $mydestination
[19:40:02] <QuantumTim> okay, well I guess I'll just add all the names once for each domain
[19:40:42] <rob0> That was just telling you what happens with unqualified addresses.
[19:40:48] <QuantumTim> but as I say, the domains entirely coincide with each other, there's no difference between them
[19:42:24] <QuantumTim> yes, but hypothetically, if I have local.example.com in $mydestination, and virtual.example.com in $virtual_mailbox_domains, unqualified lines in the virtual alias file will never be matched against the local domain
[19:42:24] <rob0> roe, my postconf -d parent_domain_matches_subdomains doesn't list mydestination
[19:42:41] <roe_> nor does mine
[19:42:58] <roe_> I had it in there at one point during the TS-ing
[19:43:06] <rob0> so maybe the ".example.com" syntax isn't supposed to work in mydestination
[19:43:12] <rob0> just a guess
[19:43:24] <rob0> maybe it's a documentation bug
[19:43:32] <roe_> to the list?
[19:44:05] <rob0> Right, but don't use the "B" word, just ask for explanation
[19:44:08] <rob0> :)
[19:44:21] *** Xjs has quit IRC
[19:44:36] <roe_> B? you mean broccoli?
[19:44:49] <rob0> Well, don't mention that either.
[19:45:03] <rob0> Cauliflower, sure.
[19:45:04] <roe_> do you know if it is highly trafficed?
[19:45:12] <roe_> trafficked?
[19:45:12] <rob0> the list is pretty busy
[19:45:24] <roe_> niether of those looks right
[19:45:27] <rob0> or do you mean Broccoli?
[19:45:34] <roe_> yup, that must be it
[19:45:48] <rob0> I just bought a couple of stalks for US$1.
[19:46:12] <roe_> nice crowns?
[19:46:21] <rob0> not bad at all.
[19:46:37] <roe_> gonna do anything exciting with them?
[19:46:49] <roe_> maybe a stir fry...
[19:47:06] <rob0> Also some smallish heads of cauliflower. We'll probably use them in a stew, but yeah, stir fry sounds good too.
[19:47:35] <roe_> does cauliflower hold up well enough for a stew?
[19:47:56] <rob0> not really, but we like it anyway :)
[19:48:05] <roe_> that's all that counts
[19:49:51] *** jeffspeff[A] has quit IRC
[19:53:34] *** wdp_ has quit IRC
[19:54:49] *** Xjs has joined #postfix
[19:56:24] *** wdp has joined #postfix
[19:59:43] *** carl- has joined #postfix
[20:03:01] *** jimpop has quit IRC
[20:03:15] *** jtaji has joined #postfix
[20:04:04] <roe_> rob0, ok, email is off, we'll see how my backside feels in a few hours
[20:05:13] <rob0> got it
[20:07:28] <roe_> even though this is OT, I just started using sieve + the sieve plugin for roundcube and tbird.  I *really* like it
[20:07:53] <roe_> it makes server side filtering easy, and therefore worthwhile
[20:10:06] *** growltiger has joined #postfix
[20:13:56] *** Severed_Head_Of_ has joined #postfix
[20:15:41] *** hark has quit IRC
[20:16:00] *** hark has joined #postfix
[20:19:23] *** Haris1 has quit IRC
[20:24:55] *** growltiger_ has quit IRC
[20:26:06] *** stas has quit IRC
[20:29:21] *** growltiger_ has joined #postfix
[20:31:27] *** growltiger has quit IRC
[20:35:17] *** Severed_Head_Of_ has quit IRC
[20:35:42] *** robink has joined #postfix
[20:35:49] <robink> Can you make Postfix defer a message?
[20:36:09] <roe_> based on what
[20:36:10] <robink> i.e. make it understand something like Deferred-delivery: ?
[20:36:18] <robink> Basically, tell it to send the mail around a certain date/time.
[20:36:49] <roe_> defer connotes failure, you want to hold the message until a certain time?
[20:36:55] <robink> roe_: Yes.
[20:37:04] <roe_> sure, based on what
[20:37:15] <robink> A header in the mail
[20:37:30] <robink> with a POSIX date as the value of the header key
[20:37:42] <robink> "Deliver around this time"
[20:37:54] <robink> It can strip the header or keep it, I don't care.
[20:38:13] <robink> Would I need to run it through an external filter?
[20:38:45] <robink> Sorry, service.
[20:39:42] <robink> s/POSIX/rfc2822/
[20:40:07] <roe_> well...
[20:40:12] <roe_> people do that all the time
[20:40:15] <roe_> !header_check
[20:40:16] <knoba> roe_: Error: "header_check" is not a valid command.
[20:40:47] <robink> header_checks?
[20:42:00] <robink> !header_checks
[20:42:01] <knoba> robink: "header_checks" : a configuration parameter in the main.cf: Optional lookup tables for content inspection of primary non-MIME message headers, as specified in the header_checks(5) manual page.
[20:42:09] <robink> That?
[20:42:54] <robink> I'd need to do a regex that would do a GEQ eval, plus be able to do a check on held messages every so often.
[20:43:46] <robink> Sorry, a GEQ eval on a date.  Since Postfix supports PCRE expressions, this is probably possible (I'm quite new to regexes), but would require an internal date comparison or a conversion of the date into something akin to UNIX time.
[20:44:20] <robink> Postfix would also need to recheck held messages every so often to see if the Deliver-later: header no longer had a date in the future.
[20:45:36] *** Fallenou has joined #postfix
[20:46:43] <robink> I'm not seeing anything in pcre(3), pcrepattern(3) or pcresyntax(3).
[20:48:13] *** daguz has quit IRC
[20:49:03] <roe_> may I ask why you are delaying mail for a certain time period?
[20:49:22] <robink> So I can send a message to someone and have them receive it on the date I specify, rather than immediately.
[20:49:24] <roe_> most people delay blast emails and such until late at night and then flush the entire queue
[20:49:49] <robink> I never force a queue flush (to my knowledge).
[20:51:21] <robink> I was thinking I could write someone a mail that would get delivered on Christmas.
[20:52:34] <robink> I could just send the mail on Christmas or try to set up my MUA to push it to the mailserver on Christmas.
[20:54:29] <robink> This isn't a blast e-mail, I just want the recipient to not get it until Christmas.
[20:55:15] <roe_> that behavior might be better done on the client
[20:55:21] <robink> roe_: OK
[20:55:39] <roe_> but hang around and see if anyone else weighs in on it
[20:56:28] *** F|oFF has quit IRC
[20:57:32] *** githogori has joined #postfix
[20:57:48] *** F|oFF has joined #postfix
[20:58:07] <Dominian> eh
[20:58:11] <Dominian> I would do it on the client side
[20:58:18] <Dominian> pointless to do that just for one email to one person..
[20:58:27] <Dominian> and the work involved.. afaik.. would be more trouble than what its worth
[21:00:09] <rob0> roe, it's ugly, but you could do a pcre: map for mydestination
[21:00:25] <Dominian> rob0: shush
[21:00:27] <Dominian> :P
[21:01:55] * cpm uses maps = rob0:/
[21:03:09] *** SARGuy1 has joined #postfix
[21:07:18] *** UQlev has joined #postfix
[21:07:50] <roe_> to many ro* in here
[21:09:08] <roe_> Weitze punted me for round 1, going in for round 2
[21:09:34] *** SARGuy1 has quit IRC
[21:10:01] <rob0> That wasn't a punt, is was an end-run.
[21:10:13] <roe_> my mistake
[21:11:28] *** SARGuy has joined #postfix
[21:14:10] *** carl- has quit IRC
[21:15:22] <roe_> is what I want wrong?  I know how often one can lose the forest for the trees
[21:16:47] *** VaNNi has quit IRC
[21:16:47] *** Severed_Head_Of_ has joined #postfix
[21:16:47] *** growltiger_ has quit IRC
[21:20:42] * cpm punts rob0
[21:21:01] *** VaNNi has joined #postfix
[21:21:59] *** wdp has quit IRC
[21:22:47] *** Nothing__ has quit IRC
[21:24:11] <jeev> lol punt
[21:24:39] <jeev> i used to punt people like crazy
[21:28:27] <jonez> is there anything in addition to "postsuper -r ALL" I should run if I need to have postfix re-evaluate .forward files for a few messages stuck in the queue?
[21:28:38] <jonez> I've tried it, and it does not seem to help.
[21:35:44] *** vertigo_ has quit IRC
[21:37:07] *** Jacolyte_ has joined #postfix
[21:38:51] *** luke-jr has quit IRC
[21:39:02] *** luke-jr has joined #postfix
[21:42:12] *** growltiger_ has joined #postfix
[21:42:25] *** Severed_Head_Of_ has quit IRC
[21:43:54] <rob0> Wow, Viktor suggested the same thing I did.
[21:43:55] *** SARGuy has quit IRC
[21:44:33] <rob0> jonez, maybe postfix flush?
[21:45:12] <Dominian> did he?
[21:45:13] *** XPCFan has quit IRC
[21:45:19] <Dominian> I missed the mailing ont he list.. either that or i ignored it
[21:45:48] <jonez> hmm... I'll try that again, but not sure if it will work
[21:45:49] <jonez> hold
[21:45:55] *** cpm has quit IRC
[21:45:55] *** SARGuy has joined #postfix
[21:46:16] <Jacolyte_> hay guise, what do I have to do to get fancy subdomains liek smtp.domain.com and imap.domain.com
[21:46:28] <Dominian> DNS
[21:46:42] <Dominian> Those are done at your DNS provider
[21:50:05] <Jacolyte_> I'm assuming that would be godaddy
[21:50:30] <Jacolyte_> which is the company I manage my domains through
[21:51:21] <Dominian> Well, then there's your answer.
[21:51:55] <Jacolyte_> but, those aren't required right?
[21:52:00] <Jacolyte_> smtp and imap subdomains
[21:52:30] <Dominian> nope
[21:52:40] <Dominian> I use my MX record for both
[21:52:45] <Dominian> but you can very well use whatever you want..
[21:52:54] <Jacolyte_> hmm..
[21:53:36] <rob0> You might want to read up on some DNS basics, it's pretty important in mail management.
[21:53:36] *** hparker has joined #postfix
[21:53:57] <Jacolyte_> yeah.. I guess I should X_X
[21:54:10] <Jacolyte_> Maybe that's why I'm completely lost when reading the documentation
[21:54:13] <Jacolyte_> lulz
[21:54:29] <rob0> Perhaps wikipedia?
[21:54:48] *** XPCFan has joined #postfix
[21:54:55] *** hark has quit IRC
[21:55:00] <rob0> I don't know, I just thrashed around aimlessly for a few years until it started making sense. :)
[21:55:20] <jonez> if delivery has been "temporarily suspended" do I need to mark the messages somehow so they will get redelivered?
[21:55:25] *** hark has joined #postfix
[21:55:27] <Dominian> jonez: no
[21:55:46] <rob0> qmgr(8) will do that for you
[21:55:55] <jonez> I swear I had this same problem before (one machine being replaced totally by another) and it was a simple solution
[21:56:26] <rob0> I swear you haven't described your real problem at all ... just a few random questions.
[21:56:34] <jonez> ok. forgive me.
[21:56:41] <rob0> too much swearing in here!!
[21:56:47] <jonez> my mail setup is a little less than straight forward
[21:57:56] <jonez> the domain is zoidtechnologies.com. mail gets delivered to my router, called eros (so jam at eros) dot . on that machine I have a .forward file that previously pointed to jam@dionysus. dionysus has been replaced with cyclops. I am attempting to get the email queued for jam@dionysus requeued to jam@cyclops
[21:59:20] *** growltiger has joined #postfix
[21:59:23] <jonez> I've tried running 'postfix flush' and 'postsuper -r ALL' a couple of times, but that does not help. not sure if the messages need to be "unstuck" by taking them off "hold" status, or what to try next
[21:59:33] *** rcsu has joined #postfix
[22:02:18] <rob0> why are they on hold?
[22:02:24] <rob0> man postsuper
[22:02:37] <Dominian> wtf.. hold status?
[22:02:41] <Dominian> They are in the hold queue?
[22:02:50] <Dominian> jonez: did you by chance at one point run MailScanner?
[22:03:09] <jonez> I do not think they are on hold.. I think they are deferred
[22:03:18] <jonez> "mailscanner"? no. never heard of it
[22:04:28] <rob0> oh, "hold" refers to one of the queues ... I thought you meant it
[22:04:29] *** hark has quit IRC
[22:04:51] *** xpoint has quit IRC
[22:05:10] <rob0> Mail that was queued with a .forward is not requeued when .forward changes.
[22:05:26] <rob0> so there's your problem, a misunderstanding
[22:05:27] <jonez> understandable :)
[22:05:28] *** havvg has quit IRC
[22:05:41] *** Severed_Head_Of_ has joined #postfix
[22:05:44] <jonez> I've had to deal with this situation in the past, but I do not remember what the solution was
[22:05:49] <rob0> postcat(1) can recover the messages
[22:05:53] *** githogori has quit IRC
[22:05:53] *** growltiger has quit IRC
[22:05:59] <rob0> or, use a transport(5)
[22:05:59] *** hark has joined #postfix
[22:06:07] <rob0> (transport_maps)
[22:06:25] <jonez> so basically configure cyclops to receive mail for dionysus?
[22:06:42] <rob0> another possibility, set relayhost=[the.one.you.want.to.deliver.to] and requeue
[22:07:14] <rob0> it would indeed have to be set up to accept the mail for domain[s] you want
[22:11:25] *** growltiger_ has quit IRC
[22:15:20] <jonez> well, it's the same domain but a different hostname
[22:16:01] *** Spec has joined #postfix
[22:16:25] *** hark has quit IRC
[22:16:29] *** hark has joined #postfix
[22:19:43] *** cryptnix has joined #postfix
[22:19:46] <cryptnix> hello.
[22:23:37] *** diazepam has joined #postfix
[22:26:35] <jangell> How do you guys handle the negative recipient address cache with virtual users/  domains?  Expire the entire cache every time a user is added?
[22:26:46] <Dominian> huh?
[22:26:48] *** luke-jr has quit IRC
[22:27:26] <jangell> Dominian: address_verify_negative_cache
[22:27:55] <jangell> Dominian: For example.  Lets say an incoming e-mail has a to of:  jesse at jesseangell dot com  if that does not exist the server caches it as not existing
[22:28:13] <jangell> Dominian: if I add that user to my LDAP structure the server doesn't think that the user exists until I run postfix reload because of that cache
[22:28:31] <jangell> Dominian: There is also an address_positive cache
[22:28:39] <Dominian> Hrm.. I just add the user
[22:28:40] * Dominian shrugs
[22:28:47] <Dominian> i never reload postfix unless I see necessary
[22:28:55] <jangell> Dominian: do you have address_verify_negative_cache set to yes?
[22:29:01] <Dominian> its defaulted to yes
[22:29:07] <Dominian> and I don't even use that parameter in my main.cf
[22:29:24] <jangell> Dominian: Okay.  So if you try to send an e-mail to that user before you add it..and then add that user..the server will bounce messages for that user until the cache expires.
[22:29:46] <Dominian> well postconf -n does not return that parameter
[22:29:50] <Dominian> so I must not be using it
[22:30:01] <jangell> Dominian: don't use -n
[22:30:21] <jangell> Dominian: postconf |grep address_verify_negative_cache  returns address_verify_negative_cache = yes for me
[22:30:29] <Dominian> yeah its defaulted to yes
[22:30:30] <rob0> That's a pretty rare problem, no?
[22:30:41] <Dominian> then again I use mysql for my back end authentication...
[22:30:46] <jangell> rob0: probably.  but I'm trying to think of a way to avoid it...
[22:30:51] <jangell> Dominian: it would happen with MySQL as well
[22:31:01] <Dominian> I've not seen what you're seeing though is my point.
[22:31:14] <jangell> Dominian: Test it on your server and it'll happen.  i can re-produce ito ver and over
[22:31:18] <Dominian> and if I did run into that issue.. I'd "postfix reload" and be done with it
[22:31:29] <Dominian> or disable cachine
[22:31:42] <rob0> Sure, grep the logs for a rejection to a user when you add that user.
[22:31:52] <jangell> Dominian: Well..the problem for me is that I'll have an entire support department and thousands of clients adding their own accounts.  I can't expect them to just figure this out on their own
[22:31:53] <rob0> or, query the cache directly with postmap
[22:32:08] <jangell> rob0: is there a way I can expire the cache for a specific e-mail address?
[22:32:18] <rob0> I doubt it
[22:32:21] <jangell> rob0: I'm writing a webapp layer that handles all of this..and web services for our products
[22:32:39] <jangell> So I'm basically forced with expiring the entire cache?  I'm afraid the performance hit from that could really hurt
[22:32:58] <jangell> it is my understanding that postfix sends a mesage through it's queue up to the LDA for the test
[22:33:14] <rob0> for what test?
[22:33:22] <cryptnix> Guys, i'm using postfix as a "catchall scan all ..." i want to try to stop dictionary attacks .. what is the preferred method ... and any "mail storms" ... looking for approaches ...
[22:34:03] <rob0> Don't use a catchall.
[22:34:24] <roe_> I never saw the attractiveness of a catchall
[22:35:02] <rob0> I thought I wanted one when I first started out. Changed my mind at the first dictionary attack.
[22:35:46] <roe_> I find people mistype the domain as much as they mistype a user name
[22:35:47] <jangell> rob0: address verification.  It doesn't just take the ldap result and then accept the message.  It sends a test message through its queue  http://www.postfix.org/ADDRESS_VERIFICATION_README.html
[22:36:30] <rob0> verify(8)
[22:36:38] <rob0> !verify
[22:36:39] <knoba> rob0: "verify" : Sender or recipient address verification features: http://www.postfix.org/ADDRESS_VERIFICATION_README.html
[22:37:08] <rob0> it doesn't queue a message, it stops after MAIL FROM
[22:37:53] <jangell> rob0: would I take much of a hit by just turning that cache off?
[22:38:02] <roe_> isn't it usually beneficial to at least get the rcpt to:?
[22:38:18] <jangell> rob0: well..it does get to the rcpt to
[22:38:34] <cryptnix> i'm talking about ... as a mail filter gateway .
[22:38:54] <cryptnix> basically the scanning mx for virus', spammy, rbl checking...
[22:38:54] <jangell> rob0: "A sender or recipient address is verified by probing the nearest MTA for that address, without actually delivering mail. The nearest MTA could be Postfix itself, or it could be a remote MTA (SMTP interruptus). Probe messages are like normal mail, except that they are never delivered, deferred or bounced; probe messages are always discarded. "
[22:39:19] *** dergringo has joined #postfix
[22:41:23] <dergringo> Hi. My mailserver based upon postfix and dovecot is up and running. I make use of virtual users and virtual domains (mysql). Now I am not able to send mails using mailx becuase it does not find user "root at example dot com". I just want to deliver it locally
[22:42:24] *** Zerial has quit IRC
[22:42:55] *** rob0 has quit IRC
[22:43:15] *** Dominian has quit IRC
[22:44:39] *** kale has joined #postfix
[22:46:14] <kale> i need to run postfix, so it delivers to another postfix over ssh. any guides or hints for that
[22:47:32] <jangell> Hmm..is there any way I could expire the verify cache using postkick?
[22:47:41] <UQlev> kale: why smtp over ssh? isn't tls working?
[22:47:41] <jangell> to force an update for a specific address?
[22:48:10] *** XPCFan has quit IRC
[22:48:31] <kale> UQlev: if, i'm on the road, i may not be able to send over smtp. ISP blocks it in some places
[22:48:45] *** XPCFan has joined #postfix
[22:49:00] <UQlev> kale: use submission port 587
[22:49:53] <kale> what is a submission port?
[22:50:19] <UQlev> kale: it is same smtp port but 587 i/o 25
[22:50:45] <UQlev> kale: usually with compulsory smtp-auth
[22:50:52] <kale> UQlev: ok, let me explain further
[22:52:22] <kale> i'm using mutt, and it wants a mailserver on the system it runs on. so i want to install postfix on my laptop, to be able to send mails from mutt. if i'm then not home, should i then open port 587 for postfix, so i can connect from the outside and relay mail this way?
[22:54:00] <UQlev> kale: it is quite often standard mail server listening 25 port for normal inward mails and 587 port for your worldwide customers to relay their mail
[22:54:32] <kale> UQlev: ok, thats great info
[22:54:48] <UQlev> 587 port called submission port
[22:55:49] <kale> now i'm just wondering if ISP is blocking that port too. one of the danish ISP's block all smtp traffic. you need to use them as relay, and to receive mail you need to set your system up as backup MX. They may have done something similar to the submission port
[22:56:13] <UQlev> kale: why dont you use on the notebook mail-client sending directly to your smtp-server?
[22:56:18] <kale> this way they get rid of spammers
[22:56:27] *** pirho has quit IRC
[22:56:38] <kale> UQlev: isp blocks that
[22:57:16] <kale> UQlev: depending on what ISP you are connected through, they have different mail "labyrinths"
[22:57:16] *** diazepam has quit IRC
[22:57:19] <UQlev> kale: very common to block 25 port because of zombies, they should not block 587 because it is useless for spammers
[22:57:51] <kale> UQlev: ok, so i should set up postfix to relay to mydomain port 587
[22:58:06] <UQlev> yes
[22:58:07] <kale> next quistion
[22:58:32] <kale> how do i make certain that it is really me that is trying to get something relayed
[22:58:51] <UQlev> kale: authenticate
[22:59:16] <kale> UQlev: is that done with keys?
[22:59:22] <UQlev> submission ports needs compulsory authentication
[22:59:37] <kale> UQlev: ah, ok, that makes sense
[22:59:47] *** Haris________ has joined #postfix
[22:59:52] <UQlev> kale: login or certs
[23:00:19] <UQlev> kale: most use user/password via tls
[23:00:27] <kale> UQlev: login? using pam?
[23:00:51] <UQlev> depends on virtual or system accounts
[23:01:21] <UQlev> sasl-auth
[23:01:21] <kale> UQlev: so i could use my virtual mailhosting system to authenticate from
[23:01:49] *** nonsequitir has joined #postfix
[23:02:32] <UQlev> kale: your postfix can work as auth server and auth client
[23:03:00] <kale> UQlev: that seems to be the correct solution
[23:03:08] <kale> makes sense
[23:03:24] <kale> thats for the theory, next the implementation ...
[23:03:38] <UQlev> kale: it works for 99% of servers
[23:04:29] <kale> i'll take your word for that
[23:04:33] <UQlev> kale: because majority ISPs block 25 port
[23:05:07] <kale> i'll have to read up on this.. but i'd like to ask something else
[23:05:13] <UQlev> they only allow smtp traffic to their own smtpd
[23:05:50] <kale> UQlev: i guess they can filter a lot of spam this way, thats fair i guess, but gives us a lot of extra trouble
[23:06:12] <UQlev> not that much trouble
[23:06:23] *** rob0 has joined #postfix
[23:07:22] <kale> earlier i set up my system so i could sort mail with wildcards. so i made a rule like joe* and then everything that fit that pattern would be delivered to joe
[23:07:46] <kale> but after i set up virtual mailhosting i have not been able to get it working
[23:09:23] *** Faust-C has joined #postfix
[23:09:29] <UQlev> kale: do you use virtual aliases?
[23:09:36] *** githogori has joined #postfix
[23:11:02] <kale> UQlev: i get the from a database
[23:12:22] <UQlev> kale: regret I have not used aliases yet on postfix, wait for someone's advise
[23:12:41] <kale> UQlev: ok, i'll do that
[23:13:10] <kale> you gave my some great tips on the relaying though
[23:13:28] <UQlev> kale: that one is very basic
[23:13:54] <UQlev> kale: I am noob yet on postfix
[23:14:07] <kale> UQlev: let me put it this way: being stupid just means that you do not yet know
[23:14:20] <kale> so i'll consider myself stupid here
[23:15:45] <UQlev> kale: mail server is not intuitive staff, it needs a lot of reading and less logic ;)
[23:16:41] *** Fallenou has quit IRC
[23:18:59] *** Haris_ has quit IRC
[23:24:37] <roe_> kale, you are interesting in virtual alias maps?
[23:24:53] <roe_> !virtual_alias_map
[23:24:54] <knoba> roe_: Error: "virtual_alias_map" is not a valid command.
[23:24:56] <roe_> !virtual_alias_maps
[23:24:57] <knoba> roe_: "virtual_alias_maps" : a configuration parameter in the main.cf: Optional lookup tables that alias specific mail addresses or domains to other local or remote address. The table format and lookups are documented in virtual(5).
[23:25:15] <kale> roe_: yep those
[23:25:19] *** war9407 has quit IRC
[23:25:37] <kale> roe_: but what i need i something that matches a wildcard
[23:26:02] <roe_> I think you'd have more luck with a regexp
[23:26:18] <kale> roe_: that would be fine too
[23:26:26] <roe_> ok what's the problem
[23:26:38] <kale> earlier i set up my system so i could sort mail with wildcards. so i made a rule like joe* and then everything that fit that pattern would be delivered to joe
[23:27:03] <roe_> ok
[23:27:11] <kale> i have not been able to figure out how to do this with virtual domains
[23:27:14] <roe_> I'm assuming "earlier" means local delivery only?
[23:27:20] <kale> roe_: exactly
[23:27:30] <roe_> right and the way to do it is with virtual_alias_maps
[23:27:56] <kale> roe_: and i do that
[23:28:20] <kale> roe_: but i do not know how to match anything but a complete email-address
[23:28:24] <roe_> just to be clear what you really want is joe* at domain dot com ->joe at domain dot com
[23:28:36] <kale> roe_: yes
[23:28:40] <roe_> you have to use regexp
[23:28:53] <kale> roe_: ok, where do i place the regexp?
[23:29:02] <rob0> !virtual
[23:29:03] <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
[23:29:52] <kale> roe_: should it be in the selectfield?
[23:30:23] <roe_> selectfield?
[23:31:20] <roe_> you need something like (and I don't know regexp too well) \^joe.* at domain dot com\
[23:31:30] <roe_> joe at domain dot com
[23:32:19] <kale> roe_: ok, i'll just try adding that, hold on
[23:32:20] <roe_> so virtual_alias_maps = regexp:/path/to/file
[23:32:33] <roe_> and put that in the file
[23:32:53] <roe_> rob0, stop me if I am off base
[23:33:30] <kale> roe_: ahh, but those are stored in a postgres database
[23:36:48] <roe_> ok... I'm not sure how to write the sql for that, but it should be the same idea
[23:37:06] <kale> roe_: had to relearn that sql, here goes ...
[23:38:38] <kale> roe_: didn't work
[23:39:08] <kale> insert into virtual (id,email,destination) values(18, '/^raket.* at kallenberg dot dk/', 'raket at kallenberg dot dk');
[23:39:09] <roe_> I think postfixadmin does something like that, you can take a look that their sql
[23:40:18] <kale> postfixadmin, is that an application?
[23:40:29] <roe_> !postfixadmin
[23:40:29] <knoba> roe_: "postfixadmin" : used for managing email accounts through a web interface (http://high5.net/postfixadmin/)
[23:41:23] *** XPCFan has quit IRC
[23:41:57] *** XPCFan has joined #postfix
[23:42:59] *** hparker has quit IRC
[23:43:21] *** rob0_ has joined #postfix
[23:45:38] <kale> roe_: i cannot seem to find any valuable information on regexp and virtual domain combination
[23:46:52] *** UQlev has quit IRC
[23:47:22] *** Dominian_ has joined #postfix
[23:50:18] *** dergringo has quit IRC
[23:52:51] *** Pazzo has quit IRC
[23:54:56] *** kale has left #postfix
[23:59:50] *** xpeed has joined #postfix





top