[00:00:29] <higuita> this way, the mail dns caches the mail info, the other server cache the websites and internal machines
[00:00:59] <jcrites> makes sense
[00:01:01] <higuita> but dont even think in forward the DNS to your ISP!!
[00:01:07] <jcrites> is it hard to run a dns server?  can I do it on any machine?
[00:01:14] <jcrites> I don't have a lot of experience with sysadmin type stuff
[00:01:17] <sysmonk> higuita: why not
[00:01:26] <higuita> yes, any machine is good
[00:01:39] <sysmonk> jcrites: then hire a sysadmin! :P
[00:03:12] <higuita> sysmonk: i dont know how are the dns in your countries, but here in Portugal the 2 main ISPs have slow DNS servers and at least the bigger one have 2 (of 4) dns servers that sometime go crazy... having all the email stopped because of something you dont control isnt funny
[00:03:41] <higuita> in your dns servers, you can reboot or check for problems, in the ISP... good luck
[00:05:24] <jcrites> ya but then you have to maintain dns servers :)
[00:05:54] <sysmonk> higuita: as far as i understand, jcrites works in a company which has it's own dns servers
[00:06:08] <sysmonk> that's why i offered him to use those as forwarders
[00:06:16] <sysmonk> IF they aren't overloaded ( see above! )
[00:07:28] <higuita> yeh, i was just justifying the "dont use the ISP dns"... i have bad experience with then :)
[00:07:45] * sysmonk works for an ISP
[00:07:57] <sysmonk> so you're saying i'm giving bad services! :P
[00:09:11] <jcrites> big american isp?
[00:09:19] <higuita> sysmonk: are you from Portugal ! :)
[00:09:19] <jcrites> maybe you should help me know how to best send mail to your customers :D
[00:09:49] <jcrites> basically I want to do it as fast as possible without causing any problems for the ISPs or running into spam filtering .... how do the mail servers judge this?
[00:10:05] <higuita> after using the DNS from ISP, i learn that its better to control the DNS servers used by the email... just in case :)
[00:10:28] <sysmonk> higuita: sure
[00:10:36] <sysmonk> that is, sure about DNS, not sure about Portugal :)
[00:10:45] <sysmonk> jcrites: i'm not even in USA
[00:11:13] <jcrites> oh.... we don't send as much mail globally
[00:11:33] <jcrites> although we do some
[00:11:43] *** F6F has quit IRC
[00:12:07] <sysmonk> oh, sure, another mass-mailing-company with a sysadmin which doesn't know mail stuff verry well o_o
[00:12:20] <sysmonk> hip hip hooray for that :)
[00:12:57] <higuita> jcrites: again, dont try to turn smtp in to a IM, email have delays... most of the time its fast, but if you give the idea to your boss/co-workers that the email is fast, they will complain because the "email took 5 minutes to delivery!!"
[00:13:08] <higuita> again, personal experience!! :)
[00:13:39] *** c00l2sv has joined #postfix
[00:14:03] <jcrites> not handling any email sent by humans :)
[00:14:12] <higuita> dns timeout, greylist, too many parallel deliveries to a ISP, etc all make the smtp slower, and all things you dont control
[00:14:15] <jcrites> which is good and bad
[00:14:22] <sysmonk> heh, personal experience - if email doesn't get in 1 minute, you got a call from a customer saying that 'email takes ages to deliver'
[00:14:25] <sysmonk> :P
[00:14:28] <jcrites> what's worse is that the automated systems keep track of the time it takes much more and complain :P
[00:14:49] <jcrites> sysmonk: what's the right way to send mail as fast as possible while still being a good citizen?
[00:15:03] <sysmonk> jcrites: use plain mail
[00:15:06] <sysmonk> go to a post office
[00:15:12] <sysmonk> get a white piece of paper...
[00:15:30] <higuita> dont increase the smtp parallel delivery unless you know the other side wants to
[00:15:38] <jcrites> now that does take too long :)(  a lot of these notices are things like "your produced has been bought by a customer, now ship it to them"
[00:15:39] <jcrites> :)
[00:15:42] <higuita> LOL
[00:18:15] <higuita> finally enable the connection reuse to all the server that get more load and that accept it
[00:18:17] <higuita> smtp_connection_cache_time_limit = 5
[00:18:19] <higuita> smtp_connection_reuse_time_limit = 600s
[00:19:41] <higuita> and if you can, send 1 email to about 50 users, instead of 50 messages to 50 users (not always possible, but some mailling lists allow this)
[00:20:19] <higuita> specially if those 50 users are from the same domain (so it takes just one connection to deliver the mail to the 50 users)
[00:24:00] <jcrites> Jul 12 15:23:44 mail-app-1102 postfix/smtpd[3431]: warning: connect #10 to subsystem private/proxymap: Connection refused
[00:24:04] <jcrites> what's that mean?  :(
[00:26:05] <higuita> strange... refue to connect to the proxymap?!
[00:26:30] <higuita> maybe the global process limit?
[00:27:11] *** j_s has quit IRC
[00:27:57] <jcrites> I think I have some kind of zombie process thing going on
[00:29:19] *** sypher has joined #postfix
[00:29:46] *** sypher has quit IRC
[00:29:49] *** sypher has joined #postfix
[00:35:25] *** sypher has quit IRC
[00:35:41] *** sypher has joined #postfix
[00:43:53] *** c00l2sv has quit IRC
[00:57:25] <jcrites> hmmm well
[00:57:35] <jcrites> postfix appears to be running fine with a memory-mapped /spool/postfix
[00:57:40] <jcrites> pretty good latency too
[00:57:58] <jcrites> lot of messages are <10ms
[00:58:13] <jcrites> so I wonder
[00:58:21] <jcrites> how silly would it be to have two postfix processes on one machine :)
[00:58:32] <jcrites> one with a memory-mapped queue, the other has a disk queue.  the memory postfix delivers into disk postfix...
[00:58:57] <jcrites> this way clients receive a super-fast response profile but mails can't be lost as easily (not that it's really a concern.... these are relay machines)
[01:02:06] <jcrites> yeah
[01:02:20] <jcrites> it looks like the memory-mapped postfix is consistently better performing than the disk one for similar loads
[01:04:00] *** zeitsofas has joined #postfix
[01:05:05] *** zeitsofa has quit IRC
[01:18:11] *** m0f0x has joined #postfix
[01:33:12] <jcrites> hmm
[01:33:24] <jcrites> at the 50th percentile and below, performance is much better
[01:33:39] <jcrites> gets worse towards 99th percentile where it's twice as slow
[01:40:03] <higuita> jcrites: really, that seens stupid, instead of duplication the process of receiving, just delivery directly to the final queue
[01:40:41] <higuita> what FS are you using for /var/spool/postfix?
[01:41:15] <shasta> just move spool to a separate, high-performance storage subsystem...
[01:41:50] <higuita> yep
[01:42:21] *** preaaaaa is now known as c0mrade
[01:43:41] <jcrites> don't got one of those
[01:43:41] <higuita> if you are using ext2/3, chattr -S /var/spool/postfix or something... IIRC, by default postfix ask for sync writes to the HD, but if you accept the risk of losing emails, remove the sync and the OS will do more cache
[01:43:45] <jcrites> this is commodity virtualized hardware
[01:43:53] <jcrites> I don't even have a dedicated FS much less a fast one :)
[01:43:59] <jcrites> and the memory fs is ....
[01:44:09] <higuita> my experience, reiserfs v3 and XFS are the faster for the mail spool
[01:44:09] <jcrites> whatever you get when you say mount tmpfs
[01:44:33] <jcrites> "sudo mount -osize=100m tmpfs /mnt/tmpfs -t tmpfs"
[01:44:34] *** xnixan has quit IRC
[01:44:37] <jcrites> however that works :)
[01:45:03] <higuita> :)
[01:45:27] <jcrites> i could route directly to a certain queue
[01:45:35] <jcrites> i'm mostly just concerned with finishing the handoff quickly
[01:45:41] <higuita> i'm talking about the real FS, not the one you are using as spool 8)
[01:45:45] <jcrites> letting the client know that it's caccepted and it can continue
[01:47:37] <jcrites> I think I am running inside Xen
[01:47:52] <jcrites> right now I'm testing the smallest instance possible
[01:48:00] <jcrites> I can get real hardware too but it's much more expensive
[01:48:40] *** sypher has quit IRC
[01:48:42] <higuita> how much time are you getting using a normal spool, how much the users are waiting?
[01:49:28] <jcrites> I need to test the normal spool again.... with my existing mail server the average mail transmission time is 25 ms
[01:49:36] <jcrites> with the postfix virtual spool it's 32ms
[01:49:50] <jcrites> however, the 25th percentile of postfix is 6ms while it's 20ms with the other solution
[01:49:57] <jcrites> so the best speed is much better
[01:50:05] <jcrites> this is probably due to the virtualization
[01:50:22] *** madrescher has quit IRC
[01:51:06] <jcrites> I'll try the normal spool
[01:53:41] <higuita> the chattr -S tip is for the normal spool, by the way
[01:54:04] <jcrites> doubt it matters for the memory one eh?
[01:54:10] <jcrites> but, I heard it was necessary for correctness
[01:55:59] <jcrites> slooks like the regular spool is better so far
[01:56:04] <jcrites> than memory-mapped, for the avg
[01:56:23] <jcrites> what's interesting is that the memory-mapped spool had a mode of 6, with an average of 32
[01:56:27] <higuita> not all FS have the sync bit, and for sure, tmpfs dont 8)
[01:56:32] <jcrites> and a median of 9
[01:56:41] <jcrites> that means it's just having really high-latency spikes but is otherwise better
[01:57:48] <higuita> ocasional load spikes are normal in a virtual machine
[01:58:18] <higuita> after all, you arent the only one using the hardware
[01:58:27] <jcrites> yeah :-(  maybe I need real hardware
[01:59:19] <higuita> 32ms is nothing to a web page!!
[01:59:41] <jcrites> oh 32ms is fine
[01:59:46] <jcrites> it's just the instability I'm worried about
[01:59:53] <jcrites> maybe 1/10000 gets totally dropped or something
[02:00:14] <jcrites> part of the reason we're reworking the infrastructure is because the existing mail servers spike up to really high latencies
[02:00:46] <jcrites> it's working really well now
[02:00:53] <jcrites> yeah, I think the disk spool is better
[02:00:55] <jcrites> more stable
[02:00:57] <higuita> postfix scales very well
[02:01:01] <jcrites> postfix is probably just written to use disk
[02:01:10] <jcrites> so if I want to have a huge fleet of postfix machines....
[02:01:22] <jcrites> should I just put them all behind a load balancer and be done wiht it?
[02:01:25] <jcrites> or try to use MX records?
[02:01:55] <higuita> mx records?
[02:02:24] <higuita> isnt the php/perl sending directly to the queue? or have they a localhost postfix running?
[02:02:50] <jcrites> considering lots of options
[02:02:56] <jcrites> the one i was testing just now is local delivery
[02:03:11] <jcrites> but our system right now has an internal load balancer interface which is sent to
[02:03:34] <jcrites> the possible advantage of a load balancer or public MX is that I can tune the mail servers more
[02:03:41] <higuita> postfix in local delivery? ok...
[02:03:51] <jcrites> if every system sends to itself, well, some of those systems might be shitty / heavily loaded virtuals
[02:03:55] <jcrites> so they might get bad spiky times
[02:04:05] <jcrites> however if they were delivering to dedicated hardware they might get better times
[02:04:21] <jcrites> by local delivery I mean the machine sends to its own postfix, then that postfix relays the mail somewhere else
[02:04:40] <jcrites> with the purpose of attmepting to give hte client the lowest latency :)
[02:04:41] <higuita> you can both use the MX or a load balance... load balance might balance more the servers, but MX perfectly fine and its alot cheaper 8)
[02:04:45] <jcrites> but not sure that will work or not
[02:05:41] <higuita> yes, the lowest latencies is to send to localhost and the delivery to a remote server
[02:05:52] <jcrites> k
[02:06:41] <higuita> even if the server is loaded, local delivery (i assume you are using /usr/sbin/sendmail command) is the fastest
[02:07:00] <jcrites> local smtp actually :)
[02:07:07] <jcrites> that's how the software communicates with remote servers now
[02:07:13] <jcrites> i just changed the config to point it to localhost
[02:07:26] *** c00l2sv has joined #postfix
[02:08:51] <higuita> trying to put a perl or php to delivery remotly to a smtp server used more resourses and would hold the webpage longer
[02:09:26] *** Southron has left #Postfix
[02:09:27] <higuita> humm... it its smtp, you can delivery directly to the remote
[02:10:00] <higuita> unless your network is slow :)
[02:10:59] <higuita> but that might increase the latency a little, after all, localhost dont use any hardware, and other machine use 2 network cards
[02:11:56] <higuita> on the other hand, remote delivery will free the load from the web server faster
[02:21:20] *** pirho has quit IRC
[02:23:11] *** Zblakany has quit IRC
[02:27:16] *** chmac has joined #postfix
[02:27:49] <chmac> I'm trying to test postfix, when I smtp to localhost I get no response to my `HELO localhost`
[02:28:15] <chmac> Yet it seems to work according to checkdns.net - is there an obvious noob mistake I could be making?
[02:30:10] <chmac> It's random as heck, the server is mule.bewelcome.org if anyone wants to try it
[02:31:49] <chmac> Ok, I've just tested again via http://www.checkdns.net/quickcheck.aspx?domain=bewelcome.org&detailed=1
[02:32:50] <higuita> no, its not working
[02:33:15] *** Zblakany has joined #postfix
[02:33:17] <rob0> logs
[02:33:21] <chmac> :)
[02:33:30] <chmac> CheckDNS was checking a different server, DOH!
[02:33:32] <thumbs> logs are overrated!
[02:34:41] <higuita> yep, the hostname you gave resolves to 91.121.158.43, the checkdns resolves to 91.121.1.157 :)
[02:35:22] <rob0> Logs are for wimps who can't fix a problem report of "it don't work".
[02:35:37] <thumbs> !tell rob0 about logs
[02:35:38] <knoba> thumbs: Error: No factoid matches that key.
[02:35:43] <thumbs> awww
[02:36:26] <higuita> one of the reasons i like postfix, its give good logs!! :)
[02:36:36] <chmac> Thanks guys, it's always the simple things :)
[02:50:17] *** McJerry has joined #postfix
[03:04:32] *** m0f0x has quit IRC
[03:04:55] <chmac> If I add relayhost = blah.domain.com - postfix will still try to deliver mail locally right?
[03:05:13] <chmac> I'd like postfix to forward all incoming messages (for our domains) to another server without any internal processing at all
[03:05:33] <chmac> We've moved mail from one host to another, and I want to forward anything that's sent to the old server while the dns propagates
[03:06:53] *** markl__ has quit IRC
[03:24:24] *** frato is now known as Gast24085
[03:25:29] *** Gast24085 is now known as frato
[03:31:21] *** Zblakany has quit IRC
[03:33:16] *** cilly has joined #postfix
[03:43:23] *** icewaterman has quit IRC
[03:43:36] *** icewaterman has joined #postfix
[03:51:51] *** m0f0x has joined #postfix
[03:58:16] *** higuita has quit IRC
[03:58:53] *** higuita has joined #postfix
[04:00:31] *** neoeinstein has quit IRC
[04:21:31] *** neoeinstein has joined #postfix
[04:27:47] *** arahman_ has joined #postfix
[04:30:18] *** arahman has quit IRC
[04:48:23] *** [raz] has joined #postfix
[05:00:03] *** makerc has joined #postfix
[05:02:17] *** raz has quit IRC
[05:02:17] *** [raz] is now known as raz
[05:03:13] *** githogori has quit IRC
[05:08:02] *** makerc has quit IRC
[05:14:14] *** Zeit|awy_ has quit IRC
[05:23:39] *** DaveH|Work has joined #postfix
[05:26:47] *** Derander has joined #postfix
[05:28:11] *** Motoko-chan has joined #postfix
[05:35:55] *** cilly has quit IRC
[05:36:19] *** DaveH_Work has quit IRC
[05:37:08] *** cilkay has left #postfix
[05:52:17] <Derander> I'm trying to configure postfix with sasl  and courier.  When I try and send a message from my email client, it fails, and SASL authentication failure: Password verification failed
[05:52:17] <Derander>  pops up in /var/log/mail.log.  Has anyone experienced this before?
[05:54:44] <Derander> (I can send messages if I telnet to localhost:25)
[05:59:01] <dragonheart> localhost will normally bypass authentication
[05:59:25] <Derander> Right
[05:59:40] <dragonheart> i'd guess maybe you're using username@dominan as a username instead of just username
[05:59:42] <Derander> I added that to let you know that the server itself is functioning, just not the authentication
[05:59:46] <Derander> Yes, I am
[06:00:11] <dragonheart> is the auth designed for username@domain?
[06:00:33] <dragonheart> can you try just username and see if that works?
[06:00:39] <Derander> I just tried username - no difference
[06:01:08] <dragonheart> the sasl daemon is started
[06:01:13] <Derander> Yeah
[06:01:25] <Derander> er
[06:01:48] <Derander> yes
[06:02:26] <dragonheart> was the error that you originally quoted a postfix error or a courier error?
[06:02:38] <Derander> postfix/smtpd[26214]:
[06:02:46] <Derander> postfix
[06:04:34] *** xnixan has joined #postfix
[06:05:10] <dragonheart> not sure - guess is to turn up the debugging. i've gotta go now
[06:05:16] <Derander> See you, thanks :-)
[06:07:33] *** chmac has quit IRC
[07:15:46] *** idle-boy` has joined #postfix
[07:32:43] *** preaaaaa has joined #postfix
[07:33:17] *** idle-boy`` has quit IRC
[07:37:29] *** kjkoster5489 has joined #postfix
[07:50:47] *** c0mrade has quit IRC
[07:51:07] *** c0mrade has joined #postfix
[08:00:59] *** preaaaaa has quit IRC
[08:20:38] *** Derander has quit IRC
[08:53:41] *** j_s has joined #postfix
[08:58:31] *** lat has joined #postfix
[08:59:06] *** zeitsofas has quit IRC
[08:59:22] *** zeitsofas has joined #postfix
[09:08:19] *** arahman_ has quit IRC
[09:15:18] *** master_o1_master has joined #postfix
[09:26:07] *** amrit|bbl is now known as amrit|zzz
[09:27:40] *** master_of_master has quit IRC
[09:32:07] *** xnixan has quit IRC
[09:33:54] *** madrescher has joined #postfix
[09:35:21] <rokra> Is someone play couter strike on ubuntu?
[09:37:42] *** Zblakany has joined #postfix
[09:40:53] <sysmonk> rokra: oh that SOOO postfix related
[09:41:04] <sysmonk> rokra: /wii rokra
[09:41:08] <sysmonk> woops
[09:46:36] *** Motoko-chan has quit IRC
[09:59:05] <Dewi> I would like to try ACCEPTing on SPF PASS result
[09:59:14] <Dewi> rather than REJECTing on SOFTFAIL or FAIL
[09:59:28] <Dewi> could anyone point me to a postfix-compatible SPF implementation that allows this?
[09:59:40] <Dewi> I don't really want to hack up the script to make it happen
[10:07:16] *** icewaterman has quit IRC
[10:07:17] *** jerlique has quit IRC
[10:07:17] *** Broken|work_ has quit IRC
[10:07:17] *** soren has quit IRC
[10:07:17] *** rabbit7 has quit IRC
[10:07:17] *** echinos has quit IRC
[10:07:17] *** tm-30740 has quit IRC
[10:07:17] *** meshugga has quit IRC
[10:07:17] *** Marticus has quit IRC
[10:07:17] *** cite has quit IRC
[10:07:17] *** saiam has quit IRC
[10:07:17] *** Rockj has quit IRC
[10:07:17] *** wedge_ has quit IRC
[10:07:17] *** rmayorga has quit IRC
[10:07:54] *** Zblakany has quit IRC
[10:08:13] *** meshugga has joined #postfix
[10:08:25] *** wedge has joined #postfix
[10:08:28] *** saiam has joined #postfix
[10:08:36] *** rabbit7 has joined #postfix
[10:08:45] *** icewaterman has joined #postfix
[10:09:06] *** rmayorga has joined #postfix
[10:09:09] *** Marticus has joined #postfix
[10:09:09] *** soren has joined #postfix
[10:09:14] *** echinos has joined #postfix
[10:10:42] *** Zblakany has joined #postfix
[10:14:57] *** Rockj has joined #postfix
[10:17:37] *** jerlique has joined #postfix
[10:17:37] *** Broken|work_ has joined #postfix
[10:17:37] *** tm-30740 has joined #postfix
[10:17:37] *** cite has joined #postfix
[10:18:35] *** jcrites has quit IRC
[10:18:41] *** columbin has quit IRC
[10:23:41] *** kjkoster5489 has quit IRC
[10:23:58] *** cite has quit IRC
[10:23:58] *** jerlique has quit IRC
[10:23:58] *** tm-30740 has quit IRC
[10:23:59] *** Broken|work_ has quit IRC
[10:28:49] *** tuxianer has joined #postfix
[10:37:11] *** idle-boy` is now known as idle-boy
[10:37:28] *** preaaaaa has joined #postfix
[10:37:36] *** lat has quit IRC
[10:40:39] *** Broken|work_ has joined #postfix
[10:40:39] *** tm-30740 has joined #postfix
[10:40:39] *** cite has joined #postfix
[10:40:54] *** kjkoster5489 has joined #postfix
[10:42:06] *** Broken|work_ has quit IRC
[10:42:47] *** Broken|work_ has joined #postfix
[10:48:05] *** Zeit|awy has joined #postfix
[10:48:38] *** kjkoster5489 has quit IRC
[10:54:42] *** Flobbie has quit IRC
[10:55:07] *** c0mrade has quit IRC
[10:59:30] *** xnixan has joined #postfix
[11:08:21] *** jerlique has joined #postfix
[11:08:45] <jerlique> Other than no MX, what would cause postfix to deliver local mail to an smtp_fallback ?
[11:08:50] *** martiancode has joined #postfix
[11:09:40] *** tuxianer has quit IRC
[11:10:50] *** sophokles has joined #postfix
[11:15:02] *** madrescher has quit IRC
[11:31:45] *** dusty has quit IRC
[11:35:35] *** Flobbie has joined #postfix
[11:37:10] *** dusty has joined #postfix
[11:48:06] *** xnixan has quit IRC
[11:57:58] *** Lap_64 has joined #postfix
[11:59:19] *** madrescher has joined #postfix
[12:03:11] *** servettas has joined #postfix
[12:11:06] *** [diablo] has joined #postfix
[12:24:32] *** madrescher has quit IRC
[12:35:27] *** _zsh has joined #postfix
[12:35:45] *** pirho has joined #postfix
[12:44:45] <dragonheart> jerlique: fatal 500* error from the primary i'm guessing
[12:46:32] *** githogori has joined #postfix
[12:51:49] *** F6F has joined #postfix
[12:52:24] *** c00l2sv has quit IRC
[12:53:32] *** idle-boy` has joined #postfix
[12:55:46] *** cilly has joined #postfix
[12:57:13] *** c00l2sv has joined #postfix
[13:03:08] *** idle-boy has quit IRC
[13:11:44] *** hever has joined #postfix
[13:18:48] <orzel> hello. I have the following problem : i would like to send mails from inside a virtual server (vserver), which is used only for php/web
[13:18:54] <orzel> php is so dumb i need a local MTA.
[13:19:21] <sysmonk> or you can use a smtp class
[13:19:32] <orzel> i've tried to configure postfix to only relay mail to my main postfix setup, hosted on the main vserver
[13:19:43] <orzel> sysmonk:  i can't patch all php applications
[13:19:56] <orzel> or did i not understand  ?
[13:20:23] <sysmonk> yeah
[13:20:28] <sysmonk> true
[13:20:44] <orzel> so, i've tried to start postfix, but it insists on having an interface to bind to.
[13:20:53] <sysmonk> but i though there's a way of altering the global function with some other, but anywya, that's not a #postfix question
[13:20:59] <sysmonk> orzel: yes
[13:21:09] <orzel> i can't because 1) 127.0.0.1 is not available on the vserver 2) there's already a postfix on the main vserver, using the port
[13:21:23] <sysmonk> orzel: comment out smtpd in master.cf
[13:21:36] <orzel> i'm trying to configure postfix as a relay only, this belongs to #postfix.. . ?
[13:21:56] <sysmonk> yeah, i was saying that changing functions is not #postfix, but relaying is #postfix
[13:22:06] <sysmonk> orzel: just comment out the smtpd in master.cf, that's all
[13:22:10] <orzel> #smtp      inet  n       -       n       -       -       smtpd
[13:22:14] <sysmonk> yes
[13:22:35] <orzel> it still says "fatal: parameter inet_interfaces: no local interface found for 127.0.0.1"
[13:22:55] *** c00l2sv has quit IRC
[13:22:58] <orzel> ah, it works if i comment this out too
[13:23:04] <orzel> "inet_interfaces="
[13:23:07] <sysmonk> orzel: inet_interfaces = all
[13:23:17] <sysmonk> yeah, defaults are = all
[13:23:32] <orzel> netstat confirms no port is opened, great
[13:23:39] <sysmonk> it's required because postfix still needs to know what interfaces it can use to send mail
[13:23:50] <sysmonk> even though it won't listen on any
[13:24:08] <orzel> i understand
[13:24:13] <orzel> and it works, mail is relayed :)
[13:24:20] <sysmonk> congrats ;)
[13:24:29] <sysmonk> another beer from you :P
[13:24:48] <orzel> postfix is going to cost me so much in beers :)
[13:25:13] <orzel> and magically, everything now works. Cool.
[13:25:19] <sysmonk> or you could just try to learn it o_O :)
[13:25:20] <orzel> thanks a lot. i would never have thought about master
[13:26:32] *** githogori has quit IRC
[13:29:11] <orzel> fun. if i stop postfix, and do  5 times "echo asdfasdf | sendmail mymail"
[13:29:18] <orzel> they are accepted
[13:29:24] <sysmonk> they will be
[13:29:25] <orzel> and when i start postfix, they are actually sent
[13:29:29] <sysmonk> yes
[13:29:36] <sysmonk> 'sendmail' injects them into the queue directly
[13:29:38] <sysmonk> as files
[13:29:42] <orzel> i see.
[13:29:43] <sysmonk> and postfix, when started, scans the queue
[13:29:55] <orzel> smart :)
[13:30:52] <_ruben> hmm .. is there a way to apply smtpd_milters to mails already in the queue (dkim-milter in this particular case)?
[13:32:31] <sysmonk> not really
[13:32:39] <sysmonk> smtpd_milters are done before queuing
[13:33:07] <sysmonk> atleast i don't know a way
[13:34:06] <_ruben> hmm .. i wonder if adding a (temp) relayhost in between which runs said milter would do the trick
[13:34:46] *** havvg has joined #postfix
[13:37:09] <sysmonk> might do, but don't forget to disable it on the reinjected smtpd
[13:43:31] *** denis has joined #postfix
[13:45:30] *** c00l2sv has joined #postfix
[13:45:56] *** worgil has joined #postfix
[13:52:01] *** Zblakany has quit IRC
[13:54:23] *** githogori has joined #postfix
[14:00:07] *** suuuper has joined #postfix
[14:04:24] *** servettas has quit IRC
[14:07:34] *** m0f0x_ has joined #postfix
[14:19:10] *** m0f0x has quit IRC
[14:26:45] *** Lap_64 has quit IRC
[14:27:52] *** Haris has joined #postfix
[14:30:39] *** Juspion has joined #postfix
[14:57:41] *** rokra has left #postfix
[15:03:37] *** romero has quit IRC
[15:11:52] *** madrescher has joined #postfix
[15:14:11] *** havvg_ has joined #postfix
[15:15:09] <Haris> Hello people
[15:15:24] <Haris> I have a postfix+mysql setup. All the email accounts, aliases info is in the db
[15:18:06] <Haris> I also have amavis-new running on a seperate box. postfix works closely with it, so its a postfix+mysql+amavis-new setup. I deleted a few accounts a few days ago. What I need to do, is to accept/allow incomming mail for one email account and allow to have it forwarded the whereever the aliases point. So far, it accepts, sends to local network, but not to remote domains, like gmail for example. So. what is happening is, when I send mail to email@mydomain.
[15:30:45] *** havvg has quit IRC
[15:32:04] <Haris> sends to local network means, it relays to domains hosted on the box, but not to outside domains
[15:34:43] <Haris> If email@mydomain is the email address this mail server is to accept mail for, I have 2 aliases against this email address pointing to outside domains like gmail or yahoo, it will reject email from amavis-new when the time comes to send it out. amavis is set as content-filter
[15:39:30] *** madrescher has quit IRC
[15:51:41] <Haris> postfix can see the alias, just doesn't send mail to them, rejects the sending
[15:51:44] <Haris> how can I make it work
[15:58:47] *** magyar has quit IRC
[16:00:01] *** HarmoniousChaos has quit IRC
[16:00:44] *** HarmoniousChaos has joined #postfix
[16:06:04] *** magyar has joined #postfix
[16:13:18] *** githogori has quit IRC
[16:15:41] <higuita> Haris: mynetwroks should have at least 127.0.0.0/8 (and probably your machine IP)
[16:15:59] <higuita> clamav should be able to relay to your postifx
[16:18:13] <Haris> amavis-new is able to relay to postfix, but postfix denies sending it out to the alias email addresses
[16:18:46] <Haris> mynetwork does have 127.0.0.0/8 in it
[16:30:40] *** denis has quit IRC
[16:34:25] *** t_rn has joined #postfix
[16:35:09] <t_rn> hello. I'm setting up postfix+tls and I'm having problems in understanding a thing.
[16:35:20] *** sin__ has joined #postfix
[16:35:44] <t_rn> Why while using ssl for ssh the public key verification is done just by verifying the fingerprint
[16:36:08] <t_rn> but for a mail server it seems mandatory to have the key signed by a CA?
[16:36:32] <t_rn> (I know I can create a CA by my own. It's full of howtos on this. but I'm not understanding the point.)
[16:42:01] *** denis has joined #postfix
[16:44:37] <t_rn> no clue?
[16:46:00] <Haris> well
[16:46:13] <Haris> its not sending to email aliases that are hosted work outside of the box
[16:46:21] <Haris> How can I fix it?
[16:49:02] <Haris> for a postfix+mysql setup, what should be set so, aliases pointing outside this box itself work?
[16:49:49] <ecrist> t_rn: two different types of keys and infrastructure.
[16:50:19] <ecrist> you can create your own CA, but that CA certificate needs to be imported to all the client machines as 'trusted'
[16:50:41] <ecrist> from there, you can use that CA to sign intermediate certificates for use on mail servers, webservers, VPN, etc.
[16:51:01] <ecrist> all intermediate certificates will be automatically trusted once the CA is trusted.
[16:54:14] <t_rn> ecrist: thank you. i was just missing the point that ssh infrastructure != tls' one
[16:54:21] <dragonheart> and by default there is not verification of CAs with SMTP
[16:54:31] *** Juspion has quit IRC
[16:54:55] <t_rn> dragonheart: you mean that the client won't complain of by self-signed certificate?
[16:55:18] <dragonheart> a mail client may, on server->server it woun't
[16:55:48] <dragonheart> should of clarified that.
[16:56:42] <Haris> it looks like
[16:56:50] <Haris> it is rejecting an outside source to relay
[16:56:51] <Haris> mail
[16:57:30] <Haris> I'm sending mail from my box with my gmail address to this mailserver, its saying it doesn't allow my email address to send email to another
[16:57:42] <Haris> I should mention here, that this box does not employ smtp auth
[16:58:44] <t_rn> Haris: look in the log file.
[16:58:53] <Haris> I have the log
[16:59:01] <Haris> I'm hesitant to paste it though
[16:59:13] <Haris> I have it copied
[16:59:42] <Haris> 454 4.7.1 <ndokovska at gmail dot com>: Recipient address rejected: Access denied (in reply to end of DATA command))
[17:00:03] <Haris> this is what I get when the amavis-new box sends the email back to box running postfix for sending it out
[17:00:35] <Haris> but it delivers ok to aliases that point to email accounts hosted on this same box
[17:00:38] <Haris> itself
[17:00:47] <Haris> so its actually denying relay
[17:01:37] *** denis has quit IRC
[17:01:52] <Haris> is it possible to tell postfix to retry a specific email ID?
[17:02:00] <Haris> retry sending
[17:02:21] <Haris> I have this control with exim and like having this conrol over mail queue
[17:02:38] <Haris> It helps me deliver deferred mail, emptying the queue
[17:07:21] <Haris> I'm sending mail from outside email account to an email account hosted on this box
[17:07:28] <Haris> mail doesn't get delivered locally, that's another problem
[17:07:40] <Haris> but it also doesn't get forwarded to outside email accounts
[17:11:16] <Haris> http://pastebin.ca/1070925
[17:14:15] <Haris> which is the real problem. Its not being forwarded to aliases hosted outside this box
[17:14:20] <higuita> put in mynetworks the 212.110.95.13 IP
[17:15:02] <higuita> you are delivering to the IP, not to localhost (as you should, to be cleanner, but not required)
[17:15:20] <higuita> that is why i said the 127.0.0.0/8 or YOUR IP :)
[17:16:41] *** t_rn has quit IRC
[17:29:40] *** madrescher has joined #postfix
[17:33:53] <Haris> ok, .13 is IP of the amavis-new dedicated box
[17:34:12] <Haris> I don't understand -> you are delivering to the IP, not to localhost (as you should, to be cleanner, but not required)
[17:35:01] <Haris> mail goes from me to postfix > postfix sends it to .13 for inspection > .13 sends it back to postfix > postfix doesn't send mail to aliases
[17:35:11] <Haris> hmm
[17:35:22] <higuita> humm...
[17:35:41] <higuita> so you have a smtp box and the amavis box?
[17:36:02] <Haris> yes
[17:36:12] <higuita> what ever you have, the smtp server is saying that the 212.110.95.13 cant relay
[17:36:15] <Haris> smtp is .3, amavis-new box is .13
[17:36:45] <higuita> you need to put in mynetworks the amavis IP, so it is allowed to accept any email from that IP
[17:36:50] <Haris> this is confusing for me, since postfix should only be referring to amavis-new for inspecting mail for spam/virus
[17:37:02] <Haris> and then when it gets the green signal that its not, it should be sending it out
[17:37:10] <higuita> or else, the smtp box will only accept $mydestination domains
[17:37:59] <higuita> (or $relay_domains)
[17:38:35] <Haris> relay_domains = $mydestination
[17:38:41] <higuita> (or virtual_alias_domain ... you get the ideia :)
[17:38:53] <Haris> postconf -n |grep mydestination gives no output
[17:39:26] <Haris> only virtual_alias_maps is set
[17:39:41] <higuita> no!! the relay domains is "what shall i accept, that i will relay to another server"
[17:39:53] <Haris> postconf -d tells me -> virtual_alias_domains = $virtual_alias_maps
[17:40:01] <higuita> the mydestination is "what domains i should accept and take care as local email"
[17:40:10] <Haris> hmm
[17:40:58] <higuita> so the relay domains is useless with mydestination, as even before relay, postfix already had taken care of it as a local delivery
[17:41:41] <higuita> in postfix, you have several parameters that define what domains your server should accept
[17:41:57] <higuita> the first is mydestination (i'm the final destinatio for this domains"
[17:42:41] <higuita> the next one is the virtual_alias_domain, is the "i also have this virtual domains"
[17:43:18] <higuita> the mydestination is the local delivery, virtual is all the rest (but isnt required to be like this, but its the more common setup)
[17:43:55] <Haris> I don't understand about virtual_alias_domain
[17:44:08] <Haris> What does i also have this domain mean?
[17:44:15] <Haris> deliver locally?
[17:45:05] <higuita> finally, you have the relay_domains, that says that the postfix should accept those domains, but they arent really taken care by this server, it should be forward to another server (relay) ... one of the uses is a secondary MX server, that needs later to send to the primary MX
[17:45:27] <higuita> you have 3 domains, with different mails
[17:45:41] <higuita> postmaster@domain1 postmaster@domain2 postmaster@domain3
[17:46:01] <higuita> they are all different, for different users
[17:46:41] <higuita> the mydestination is just the local domain (usually just the $hostname or $domain)
[17:46:45] <higuita> lets say its domain1
[17:46:47] *** ekimus has quit IRC
[17:47:38] *** neoeinstein has quit IRC
[17:47:41] <higuita> now how about the domain2 and domain3? you cant put this domains as mydestination, the postmaster of those domains would be delivered to the postmaster@domain1
[17:47:49] <higuita> you setup the virtual domains
[17:48:28] <Haris> that bit is ok
[17:48:41] <higuita> in there you say that @domain2 is mapped to another server, to a different user, to a $user-domain2
[17:48:43] <higuita> the samething for domain3
[17:49:37] <Haris> yes, we have that mapped in database. I understand that much
[17:49:50] <higuita> domain1 is local, domain2 is set in virtual, lets say to users *-domain2 and domain3 is a alias to @internal-server.domain3
[17:50:27] <higuita> relay_domains is neither the local delivery, neither the virtual delivery
[17:52:12] <higuita> please note that domain3 could also be setup as a relay domain, as its really being forward to another server, but usually its a matter of taste to help identify better the way the email is working
[17:53:20] <higuita> anyway, the server should accept any of the mydestination, virtual_alias_domains and relay_domains and take care of then
[17:54:17] <Haris> ok, so one thing I understand is, that the domain name is either not showing up in relay_domains or virtual_alias_domains
[17:54:20] <higuita> the amavis is delivering emails that arent in any of this domain lists, so you need to say that amavis IP have permission to relay to other domains
[17:54:26] <Haris> that is why its not relaying to aliases
[17:54:51] <Haris> hmm
[17:55:05] *** madrescher has quit IRC
[17:56:10] <higuita> amavis is delivering a email from=<hariskhan at gmail dot com> to=<ndokovska at gmail dot com>
[17:57:00] <Haris> I understand the relay part here
[17:57:01] <higuita> the postfix server isnt set to accept emails for @gmail.com (and shouldnt), it should be setup to accept relay from amavis (amavis IP in mynetworks)
[17:57:21] <Haris> what I don't understand is, why should amavis-new be trying to deliver, why isn't it just regarded by postfix as a content-filter only
[17:57:46] <Haris> hmm
[17:57:49] <higuita> the first time your server accept the email, it already transform the email to the alias, before being scanned by amavis
[17:57:49] <Haris> that was Q2
[17:57:59] <Haris> yep, true
[17:58:56] *** HarmoniousChaos has quit IRC
[17:58:56] <higuita> that is the way you set it up :)
[18:00:24] <higuita> usually the amavis is setup as a "new smtp" server that do its scan and then redelivers the email back to the first server (with special config to avoid a loop smtp<->amavis)
[18:01:59] *** ecrist has quit IRC
[18:02:41] <higuita> i dont know if content_filter= can be used in amavis, but it if can, it probably will not have this problem, as the email keeps in the same session in the postfix (vs 1 incoming email+ another incoming email from amavis=2 emails)
[18:02:48] <Haris> postfix <- lmtp -> amavis-new
[18:03:12] *** c00l2sv has quit IRC
[18:03:42] <Haris> I have -> mynetworks_style = subnet
[18:03:51] <Haris> how do I add one IP in mynetworks with this ?
[18:04:25] <higuita> ?
[18:04:41] <higuita> edit the /etc/postfix/main.cf
[18:04:47] <Haris> mynetworks_style is not set to host
[18:04:49] <higuita> search the mydestination and put in that line the IP :)
[18:04:52] <Haris> its set to subnet
[18:05:08] <higuita> 212.110.95.13/32
[18:06:56] *** makerc has joined #postfix
[18:07:08] *** ecrist has joined #postfix
[18:07:10] <Haris> well
[18:07:16] <Haris> I put .13 in mynetwrks
[18:07:24] <Haris> it still isn't being allowed to send
[18:07:26] *** c00l2sv has joined #postfix
[18:08:42] <higuita> subnet=host say it just accept emails from the local server
[18:09:00] <higuita> Haris: you have the permit_mynetworks in the smtpd_recipient_restrictions, correct?
[18:09:27] <Haris> nope
[18:09:41] <higuita> :)
[18:09:48] <Haris> smtpd_recipient_restrictions = check_recipient_access mysql:/etc/postfix/mysql_check_sender_access.cf, check_sender_access mysql:/etc/postfix/mysql_check_sender_access.cf, reject_unknown_sender_domain, reject_unknown_sender_domain, reject_unknown_recipient_domain, reject_invalid_hostname, reject
[18:10:27] <higuita> so you never check for mynetworks, you are just checking for user lists
[18:11:10] <Haris> mynetworks = 127.0.0.0/8, <- so this bit gets ignored as well?
[18:11:14] <higuita> put in the permit_mynetworks as the first entry, to allow the relay from those IPs (whatever or not the to email is listed in the user list)
[18:11:18] <higuita> yep
[18:11:51] <higuita> if the email isnt in those mysql lists, the email is rejected
[18:13:09] <Haris> and if they are, it is possible to forward spam as well
[18:14:19] <higuita> i have this restrictions as first level filtering:
[18:14:21] <higuita>  reject_non_fqdn_sender,  reject_non_fqdn_recipient,      sleep 1, reject_unauth_pipelining,      reject_unknown_sender_domain,    permit_sasl_authenticated,      permit_mynetworks,      reject_unauth_destination,      reject_unknown_recipient_domain,        reject_unlisted_recipient
[18:14:35] <Haris> also
[18:14:39] *** dusty has quit IRC
[18:15:00] <Haris> if I enable mynetworks with 127.0.0.0/8 in it, this mail should get through to aliases?
[18:15:04] <higuita> i dont want to accept emails from incomplete or unknown domains ... that is, invalid emails :)
[18:15:17] *** dusty_ has joined #postfix
[18:15:49] <higuita> no, you need also the amavis IP, as the email is comming from that IP
[18:17:15] *** sophokles has quit IRC
[18:17:31] *** ecrist has quit IRC
[18:32:20] *** [diablo] has quit IRC
[18:38:33] <Haris> Dheapmon - Software is not installed or Driver is not running (1012)
[18:39:13] <Haris> dheapinst - Win32k.sys symbol load error, Correct symbol required
[18:39:37] <adaptr> Haris: and we care about that because ?
[18:39:47] <adaptr> go complain in #windows
[18:39:53] <Haris> oops
[18:39:54] <Haris> wrong window
[18:39:58] <adaptr> no, wrong head
[18:43:17] <Haris> ok, that relay problem was fixed
[18:47:32] <Haris> I have a list of domains in the db as well
[18:47:36] <Haris> in the transports
[18:47:43] <Haris> if I put it as remote, what happens
[18:53:10] *** rokra has joined #postfix
[18:59:31] *** rokra has quit IRC
[19:07:49] *** Zblakany has joined #postfix
[19:11:48] *** madrescher has joined #postfix
[19:30:33] *** Haris has quit IRC
[19:31:50] *** Draecos has joined #postfix
[19:32:00] *** m0f0x_ is now known as m0f0x
[19:35:15] *** worgil has quit IRC
[19:44:18] *** havvg_ has quit IRC
[19:49:08] *** raz has left #postfix
[19:50:03] *** servettas has joined #postfix
[19:54:57] *** capt_rogers has joined #postfix
[19:57:23] *** Deffie has joined #postfix
[20:01:01] *** preaaaaa is now known as c0mrade
[20:03:29] *** denis_ has joined #postfix
[20:14:24] *** rokra has joined #postfix
[20:14:55] *** rokra has left #postfix
[20:20:59] *** jeffi has joined #postfix
[20:25:02] *** hever has quit IRC
[20:32:38] *** Derander has joined #postfix
[20:32:45] *** servettas has quit IRC
[20:40:01] *** cssbkgn has joined #postfix
[20:42:10] *** servettas has joined #postfix
[20:48:27] *** Deffie has quit IRC
[20:51:34] *** idle-boy`` has joined #postfix
[20:57:34] *** tom_ has joined #postfix
[20:58:20] <tom_> Can someone help me out with postfix/procmail/spamassassin. it's all working, but i think it's doing procmail *before* spamassassin, which obviously doesn't have the effect I need....I need to switch it and do procmail AFTER spamassassin...
[20:58:21] *** wlmttobks has joined #postfix
[20:59:33] *** deftunix has joined #postfix
[20:59:41] <deftunix> hi all
[20:59:53] <deftunix> i've some question about postfix performance
[21:00:10] <deftunix> with spamassassin and clamav
[21:00:18] <deftunix> have some council
[21:00:19] <deftunix> ???
[21:02:21] *** wedge has quit IRC
[21:02:25] *** wedge has joined #postfix
[21:02:34] <xpoint> 42
[21:04:02] <deftunix> xpoint: 42?
[21:04:56] <tom_> no one's seen my issue before?
[21:05:00] <xpoint> its always the answer to undefined questions
[21:05:22] <tom_> xpoint: :-)
[21:08:21] <deftunix> another question... is possible run postfix queue on NFS and run 2 different postfix server on the some queue?
[21:08:34] <deftunix> sorry for my bad english
[21:08:36] <deftunix> :)
[21:09:16] <xpoint> deftunix, better questions get better answers, but no, postfix cant share spool dir
[21:09:37] <deftunix> xpoint: ok
[21:09:45] *** idle-boy` has quit IRC
[21:15:02] *** v0idnull has joined #postfix
[21:15:47] *** Deffie has joined #postfix
[21:16:42] *** wlmttobks has quit IRC
[21:16:46] <v0idnull> when starting postfix, postfix-script seems to scan my entire filesystem for permissions. This has not happened in the past and I'm not entirely sure what changed.
[21:17:05] <xpoint> deftunix, read NFS_README.html in postfix tarball
[21:17:36] <deftunix> xpoint: thanks
[21:17:47] *** servettas has quit IRC
[21:17:52] <v0idnull> http://pastebin.com/meb3aedf - that's my conf. http://pastebin.com/m1a6a32e6 - that's a small sample of the error messaegs
[21:18:02] *** tom_ has quit IRC
[21:18:59] *** cssbkgn has quit IRC
[21:21:01] <xpoint> v0idnull, hmm funny logs
[21:21:18] *** Deffie has quit IRC
[21:21:27] <v0idnull> xpoint: yeah dude, I haven't a clue what's causing this
[21:22:45] <v0idnull> a few google searches, people seemed to have similar problems, but nothing that that scans /*
[21:23:17] <xpoint> freebsd ?
[21:23:51] <v0idnull> v6
[21:24:06] <xpoint> complain to ports maintainers so
[21:25:09] <xpoint> postfix does not need any python libs or code to run
[21:25:30] <xpoint> so the permissions is a bit bogus
[21:25:55] <xpoint> v0idnull, okay dude
[21:26:14] <v0idnull> xpoint: those lines are just a small bit. it's checking EVERYTHING
[21:27:20] <xpoint> your postfix is installed wroung, dont expect right results from it
[21:28:25] <v0idnull> yes, I get it
[21:28:44] <v0idnull> the question is, what is wrong?
[21:29:13] <v0idnull> postfix/postfix-script: warning: not owned by postfix: /./dev
[21:29:16] <v0idnull> heh
[21:29:45] <xpoint> chroot ?
[21:30:50] <v0idnull> no
[21:34:57] <xpoint> tore_, you need to call spamassassin in a procmailrc file to have it add the headers you like to test in procmail, else you must have headers added external before it goes to local lda in postfix
[21:38:08] <v0idnull> ok well, whatever, I just commented those scans out of postfix-script since they don't really do anything
[21:41:27] <deftunix> xpoint: is not possible sharing postfix queue between multiple instances... thank you
[21:44:22] <xpoint> deftunix, currect
[21:44:42] <xpoint> deftunix, but spool can be on nfs drive, but not shared
[21:45:15] <deftunix> xpoint: ok... and for guantee high availability?
[21:45:41] <xpoint> depends
[21:46:15] <deftunix> xpoint: have you some example for me
[21:46:15] <deftunix> ?
[21:46:33] <xpoint> ha is another issue that should be solved diffrently
[21:47:24] *** xnixan has joined #postfix
[21:47:43] <deftunix> xpoint: my idea is sharing mail spool for improve performances and guarantee high availability
[21:47:48] <xpoint> hardware loadbalancers with more then one postfix behind nat is better, just remember one smtp_helo then
[21:48:09] <xpoint> forget it
[21:48:39] *** ttf has joined #postfix
[21:49:41] <xpoint> if you want speed try ramdisk
[21:50:03] <xpoint> but if reboot or loose of power then you loose mails
[21:50:28] <deftunix> xpoint: ok...
[21:52:19] <xpoint> its better to get seperate disks for the spool dir
[21:56:41] <deftunix> xpoint:  and for integration with content filters?? is better to use amavis or configuring filters as befor and after queue action in postfix?
[22:03:40] *** DigitalNinja has joined #postfix
[22:04:31] *** ttf has left #postfix
[22:04:38] <DigitalNinja> Is postgrey and greylisting still an effective spam prevention technology or have the spamers found a way around it?
[22:06:50] <xpoint> deftunix, amavis can have temp files in ramdisk
[22:07:06] *** deftunix has left #postfix
[22:07:16] *** deftunix has joined #postfix
[22:07:26] <xpoint> deftunix, amavis can have temp files in ramdisk
[22:07:37] *** sin has joined #postfix
[22:07:56] *** hark has joined #postfix
[22:11:14] *** deftunix has quit IRC
[22:14:13] <xpoint> DigitalNinja, greylist still helps
[22:14:31] <DigitalNinja> cool
[22:14:36] <xpoint> DigitalNinja, but not as much as before
[22:14:41] <DigitalNinja> That's what I wanted to hear
[22:14:44] <DigitalNinja> Hmm...
[22:15:15] <DigitalNinja> Should I set the wait time high or is that not a factor
[22:15:50] <xpoint> it can be, but set it to 300 sec is the best
[22:16:26] <xpoint> 300 so spam domains is listed in rbl :-)
[22:16:28] <DigitalNinja> okay
[22:16:34] *** deftunix has joined #postfix
[22:16:57] <DigitalNinja> okay that brings me to my next question
[22:17:27] <DigitalNinja> is there a good site that lists all the rbl's? I would like to add a few of the good ones.
[22:17:46] <adaptr> zen
[22:18:01] <adaptr> you don't need any other
[22:18:10] <DigitalNinja> zen
[22:18:12] <DigitalNinja> okay
[22:18:14] <xpoint> !zen
[22:18:15] <knoba> xpoint: "zen" : http://www.spamhaus.org/zen/ : A composite of all Spamhaus DNSBLs: SBL, XBL and PBL
[22:18:33] *** Draecos has quit IRC
[22:18:35] <DigitalNinja> Okay
[22:18:38] <DigitalNinja> Thanks!
[22:18:44] <DigitalNinja> I'm checking out the site now
[22:20:59] <xpoint> DigitalNinja, http://www.robtex.com/rbl/127.0.0.2.html
[22:21:38] <xpoint> just make sure you not add RIP :-)
[22:22:25] <xpoint> and dont use this list if you are not using eg policyd-weight or other weighed rbl tester
[22:23:16] <DigitalNinja> I'll do my homework before I add anything.
[22:23:19] *** denis_ has quit IRC
[22:23:29] <DigitalNinja> I should get back on the postfix mailing list
[22:24:23] *** sin__ has quit IRC
[22:24:28] *** pirho has quit IRC
[22:25:21] <xpoint> DigitalNinja, super
[22:25:55] *** master_o1_master is now known as master_of_master
[22:33:23] *** F6F has quit IRC
[22:36:12] *** madrescher1 has joined #postfix
[22:36:50] *** madrescher has quit IRC
[22:45:34] *** hever has joined #postfix
[22:46:03] *** Broken|work__ has joined #postfix
[22:47:57] *** hever has quit IRC
[22:52:31] *** mosez has quit IRC
[22:54:02] *** DigitalNinja has quit IRC
[22:56:22] *** F6F has joined #postfix
[22:57:59] *** xpoint has quit IRC
[23:03:17] *** Broken|work_ has quit IRC
[23:18:15] *** martiancode has quit IRC
[23:18:50] *** tom_ has joined #postfix
[23:19:35] <tom_> i'm trying to get procmail/spamassassin working in postfix. Spamassassin to tag e-mail and then procmail to move it to a junk folder. They are both working, but it's calling procmail first -- so the tagged e-mail isn't being moved. Can someone help please i'm at the end of my rope...
[23:21:45] <adaptr> *why* is it calling procmail ?
[23:21:58] <rob0> I recommend amavisd-new, user+spam tagging, and a ~/.forward+spam file, no procmail needed.
[23:22:10] *** j_s has quit IRC
[23:22:12] <tom_> adaptr: the idea was to have spamassassin tag the e-mail, then have procmaail check for tagged e-mail and move to a folder
[23:23:23] <tom_> it *is* tagging the file, but not moving it.  If I send a file to myself with the tag in the subject line -- it gets moved, so i'm guessing it's sending it to procmail before running spamassassin on it
[23:23:38] <adaptr> tom_: I get that, and I asked *why* is it calling procmail
[23:23:59] <adaptr> if you don't actually *know* why, procmail is like a chainsaw in the hands of a 3-year-old
[23:24:09] <adaptr> and you would do well to stay far away from it :)
[23:24:11] <tom_> adaptr: because simply tagging spam isn't enough for me -- I want to move the files into a "Junk" Folder
[23:24:17] <adaptr> sigh
[23:24:32] <adaptr> man procmail, learn the difference between global and per-user processing
[23:24:40] <tom_> i've never done this before -- it's sorta my first attempt
[23:24:42] <tom_> alright
[23:24:54] <adaptr> if *you* deliver to global procmail, then OF COURSE procmail comes first - YOU DELIVER TO IT
[23:25:14] <adaptr> per-user procmail is only executed at the very last step
[23:25:20] *** deftunix has quit IRC
[23:25:55] <tom_> adaptr: ahh! that makes sense than. So /etc/procmailrc gets executed first, and then ~/.procmailrc gets executed last?
[23:26:28] <adaptr> /etc/procmailrc gets parsed IFF something actually executes procmail
[23:26:34] <adaptr> and that would be under your control
[23:27:04] <adaptr> if you want to use procmail to send mali to SA and then again process the mail into folders, you need to call it twice
[23:27:10] <adaptr> I would think this is obvious
[23:27:29] <rob0> Another reason why I would ... oh never mind.
[23:27:48] <adaptr> heheh
[23:27:51] <adaptr> I know, I know
[23:28:04] <adaptr> sssh you'll spoil it all
[23:28:10] <tom_> alright..well.thanks for the help...
[23:29:07] *** tom_ has quit IRC
[23:31:11] *** JoeWulf has joined #postfix
[23:42:23] *** F6F has quit IRC
[23:44:28] *** Joe_Wulf has quit IRC





top