May 23, 2011 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
May 23, 2011 [00:00:49] *** Mac_Weber has joined #postfix[00:01:25] *** codeshah has quit IRC[00:06:16] <Mac_Weber> I'm getting error 554 Relay access denied. I tried google, but I could not fix. Anyone can help me please?[00:12:33] <kale> how do i get postfix to add something to the value i get from ldap ... result_attribute = homeDirectory. i want result_attribute = homeDirectory/../maildir/ ?[00:12:34] *** carloimperia has quit IRC[00:14:17] *** brancaleone has quit IRC[00:16:09] *** verywiseman has quit IRC[00:19:16] *** sphenxes has quit IRC[00:20:17] *** Zeit|awy has quit IRC[00:23:12] <kale> ahh, result_format does it[00:28:19] *** Timzzzz is now known as Timmooo[00:31:21] *** locohost has joined #postfix[00:31:32] <locohost> no way to enter multiple addresses in bccmaps?[00:32:17] *** gremset_ has joined #postfix[00:32:18] <locohost> i have a postfix server that is relaying mail and I would like to fwd a copy of a message from a certain address to several sms email addresses (555555555 at vtext dot com)[00:35:44] *** gremset has quit IRC[00:36:29] *** e-anima has quit IRC[00:50:52] *** brancaleone has joined #postfix[00:52:21] *** Cain has quit IRC[00:53:32] *** Cain has joined #postfix[00:54:47] <kale> Aprogas: i finally got it. thank you very much for the help![00:56:31] *** marchelly has quit IRC[00:56:37] <locohost> anyonre?[00:56:50] <locohost> may have to look into getting procmail setup on this box...[00:57:07] <locohost> been a while since i touched procmail, heh[00:57:26] <locohost> though, im not even sure if that would do what i want if im not doing local delivery[01:00:16] *** dimm has quit IRC[01:02:12] *** wdp_ has quit IRC[01:03:21] <locohost> well., im to go home for now, thks anyway, maybe ill try back later when people are here[01:11:55] <standon> people are here.[01:32:17] *** lunaphyte_ has quit IRC[01:33:34] *** brancaleone has quit IRC[01:33:50] <jimpop> haha loco-host[01:34:41] <jimpop> !knoba[01:34:41] <knoba> jimpop: "knoba" : an informational bot in this channel (see http://workaround.org/f=postfix)[01:35:44] <jimpop> !:)[01:35:44] <knoba> jimpop: ":)" : oh, you say that now![01:36:09] <jimpop> !pie[01:36:09] <knoba> jimpop: "pie" : Postfix Integration Expert: A Postfix certification. The practical portion of the exam involves washing and waxing Dr. Venema's car. See also: !PEE[01:36:17] <jimpop> haha[01:38:01][01:38:01][01:38:08] <jimpop> !StevenSeagal[01:38:09] <knoba> jimpop: "StevenSeagal" : The CockPuncher[01:45:14] *** Gatto has quit IRC[01:47:06] *** lunaphyte has quit IRC[01:49:55] *** dragonheart has joined #postfix[01:51:37] *** lunaphyte has joined #postfix[01:57:05] <adaptr> !pee[01:57:05] <knoba> adaptr: "pee" : Postfix Engineering Expert: An advanced Postfix certification. The practical portion of the exam involves causing physical harm to Dan Bernstein (DJB). See also: !PIE[01:57:34] <jimpop> nice[01:58:03] *** Timmooo is now known as Timzzzz[01:58:46] <adaptr> there should also be be a !PUL[01:58:47] <jeev> lol[01:59:04] <jimpop> Postfix User Loser?[01:59:17] <jimpop> :-)[01:59:29] <adaptr> Postfix Uber-Lord: the only portion of the exam is getting DJB to wash Wietse's car[02:00:18] *** udoprog has quit IRC[02:00:18] *** _nalle_ has quit IRC[02:01:03] <jimpop> hah![02:01:42] *** _nalle has joined #postfix[02:01:46] *** udoprog has joined #postfix[02:13:51] *** eckirchn has quit IRC[02:15:20] *** wdp has quit IRC[02:15:38] *** eckirchn has joined #postfix[02:30:07] *** ziroux has joined #postfix[02:51:46] *** eckirchn_ has joined #postfix[02:53:23] *** Cain` has joined #postfix[02:54:36] *** Cain has quit IRC[02:54:37] *** Cain` is now known as Cain[02:56:35] *** eckirchn_ has quit IRC[02:59:13] *** hparker has joined #postfix[02:59:13] *** hparker has joined #postfix[03:06:49] *** master_of_master has quit IRC[03:07:55] *** gremset_ is now known as gremset[03:08:12] *** asmogator has quit IRC[03:08:54] *** eckirchn_ has joined #postfix[03:10:39] *** master_of_master has joined #postfix[03:26:44] *** etaylor has quit IRC[03:26:59] *** etaylor has joined #postfix[03:30:08] *** etaylor has quit IRC[03:31:56] *** etaylor has joined #postfix[03:48:35] *** mu574n9 has joined #postfix[03:49:46] *** hparker has quit IRC[03:49:59] *** mu574n9 has quit IRC[03:50:29] *** mu574n9 has joined #postfix[03:54:33] <standon> *yawn*[04:17:03] *** Linex has joined #postfix[04:19:36] *** Linex has left #postfix[04:23:40] *** higuita has quit IRC[04:30:04] *** higuita has joined #postfix[04:32:34] *** kuhkatz has quit IRC[04:42:43] *** EugeneKay has joined #postfix[04:45:07] <EugeneKay> Hi, All. Exploring hosting my own email for various domains I own, rather than using Google Apps.[04:45:34] <EugeneKay> My goal is to have IMAP access to a centralized mail server, with all the mail being stored in an $arbitrarydir so that I may integrate backing-up, archival, etc into my existing processes.[04:45:44] *** eckirchn_ has quit IRC[04:46:31] <EugeneKay> After doing some poking about(reading anecdotal evidence), I've decided upon a combo of Postfix and Dovecot, though I'm still looking at webmail packages.[04:46:32] <locohost> need, lots of things can do that, zenoss is cool, im sure lots of people have diff ops here, kinda new to this room[04:47:00] *** flaccid has joined #postfix[04:47:14] <EugeneKay> I found a coupla different guides, want to get a recommendation on a good(correct, complete, secure) one before I proceed with fiddling.[04:47:16] <locohost> no zenoss[04:47:18] <locohost> zimbra[04:47:30] <flaccid> mail is not working because of warning: mail_queue_enter: create file incoming/825350.27463: No space left on device[04:47:41] *** Sugarat has joined #postfix[04:47:41] <flaccid> however there is lots of space; wtf?[04:47:44] <locohost> postfix is always cool and not too much exp with dovcot but here good things about it...light and stable[04:47:55] <locohost> zimbra is closer to outlook/exchange functionality[04:48:23] <locohost> df -hP, flaccid?[04:48:27] <Sugarat> hi all.. I can't seem to get postfix to authentication to Gmail as a relayhost for some reason. I am using a sasl_password which has been hashed, but I still get Gmail saying 'authentication required' ?[04:48:32] <locohost> is it a vm?[04:48:34] <EugeneKay> I'm looking for simple IMAP. Want to give out free email addresses for novelty domain names, possibly from a simple PHP "register" form :-p[04:48:34] <locohost> think provisioned?[04:48:50] <locohost> hrm...would probably get an io error rather then no space on decice, not 100% sure[04:49:02] <locohost> s/think/thin[04:49:08] <EugeneKay> I found this guide, reading it now, but I'm a Red Hatter(my boxen are CentOS 5 and SL 6), rather than a Debnut. http://workaround.org/ispmail/squeeze[04:49:16] <EugeneKay> I should be able to manage, but suggestions are welcome[04:49:19] <flaccid> locohost: plenty http://dpaste.org/y4Zs/[04:49:39] <flaccid> locohost: this is an ec2 instance yes[04:49:55] <locohost> eugene, i might spin up a few in virtual dev environments before deploying on your vps[04:50:02] <locohost> and see what you like[04:50:30] <EugeneKay> I was planning to use my desktop's VMware lab network, yeah.[04:50:30] <locohost> i would probably suggest trying what you tried and check out the zimbra virtual appliance or community rpm packages[04:50:40] <Sugarat> if I am using sasl_passwd how come Postfix is still not authenticating with Gmail ?[04:50:40] <EugeneKay> And I roll with physical boxen. VPSes are for the poor.[04:50:50] <locohost> ran zimbra extensivly under centos in a very large cluster, it is really cool[04:50:59] <locohost> it has its own mta, but you can disable it and use something else like postfix[04:51:07] <EugeneKay> It looks to be more groupware-y.[04:51:28] <locohost> we used postfix for a bit but switched to cloudmark eclarity, which is a not-cheaped pretty sweet solution[04:51:41] *** etaylor has quit IRC[04:51:41] <Sugarat> ignore me... problem solved[04:51:44] <Sugarat> :-)[04:51:47] *** Sugarat has quit IRC[04:52:12] <flaccid> cannot open /etc/mtab.tmp (No space left on device) - mtab not updated[04:52:20] <flaccid> it definately thinks there is no space on the root part[04:52:26] <flaccid> this has to be ec2 failure (again)[04:52:50] <locohost> grep -v rw /proc/mounts[04:53:02] <locohost> thats weird[04:53:13] <flaccid> locohost: doesn't return anything[04:53:20] <flaccid> but i can write to /[04:53:51] <locohost> wtf...[04:54:16] <flaccid> i will reboot again to fix i finks[04:55:00] <locohost> how are these disks connected if its emc and physical? iscsi? hba/mbs fibre channel fabric?[04:55:29] <flaccid> its xen/ec2[04:55:36] <flaccid> ebs[04:55:42] <jimpop> flaccid: what's the output of "df -k"[04:55:55] <locohost> any multipath failovers or anything in kernlog/messages/dmesg[04:56:17] <locohost> can you look on hte host rather then the guest?[04:56:21] <locohost> no much exp with xen[04:56:27] <flaccid> jimpop: http://dpaste.org/QAki/ rebooting[04:57:30] <locohost> is the system navigatable via ssh and such? or screeching to a hault?[04:57:51] <flaccid> system is fine[04:58:01] <locohost> ahh, when you said ec2 i was thinking emc^2 san[04:58:30] <locohost> are you out of inodes?[04:58:45] <locohost> i think thats df -i, im not in a shell[04:58:47] <flaccid> no idea[04:58:59] <flaccid> this happened recently[04:59:01] <flaccid> reboot fixed[04:59:16] <jimpop> flaccid: that's strange[04:59:21] <locohost> getting spammed, lots and lots of small files?[04:59:28] <locohost> would have to be a whole lot to run out of inoes...[04:59:29] <flaccid> yeah i'd say problem is either ec2 or ubuntu[04:59:42] <flaccid> nah its low volume and has filters[04:59:52] <jimpop> i wonder if the disk problems are due to lack of space on the main host (physical)[05:00:17] <locohost> yeah jim, thats why i asked if it was thin provisioned[05:00:23] <flaccid> i guess its possible, but getting aws to give a shit without a paid support plan will be difficult[05:00:36] <jimpop> indeed[05:00:44] <locohost> i assumed you try stopping and starting the services ur using right, not sure what your doing on the application side[05:00:52] <locohost> and then lsof|grep $path[05:00:55] *** mu574n9 has quit IRC[05:01:04] <flaccid> yeah[05:01:05] <locohost> and see if something is using, locking/futex etc[05:01:19] <flaccid> considering you get out of disk space on paths like /etc/foo on root[05:01:21] <flaccid> its pretty silly[05:01:36] <flaccid> /dev/sda1 10G 5.9G 3.6G 63% /[05:01:43] <locohost> yeah, if /r was out of disk space, you couldnt sssh in[05:01:44] <flaccid> lets see how it goes after reboot[05:01:52] <jimpop> flaccid: what size ec2 instance is it?[05:02:04] <locohost> would get "ssh identifier failed" at best[05:02:09] <flaccid> m1.small[05:02:10] <jimpop> i recently had trouble with a micro instance[05:02:12] <flaccid> same thingon reboot[05:02:18] <jimpop> yep, that's what i had m1[05:02:29] <locohost> yeah, maybe an ubuntu bug..havent really seen that in centos/rhel[05:02:36] <jimpop> i was trying to index 63,000 html files. ec2 instance choked[05:03:37] <flaccid> hmm yeah this is bullshit[05:03:39] <flaccid> No space left on device[05:04:10] <locohost> sorry, what exacly is ec2?[05:04:23] <jimpop> amazon ec2 (virtual machines)[05:04:24] <locohost> in 10 words or less[05:04:26] <locohost> ahh[05:04:46] <jimpop> flaccid: google seems to show a lot of hits for "ec2 no space left device"[05:04:49] <flaccid> a public cloud[05:05:02] <flaccid> yeah[05:05:13] <locohost> yeah, totally must be thin provisioning issue[05:05:40] <flaccid> root@starbug:~# df -i[05:05:40] <flaccid> Filesystem Inodes IUsed IFree IUse% Mounted on[05:05:46] <flaccid> i fink you are right on the inodes[05:05:50] <flaccid> but how to fix?[05:05:54] <locohost> it hangs?[05:06:04] <flaccid> no hanging[05:06:14] <jimpop> inodes can only be changed by re-formating the partition[05:06:16] <flaccid> but i need inodes, how to get?[05:06:18] <locohost> you can set the inode size when you create the partition...i think...dont remember syntax[05:06:19] <flaccid> wtf[05:06:31] *** EugeneKay has quit IRC[05:06:52] <locohost> i think u would have to move each volume to a new partition[05:07:16] <locohost> create new /root /opt/ etc, mount from recovery and dd[05:07:20] <locohost> if that is the issue[05:07:31] <flaccid> thats ridiculous[05:07:48] <locohost> df does not hang and df -i does?[05:07:52] <locohost> i woudlnt relaly expect that[05:08:00] <flaccid> no hanging[05:08:12] <locohost> maybe just the timing of commands? df -hP or something does work?[05:08:26] *** gremset has quit IRC[05:08:31] <jimpop> what is the output of "df -i" that should show inode usage[05:09:12] <flaccid> i pasted that above[05:09:16] <flaccid> https://forums.aws.amazon.com/search.jspa?q=No+space+left+on+device&x=0&y=0[05:09:28] <flaccid> i didn't come through sorry[05:09:29] <flaccid> /dev/sda1 655360 655360 0 100% /[05:09:32] <locohost> root@regnar ~]# df -hPi[05:09:32] <locohost> Filesystem Inodes IUsed IFree IUse% Mounted on[05:09:33] <locohost> /dev/mapper/vg_regnar_00-lv_regnar_root 256K 7.6K 249K 3% /[05:09:33] <locohost> /dev/mapper/vg_regnar_02-lv_regnar_nfs01 2.5M 4.5K 2.5M 1% /data/nfs01[05:09:33] <locohost> /dev/mapper/vg_regnar_01-lv_regnar_install 9.5M 21K 9.5M 1% /data/netinstall[05:09:33] <locohost> /dev/mapper/vg_regnar_00-lv_regnar_var 96K 571 96K 1% /var[05:09:35] <locohost> /dev/mapper/vg_regnar_00-lv_regnar_vartmp 40K 10 40K 1% /var/tmp[05:09:37] <locohost> /dev/mapper/vg_regnar_00-lv_regnar_varlog 64K 95 64K 1% /var/log[05:09:39] <locohost> /dev/mapper/vg_regnar_00-lv_regnar_usr 384K 36K 349K 10% /usr[05:09:43] <flaccid> so how can i free inodes?[05:09:46] <flaccid> delete files?[05:09:51] <flaccid> i can't scrap this instance[05:09:53] <locohost> yeah, file references[05:10:15] <jimpop> yep, delete files[05:10:23] <locohost> a 1 tb file will ues not a lot of inodes (1?) and a million files that equial 1tb will use 1m[05:10:32] <locohost> or something[05:11:04] <locohost> u got a spam quarenteen folder or something?[05:11:12] <locohost> running like amavis or something?[05:11:14] <jimpop> lots of log files?[05:11:17] <flaccid> amvis yes[05:11:32] <locohost> is there like a /var/amavis/quarenteen or something[05:11:34] <jimpop> check the amavis conf file for quarantine settings[05:11:53] <locohost> ive seen it called different things in different versions[05:12:11] <locohost> i had a kinda similar problem...wasnt getting no space on device...with amavis once[05:12:29] <locohost> it had 4m+ files in a dir, that were gzipped blocked emails[05:12:50] <locohost> ls -ltah /var/amavis/quarenteen took the box down[05:13:20] <locohost> suse 7.3 production server....yeah...and that was 3 months ago...kinda sad[05:13:44] <flaccid> root@starbug:/var/lib/amavis# ls[05:13:44] <flaccid> Maildir db tmp virusmails[05:14:42] <locohost> du -sh /var/lib/amavis/virusmails[05:14:54] <flaccid> yeah doing a du -cah[05:15:03] <locohost> be aware, thats what hung my session on the box i was remembering[05:15:05] <flaccid> question is, how to remove all the virus mails but not the directories...[05:15:39] <locohost> pffft[05:15:47] <locohost> does * say there is too much shit?[05:15:48] <rob0> !postsuper[05:15:48] <knoba> rob0: "postsuper" : the queue supervision tool for postfix. Use it with the option "-d" to remove mails from the queue. See 'man postsuper' for more information.[05:16:43] <locohost> /usr/sbin/postsuper -d ALL or /usr/sbin/postsuper -d E0156ca[05:16:46] <locohost> or something[05:16:58] <locohost> carefull with all[05:17:04] <jimpop> flaccid: find /var/lib/amavis/virusmails -type f -exec rm -f {} \;[05:17:51] <flaccid> thanks[05:17:53] <locohost> if he is out of inodes, i wonder if any of these commands will run[05:17:57] <flaccid> yeah i'lll keep queue and do what jimpop said[05:18:09] <rob0> nono, use postsuper[05:18:13] <locohost> can you see how many files are in there?[05:18:24] <flaccid> there is many[05:18:34] <rob0> if they're in the queu, that is[05:18:37] <jimpop> flaccid: find /var/lib/amavis/virusmails -type f | wc -l[05:18:43] <locohost> if you can see them thats a good sign[05:18:59] <locohost> yeah, i was just ganna say |wc -l jimpops find[05:18:59] <flaccid> its freeing up[05:19:05] <rob0> oh, they're in amavisd quarantine, disregard[05:19:27] <jimpop> ;-)[05:19:35] <locohost> yeah, rob, lol, i though you were asking the bot for help on a totally different issue, lol[05:19:46] <jimpop> rob0: we are masking as #amavis and #ec2[05:19:51] <rob0> ah[05:19:52] <locohost> holy shit, flaccid, dont postsuper -d[05:20:03] <flaccid> locohost: thats what i said[05:20:09] <flaccid> flaccid: yeah i'lll keep queue and do what jimpop said[05:20:23] * standon gets some popcorn, this is getting good[05:20:26] <flaccid> removing thousands of virusmails now[05:20:28] <flaccid> thanks so much guys![05:20:50] <jimpop> just be careful, after jimpop has had a few beers, he liberally advises rm -rf /usr[05:21:01] <flaccid> i added -v[05:21:03] <locohost> flaccid, when i ran into a similar problem with that dir, was was an old school server that was not maintained for years, and my wc-l would come back with like 4m before completing[05:21:10] <flaccid> removed `/var/lib/amavis/virusmails/C/spam-Cc5jYlErFra5.gz'[05:21:10] <flaccid> removed `/var/lib/amavis/virusmails/C/spam-CW89W81wP76w.gz'[05:21:11] <flaccid> etc...[05:21:17] <locohost> ..like, count some and then error and 4m was part of the error[05:21:20] <rob0> I'll try that. Should I be root for "rm -rf /usr"?[05:21:26] <locohost> i ahd to do like rm spam-A[05:21:28] <locohost> then b[05:21:29] <locohost> then c[05:21:35] <flaccid> this server is uptime of like 2 years[05:21:36] <locohost> and shit[05:21:38] *** EugeneKay has joined #postfix[05:21:39] <flaccid> pretty good for ec2[05:21:40] <locohost> consider yourself lucky[05:22:00] <jimpop> rob0: yes, or sudo ;-)[05:23:07] <locohost> well, clearly the people im talking to are not idiots, lol, it sounded like they would catch a red flag on -d ALL and super, but im a little tired after a long weekend and though you were asking for an example of postsuper...didnt scooll up cus you were afk, got it[05:23:49] <locohost> just finished a successfull email migraion[05:23:58] <locohost> from novell GW 7.1[05:24:14] <locohost> saturday was a 19 hour day[05:24:44] *** MAAAAD has joined #postfix[05:25:14] <locohost> man, i got lucky with that one, calling out amavis[05:25:31] <locohost> i think there may be a retention setting in amavis.conf[05:25:46] <locohost> ....i may have a cron somewhere that cleans that, lol[05:27:31] *** MAAAAAD has quit IRC[05:28:07] <locohost> [root@nx01 jslinger]# ls -ltah /var/virusmails/|wc -l[05:28:07] <locohost> 63872[05:28:44] <locohost> [root@nx01 jslinger]# df -i /var/[05:28:44] <locohost> Filesystem Inodes IUsed IFree IUse% Mounted on[05:28:44] <locohost> /dev/mapper/vg_lx03a_00-lv_nx01_var[05:28:44] <locohost> 2097152 65148 2032004 4% /var[05:31:08] <locohost> [root@nx01 jslinger]# crontab -l|grep nightlystats[05:31:08] <locohost> 30 23 * * * /usr/local/bin/nightlystats[05:31:08] <locohost> [root@nx01 jslinger]# grep rm /usr/local/bin/dailystats[05:31:08] <locohost> rm /var/tmp/spamstats.$tdy[05:31:08] <locohost> find /var/virusmails/ -mtime +7|xargs rm[05:32:06] <flaccid> still deleting virusmails lol[05:32:38] <locohost> if it fixes your problem, try something like i did if you want[05:33:01] <locohost> i think you can set retention in the amavis config, but this way you get to chose when the io happens[05:33:13] <locohost> unless you dont care or just want to discard them[05:33:42] <locohost> its resonable that you may need to pick an email out of there[05:33:43] <flaccid> yeah i'm sure there are amavis.conf settings for pruning and such; just not something i have time to look at[05:34:15] <locohost> yeah, find /var/virusmails/ -mtime +7|xargs rm is easier then polking through config[05:34:45] <locohost> durring your period of lowest io[05:35:06] <locohost> and it should not take long if done taily[05:35:08] <locohost> daily[05:36:04] <locohost> im having trouble with exchange letting me put sms emails in disto lists, ie 555555555 at vtext dot com[05:36:30] <locohost> i want to catch it on a relay, i was trying to figure out how to do this before, but now im thinking[05:36:50] <locohost> if i bcc_map it to localhost.localaccount.domain[05:37:53] <locohost> i should be able to use /etc/aliases and /bin/newaliases and fwd localaccount:to 55555555 at vtext,com,6666666@sms dot sprintpc.com[05:37:55] <locohost> etc[05:37:58] <locohost> that should work, right?[05:40:14] <Tabmow> why don't you just create a normal alias to do that?[05:41:23] <locohost> what do you mean tab?[05:41:31] <locohost> with bcc_map i cant seem to send to more then one address[05:41:41] <locohost> and with alias, i cant seem to catch an account not bound for that machine[05:42:03] <locohost> the email address is an email distro passing through the postfix relay[05:42:35] <locohost> exchange wont let me add addresses to the distro that are not specifically associated with an ad account[05:42:52] <locohost> and alaiases wont let me alias an non local account[05:43:46] <Tabmow> locohost: you can add users as contacts into your Exchange environment...[05:44:52] <locohost> i guess i should be in the windows-server room for that...but that is something you can change in exchange MMC under distrubition lists? cus i could not find anything like that[05:45:17] <locohost> other then per account fwd rules[05:45:30] <locohost> just want to be able to control it in 1 place[05:45:43] <locohost> and turn it off and on in a maintanace window w/o touching many places[05:46:14] <locohost> was considering going the procmail route but thts dirty in this env[05:46:54] <locohost> in the exchange console for a distro, my only options for adding a contact email are whats in AD for that users email ddy[05:48:55] <locohost> so for example, if i want to send a message from my nms to be popped and parsed by some script somwhere[05:49:31] <locohost> i digress[05:49:41] <locohost> how are the rm xargs flowing?[05:50:05] <locohost> can you df -i yet in your other screen?[05:57:36] <flaccid> root@starbug:/var/lib/amavis/virusmails# df -i[05:57:37] <flaccid> Filesystem Inodes IUsed IFree IUse% Mounted on[05:57:43] <flaccid> /dev/sda1 655360 81801 573559 13% /[05:57:52] <flaccid> lots of inodes now :)[05:59:26] <locohost> sweet[05:59:36] <locohost> do future sysadmins a favor and cron that[05:59:50] <flaccid> this is my personal server[06:00:08] <flaccid> and i'd rather look at a better solution than cron[06:00:15] <flaccid> it should be managed by amavis[06:01:45] <locohost> yeah, have you looked at /etc/amavis.conf, its like 3000k lines[06:02:57] <flaccid> well this is debian so it doesn't use that[06:03:25] <locohost> well, let me know where you plan to maintain that, i may be interested[06:03:43] <locohost> i did that as a quick solution cus there were more pressing matters and that seemed fine at thte time[06:03:48] <flaccid> probably in about 6 months i'll get around to it[06:03:56] <flaccid> i.e. amavis chef cookbook[06:04:17] <locohost> lol[06:04:54] <flaccid> because i work in the cloud, my personal stuff gets the least attention unfortunately[06:05:21] <locohost> just cron the rm and be done with it[06:06:10] <locohost> my personal server is a mess[06:06:22] <locohost> my 2ndary ldap and dns have not replicated in like a year[06:07:04] <locohost> and serve differnt zones that dont exist locally anomore[06:07:29] <locohost> ldap=linux openldap central auth[06:08:56] <flaccid> well cron rm command we can both do, so nothing for us to do there...[06:09:22] <flaccid> sounds like you need to start using config management instead of doing it the old way[06:10:43] <locohost> well, my puppet sever at home points to a custom mysql cmdb that i build to integrade with zenoss that is broke at the moment[06:10:57] <locohost> because zenoss totally changed a ton from 2.x to 3.x[06:11:02] <locohost> and it never got updated[06:11:48] <flaccid> yeah which is why some kind of freeze/vcs control is needed so variables dont' change to screw things[06:12:17] <locohost> u dont know what i had to go through to get zenoss to pull from it[06:12:42] <locohost> i like modified a tab in the controll pannel that was referenced by like 10 dtml/zope wrappers[06:12:47] <flaccid> i don't even know what zenoss is but it doesn't sound fun[06:13:06] <locohost> its an open source network monitoring platform[06:13:12] <locohost> like cacti or nagios or something[06:13:31] *** _justdave is now known as justdave[06:13:49] <locohost> more for alerting then cacti, like for a helpdesk or noc to watch to get alerts when a server is broke[06:13:58] <flaccid> ah i use collectd and rightscale[06:14:19] <locohost> for example, will telnet and give a helo to your postfix mta every 5m and graph the responce speed[06:15:25] <flaccid> you can do that with a collectd custom exec plugin in a few lines of code[06:15:34] <locohost> for example, i have my new echange servers set to email my distro list when D: hits 3k iops or 30MB a second or something[06:15:59] <locohost> you can manage lots of systems with that?[06:16:10] <flaccid> as many as you want[06:16:18] <flaccid> zynga do nearly 100K[06:16:45] <locohost> crossplatform...[06:17:08] <locohost> in my env, snmp makes hte most sence[06:17:26] <locohost> with additional port checks and such[06:17:37] <locohost> and thresholds from an external system[06:17:51] <flaccid> you can code any plugin you want[06:18:09] <locohost> have you head of graphite?[06:18:17] <flaccid> nope[06:18:18] <flaccid> http://collectd.org/wiki/index.php/Plugin:Exec[06:18:23] <flaccid> RRD for graphs[06:18:24] <locohost> cacti?[06:18:42] <locohost> rrdtool dump meyDC.rrd ftw[06:18:53] <flaccid> heard of cacti but not used[06:19:32] <locohost> graphite is suppsed to be really cool, i messed iwth it for a few hours[06:19:45] <locohost> but did not get too far past the install and getting the gui up[06:19:58] <locohost> its what orbits.com build for tracking their flights and stuff[06:20:04] <locohost> really cool looking graphs for realtime shit[06:20:22] <flaccid> rrdtool works fine..[06:20:22] <locohost> but, it has nothing for data colection at all[06:20:35] <flaccid> collectd for data collection[06:20:45] <locohost> so, you have to do all the data collection yourself[06:21:04] <locohost> yeah, i like RRD, i have had hidden urls on my cacti server before actualy[06:21:29] <locohost> and just wrote my own scripts that pulled data out of cactis rrd and generated my own graphing statemens[06:21:43] <locohost> for example, if you had 50 mail drops on the same san[06:21:56] <flaccid> yeah thats what the exec plugin is for with collectd. you can script whatever - you just make it output in rrd format[06:21:58] <locohost> and wanted to make a stacked graph of their system iops[06:22:29] <locohost> so, it writes the graphing statements for you?[06:22:45] <locohost> for me, the gathering of the data is the more difficult part[06:23:02] <locohost> i almost prefer to write hte rrd graph statements myelf[06:23:56] <locohost> cacti does that, and its kinda convienent, cus it will just generate the .png files when your navigating the interface[06:24:22] <locohost> so, if someone isnt using the gui, your script isnt writing hte image files every 5m[06:24:25] <flaccid> an example: http://support.rightscale.com/12-Guides/RightScale_Methodologies/Monitoring_System/Writing_custom_collectd_plugins/Custom_Collectd_Plug-ins_for_Linux[06:25:16] <locohost> OoO ruby[06:25:37] <flaccid> you can use anything of course or a binary - it just has to output the RRD to stdout..[06:26:27] <flaccid> i wrote a simple bash one: http://support.rightscale.com/12-Guides/RightScale_Methodologies/Monitoring_System/Writing_custom_collectd_plugins/Example%3A_memory_buffers_collectd_exec_plugin[06:30:56] <locohost> http://imageshack.us/f/233/cacti.png/[06:31:51] *** Mac_Weber has quit IRC[06:32:30] <flaccid> locohost: ah yep that is rrd graphs. looks like its from the disk plugin of collectd or similar[06:32:47] <flaccid> we got that by default[06:33:15] <flaccid> http://collectd.org/wiki/index.php/Plugin:Disk[06:33:28] <flaccid> they probably just added more types to the types.db or something[06:34:52] <locohost> u use the UCD-mib in linux and snmp informant in suckdows?[06:37:15] <flaccid> can't say i've heard of any of that, no[06:38:59] <locohost> http://www.opennms.org/wiki/Monitoring_disk_IO_with_Net-SNMP[06:39:47] <locohost> yeah, thats what it uses[06:40:04] <flaccid> though collectd disk plugin works out of box but yeah[06:40:19] <locohost> yeah, you configure net-snmp on the linuxs clientsw[06:40:26] *** dragonheart has quit IRC[06:40:37] <locohost> and it uses the ucd diskio mib which has the variables and api and crap for net-snmp[06:40:45] *** mu574n9 has joined #postfix[06:41:06] *** asb has quit IRC[06:41:18] <flaccid> i work in the cloud so i don't really have a need for it[06:41:29] *** asb has joined #postfix[06:41:36] <jimpop> two others: monit and munin[06:41:53] <jimpop> both extremely simple, and very low overhead[06:42:01] <flaccid> thats for monitoring processes[06:42:35] <flaccid> pretty limited. collectd does this though http://collectd.org/wiki/index.php/Plugin:Processes[06:42:36] <jimpop> and system stuff[06:42:39] <locohost> yeah, ill bet you will find .1.3.6.1.4.1.2021 somwherein its config :)[06:43:25] <locohost> yeah, there are tons of things in the udc mib[06:43:26] <locohost> all that[06:43:39] <locohost> and context swiches and stuff[06:45:05] <locohost> i get postfix stats with it?[06:45:09] *** hever has joined #postfix[06:45:24] <locohost> my postfix servers just get a daily email....[06:45:59] <locohost> i mean..[06:46:08] *** hever has quit IRC[06:46:19] <locohost> for monitoring/trending. they send a nightly email that is a summary of a parsse of /var/log/maillog[06:46:31] *** hever has joined #postfix[06:47:44] <locohost> perl /usr/local/bin/pflogsumm.pl -d today -h 100 /var/log/maillog > /var/log/mailstats/postfixstats.$tdy[06:47:45] <locohost> head -n 51 /var/log/mailstats/postfixstats.$tdy|mail -s "End of day inbound Email Stats Summary for nx01 $tdy" -c $cc $rcptto[06:48:49] <locohost> pflogsumm.pl is pretty sweet actually but to integrate it with rrd myeslf would sure be a chore[06:50:02] <locohost> well, its late. good luck with your inodes, see u guys around[06:50:49] *** basho__ has joined #postfix[06:51:15] <flaccid> same i gotta run; thanks locohost and all[06:51:17] <flaccid> cyas again soon[06:51:22] *** flaccid has left #postfix[06:55:47] *** Timzzzz is now known as Timmooo[07:07:16] *** twobithacker has quit IRC[07:18:26] *** twobithacker has joined #postfix[07:22:15] *** Niemi has joined #postfix[07:26:47] *** krzee has quit IRC[07:32:47] *** dimm has joined #postfix[07:38:05] *** loddafnir has joined #postfix[07:40:21] *** dimm has quit IRC[07:40:21] *** dimm has joined #postfix[07:41:00] *** Timmooo is now known as Timzzzz[07:51:09] *** twobithacker has quit IRC[07:52:48] *** twobithacker has joined #postfix[07:57:35] *** twobithacker has quit IRC[07:58:28] *** krzee has joined #postfix[07:58:30] *** twobithacker has joined #postfix[08:27:29] *** Zeit|awy has joined #postfix[08:29:43] <dimm> can i use Maildir/ with virtual accounts ?[08:30:06] <Tabmow> dimm: yes[08:30:12] <pj> dimm: sure you can[08:30:27] <pj> !maildir[08:30:27] <knoba> pj: "maildir" : a mailbox format introduced by qmail where emails are saved as individual files in a directory structure rather than into a single flat text file. Postfix settings like !home_mailbox and !virtual_mailbox_maps will allow delivery to maildir if the path value returned ends in /[08:30:37] <pj> dimm: ^^^^^^^^^^^^[08:31:43] <dimm> !virtual_mailbox_maps[08:31:43] <knoba> dimm: "virtual_mailbox_maps" : a configuration parameter in the main.cf: Optional lookup tables with all valid addresses in the domains that match $virtual_mailbox_domains.[08:39:27] *** e-jones has joined #postfix[08:40:23] *** gerhard7 has joined #postfix[09:01:41] *** joschi_ is now known as joschi[09:05:45] *** mu574n9 has quit IRC[09:08:56] <dimm> it is workin' :)[09:09:18] <pj> cool, glad to hear it :-)[09:13:48] *** Guip`aw is now known as bezourox[09:14:01] *** zorg1 has joined #postfix[09:17:20] *** e-anima has joined #postfix[09:26:13] *** cilly has joined #postfix[09:28:41] *** jujugre has joined #postfix[09:33:45] *** breaker has joined #postfix[09:34:56] *** breaker has quit IRC[09:35:04] *** breaker has joined #postfix[09:35:40] *** breaker has quit IRC[09:35:47] *** breaker313 has joined #postfix[09:50:32] *** sphenxes has joined #postfix[10:15:19] *** sep has quit IRC[10:16:37] *** chrisq has quit IRC[10:17:12] *** Niemi has quit IRC[10:17:13] *** brancaleone has joined #postfix[10:18:34] *** TomHome has joined #postfix[10:28:22] *** sep has joined #postfix[10:30:26] *** dimm has quit IRC[10:31:27] *** dimm has joined #postfix[10:34:12] *** hever has quit IRC[10:40:54] *** hever has joined #postfix[10:42:02] *** beginer has joined #postfix[11:05:45] *** UQlev has joined #postfix[11:08:23] *** marchelly has joined #postfix[11:18:04] *** breaker313 has quit IRC[11:18:35] *** matt1982 has joined #postfix[11:19:51] *** EugeneKay is now known as EugeneKaway[11:21:53] *** hever has quit IRC[11:22:18] *** hever has joined #postfix[11:35:10] *** pj has quit IRC[11:38:04] *** n0sq has quit IRC[11:46:56] *** stefano has joined #postfix[11:47:54] <stefano> Is it possible to forward per outgoing sender all emails to an other destination?[11:48:21] <tuxick> ??[11:48:52] *** busta has joined #postfix[11:49:18] <stefano> well, i want to forward all email of one account to an other destination, outgoing and incoming[11:49:36] <stefano> actually i can forward incoming email by setting an alias up[11:49:44] <stefano> what about ougoing email?[11:49:47] <stefano> outgoing[11:50:05] <Aprogas> Usually you let outgoing mail go to the intended destination.[11:50:17] <Aprogas> Do you want to intercept and redirect that mail?[11:50:46] <Aprogas> Effectively shutting down email-service for that user?[11:50:51] *** marchelly has quit IRC[11:50:52] *** n0sq has joined #postfix[11:50:57] <stefano> i want to set up an alias for outgoing emails too, i want the mail to go to the intended destination, but forward it to an other destination too[11:51:24] <Aprogas> !tell stefano sender_bcc_maps[11:51:24] <knoba> stefano: "sender_bcc_maps" : a configuration parameter in the main.cf: Optional BCC (blind carbon-copy) address lookup tables, indexed by sender address. The BCC address (multiple results are not supported) is added when mail enters from outside of Postfix.[11:51:46] <Aprogas> !tell stefano recipient_bcc_maps[11:51:46] <knoba> stefano: "recipient_bcc_maps" : a configuration parameter in the main.cf: Optional BCC (blind carbon-copy) address lookup tables, indexed by recipient address. The BCC address (multiple results are not supported) is added when mail enters from outside of Postfix.[11:51:50] <Aprogas> Not sure which one you will need.[11:52:07] <stefano> thanks, i'll give a look[11:55:30] *** chrisq has joined #postfix[11:58:16] *** weedar has joined #postfix[12:00:12] *** weedar has quit IRC[12:01:09] *** weedar has joined #postfix[12:02:57] *** weedar has quit IRC[12:03:03] *** wdp has joined #postfix[12:03:03] *** wdp has joined #postfix[12:03:23] *** weedar has joined #postfix[12:04:37] *** weedar has quit IRC[12:05:05] *** weedar has joined #postfix[12:06:37] *** weedar has quit IRC[12:06:55] *** weedar has joined #postfix[12:08:23] *** ionte has quit IRC[12:09:16] *** ionte has joined #postfix[12:10:32] *** weedar has quit IRC[12:18:05] *** shal3r has joined #postfix[12:20:23] *** Twinkletoes has joined #postfix[12:22:54] *** zorg1 has quit IRC[12:24:18] *** iam126 has joined #postfix[12:25:11] <iam126> !debug[12:25:11] <knoba> iam126: "debug" : http://www.postfix.org/DEBUG_README.html : a good starting point for how to deal with problems and to report information to those who might help. Post your information in a pastebin such as http://pastebin.ca/ or http://dpaste.com/[12:29:42] *** cpm has joined #postfix[12:29:42] *** cpm has joined #postfix[12:35:16] *** zorg1 has joined #postfix[12:38:19] *** dimm has quit IRC[12:39:03] *** dimm has joined #postfix[12:41:20] *** gerhard7 has quit IRC[12:49:41] *** shoonya has joined #postfix[12:55:10] *** wdp_ has joined #postfix[12:55:11] *** shoonya has quit IRC[12:58:29] *** wdp has quit IRC[13:02:10] *** busta has quit IRC[13:06:00] *** UQlev has quit IRC[13:07:48] *** hever has quit IRC[13:17:31] *** stefano has left #postfix[13:20:11] *** cilly has quit IRC[13:20:51] *** dimm has quit IRC[13:20:51] *** dimm has joined #postfix[13:22:24] *** jfried has joined #postfix[13:24:41] *** JoKoT3 has joined #postfix[13:24:45] *** EugeneKaway is now known as EugeneKay[13:25:36] *** busta has joined #postfix[13:37:48] *** cilly has joined #postfix[13:49:41] *** cilly has quit IRC[13:52:36] *** mambaw has quit IRC[13:53:15] *** plee has quit IRC[13:55:02] *** kuhkatz has joined #postfix[13:56:39] *** cilly has joined #postfix[13:56:59] *** plee has joined #postfix[13:57:15] *** Lujeni has joined #postfix[14:04:02] *** amir has quit IRC[14:09:50] *** TomHome has quit IRC[14:10:34] *** ketema has joined #postfix[14:11:00] *** dimm has quit IRC[14:12:26] *** dimm has joined #postfix[14:16:53] *** amir has joined #postfix[14:18:48] *** ketema has left #postfix[14:18:55] *** lunaphyte_ has joined #postfix[14:27:35] *** beginer has quit IRC[14:28:05] *** cilly has quit IRC[14:30:15] *** LinuxCode has joined #postfix[14:51:18] *** raavan has quit IRC[15:02:14] *** gerhard7 has joined #postfix[15:12:52] *** weedar has joined #postfix[15:17:22] *** mandragor has joined #postfix[15:17:22] *** weedar has quit IRC[15:18:38] *** mandragor has quit IRC[15:19:08] *** mandragor has joined #postfix[15:22:12] *** mandragor has quit IRC[15:22:43] *** mandragor has joined #postfix[15:22:57] *** mandragor has quit IRC[15:23:12] *** mandragor has joined #postfix[15:24:36] *** mandragor has quit IRC[15:34:26] *** hever has joined #postfix[15:35:15] *** hever has quit IRC[15:35:41] *** hever has joined #postfix[15:44:24] *** empity has quit IRC[15:45:43] *** wame has joined #postfix[15:46:09] *** wdp_ has quit IRC[15:56:47] *** Section1 has joined #postfix[16:01:22] *** Gatto has joined #postfix[16:02:42] *** bodom has joined #postfix[16:03:20] *** dimm has quit IRC[16:08:14] *** twobithacker has quit IRC[16:15:50] *** cilly has joined #postfix[16:17:23] *** busta has quit IRC[16:20:59] *** hparker has joined #postfix[16:21:00] *** hparker has joined #postfix[16:22:45] *** morphine has joined #postfix[16:24:32] *** twobitha1ker has joined #postfix[16:26:01] <morphine> so I implemented SPF, and nearly everything is okay[16:26:30] *** twobitha1ker is now known as twobithacker[16:26:37] <morphine> except that I still see the odd message or two getting past it, apparently having the SPF check OK, but still spoofing the "From" field[16:26:49] <morphine> did I do anything wrong, or SPF simply doesn't do that?[16:27:02] <morphine> (I know this doesn't pertain Postfix, but I thought it'd be the place to ask)[16:28:57] *** mike_ has joined #postfix[16:29:19] *** mike_ is now known as Guest27745[16:30:35] <Guest27745> Does anyone know if there is a way to direct emails from a specific address to another relay? I know you can do it by domains via the transport mapping file but not sure if you can specify particular address?[16:31:01] *** Guest27745 is now known as myki3[16:32:22] *** f3ew has quit IRC[16:38:09] *** EugeneKay has quit IRC[16:45:17] *** f3ew has joined #postfix[16:46:29] *** bodom has quit IRC[16:46:33] *** EugeneKay has joined #postfix[16:48:56] *** empity has joined #postfix[16:49:51] <myki3> nevermind I had miss read the address and can use a transport map[16:56:39] *** Lenhix has joined #postfix[16:58:22] *** MrRagga has quit IRC[17:00:02] *** UQlev has joined #postfix[17:11:11] *** e-anima is now known as e-anima_afk[17:26:01] *** e-jones has quit IRC[17:30:27] *** seekwill has joined #postfix[17:53:48] *** mambaw has joined #postfix[18:03:02] *** jujugre has left #postfix[18:03:19] *** bio__ has quit IRC[18:06:10] *** zorg1 has quit IRC[18:06:13] *** ced117 has joined #postfix[18:06:13] *** ced117 has joined #postfix[18:08:09] *** e-anima_afk has quit IRC[18:13:41] *** e-anima has joined #postfix[18:18:08] *** EugeneKay is now known as EugeneKaway[18:19:16] *** jfried has quit IRC[18:21:00] *** cilly has quit IRC[18:25:35] *** myki3 has quit IRC[18:30:04] *** brancaleone has quit IRC[19:03:18] *** carl- has joined #postfix[19:05:08] *** Toerkeium has joined #postfix[19:13:15] *** e-anima has quit IRC[19:13:25] *** e-anima has joined #postfix[19:15:02] *** Lenhix has quit IRC[19:15:18] *** sep has quit IRC[19:19:33] *** sep has joined #postfix[19:20:26] *** busta has joined #postfix[19:26:49] *** Verilium has quit IRC[19:32:38] *** UQlev has quit IRC[19:34:09] *** Twinkletoes has quit IRC[19:45:20] *** Lujeni has quit IRC[19:50:38] *** shoonya has joined #postfix[19:51:27] <seekwill> Does anyone do any sort of spam checks on the From header, even if there is a reply-to?[19:56:10] <Aprogas> jNews claims that having different From and Reply-To is scored by some spamfiltering software, but anything jNews developers claim should be taken with a grain of salt[19:58:23] <seekwill> What we've been seeing is a lot of people using "noreply at company dot com", and a reply-to that sets to custservice@, sales@, etc. A lot of the big receivers really like having consistant From headers[19:58:46] <seekwill> So just wondering what do people do for noreply at company dot com. if people check to see if that's a real address, etc.[19:59:24] <tharkun> No one hass complained yet. So I can't tell you anything useful[19:59:51] <jimpop> if From != Reply-to, assume Bulk sidetrack/filter accordingly[20:01:16] <jimpop> of course, if From == Reply-to, don't assume it's not Bulk, and filter accordingly ;-)[20:06:47] <seekwill> jimpop: That's not what Yahoo and Gmail are saying...[20:07:12] <seekwill> Even for transactional mail. I guess it's the interpretations of their guidelines...[20:18:09] <jimpop> they have guidelines? :-)[20:31:55] *** tris has quit IRC[20:33:10] *** tris has joined #postfix[20:36:45] *** iam126 has left #postfix[20:45:22] *** mu574n9 has joined #postfix[20:47:07] *** digitolx has quit IRC[20:47:43] *** ssureshot has joined #postfix[20:50:31] *** _Captain has joined #postfix[20:56:02] *** morphine has quit IRC[20:56:46] *** Motoko has joined #postfix[21:00:00] *** KaiForce has joined #postfix[21:07:32] *** cpm has quit IRC[21:11:15] *** Matic`Makovec has joined #postfix[21:12:45] *** ketema has joined #postfix[21:13:52] *** marchelly has joined #postfix[21:16:35] *** busta has quit IRC[21:18:41] *** Matic`Makovec has quit IRC[21:25:29] *** pj has joined #postfix[21:34:00] *** lunaphyte_ has quit IRC[21:41:28] <seekwill> omg[21:41:32] <seekwill> I need to kill someone[21:41:36] <seekwill> thumbs![21:42:28] *** wdp has joined #postfix[21:42:28] *** wdp has joined #postfix[21:42:56] *** Matic`Makovec has joined #postfix[21:44:35] *** marchelly has quit IRC[21:47:06] *** jan__ is now known as gothos[21:48:28] *** lunaphyte_ has joined #postfix[21:48:28] *** lunaphyte_ has joined #postfix[21:56:08] *** Timzzzz is now known as Timmooo[22:01:57] *** brancaleone has joined #postfix[22:02:25] *** carl- has quit IRC[22:03:01] *** marchelly has joined #postfix[22:04:10] *** busta has joined #postfix[22:04:14] *** EugeneKaway has quit IRC[22:07:46] *** marchelly has quit IRC[22:10:04] *** EugeneKaway has joined #postfix[22:11:09] *** hever has quit IRC[22:14:15] <Aprogas> I use noreply for return-path on automated mailings because I am a dirty spammer who doesn't handle bounces.[22:15:21] <seekwill> Huh?[22:15:25] *** sbeam has joined #postfix[22:16:28] <Aprogas> Not sure why one would set From to noreply but Reply-To to the real address.[22:17:15] <seekwill> So across several mailings you would have a consistant From[22:17:19] <seekwill> header[22:17:41] <seekwill> 'Nevermind[22:20:55] <Aprogas> I should probably raise the issue of not using a bounce handler again. Except your subscription list is provided by a parent organisation, which has a crappy CRM, so it won't really do any good since we can't purge the invalid addresses from the database.[22:21:02] <Aprogas> s/your/our/[22:22:06] *** ced117 has quit IRC[22:22:29] <seekwill> huh?[22:22:55] <Aprogas> I'm just venting some frustration.[22:23:06] <seekwill> Oh... about what?[22:23:44] <Aprogas> About sending repeated mailings to nonexistent addresses.[22:24:24] <seekwill> Oh[22:24:29] <seekwill> Who does that? :P[22:24:53] <Aprogas> I do, but I'm too lazy to fix it the hard way, and the easy way isn't an option.[22:25:14] <seekwill> So you're sending messages to invalid addresses? Yeah, that's spammy...[22:25:40] <seekwill> Really really spammy[22:26:05] <seekwill> You can't use a suppression or exclusion list?[22:26:29] *** ketema_ has joined #postfix[22:26:39] <Aprogas> I could, but I'm not really familiar with the mass-mailing software, which is a plugin to the CMS.[22:27:47] <seekwill> How about something in Postfix?[22:27:53] * seekwill notes channel topic[22:27:55] <Aprogas> I think we use custom software to parse the Excel-file provided by the parent organisation to fill the subscription lists, but I think any changes on our end are just overwritten on the next sync.[22:28:23] <seekwill> Can you explain to me your process?[22:28:37] <Aprogas> jNews aborts the entire mailing if the SMTP-server throws a single error. And when it retries, it starts back from scratch, rather than skipping the recipients that were succesful.[22:28:56] <seekwill> oh...[22:28:56] <Aprogas> Well, actually I haven't tested that under jNews, but Acajoom did that.[22:29:05] *** ketema has quit IRC[22:29:05] *** ketema_ is now known as ketema[22:29:07] <seekwill> I thought you were making fun of me all this time[22:29:11] <Aprogas> I just realised I can disable emailaddresses without removing them, maybe that will keep the changes.[22:29:18] <Aprogas> No, I'm talking about myself.[22:30:16] <Aprogas> The envelope-sender (and hence return-path) used to be set to postmaster@ to I would receive all bounces personally.[22:30:58] <seekwill> yikes![22:31:37] <Aprogas> +- 5000 recipients, +- 120 bounces I think.[22:31:51] <seekwill> That still sucks[22:32:41] <Aprogas> Hence why I now use noreply@ and sink bounces into /dev/null[22:32:49] <seekwill> oh no[22:33:27] <seekwill> If I were you, I'd outsource that[22:33:41] <seekwill> If you get an Excel spreadsheet, just upload that to the ESP[22:34:06] <seekwill> It would be more expensive, but will make your mailing campaign much more effective[22:34:27] <Aprogas> It's a non-profit volunteer organisation.[22:34:53] <seekwill> There are many ESPs giving out heavily discounted rates[22:40:00] <seekwill> Aprogas: URL?[22:40:07] <adaptr> seekwill isn't one of them[22:40:15] <Aprogas> It's a secret.[22:40:30] <seekwill> Spammer[22:40:55] <Aprogas> I know there are multiple technical solutions to this, I just didn't get around to working on them yet.[22:41:18] <Aprogas> The mailserver is also a forwarder that doesn't do SRS even though most of the forwarding is done to GMail and Hotmail accounts.[22:41:44] <seekwill> Spammer[22:41:46] <seekwill> :P[22:42:27] <seekwill> But seriously, it's understandable.[22:42:40] <seekwill> But you really would be marked as a spammer by the elitist of the channel...[22:43:07] <seekwill> And various antispam enthusiasts :)[22:44:32] <Aprogas> Have I told you the 5000 recipients didn't even explicitly opt-in, other than signing up for general membership of the organisation? :) Or that our opt-out link is malfunctioning? :P[22:44:34] <adaptr> I thought you were the elitist of the channel![22:44:57] <Aprogas> I exaggerate a bit now, it's time for my yoghurt.[22:46:18] <adaptr> lack of yoghurt makes you lie ?[22:46:26] <adaptr> score one for science[22:48:35] <Aprogas> Lack of sleep makes me lie and I eat yoghurt before I go to bed.[22:50:04] <seekwill> So you're sending mail to 5k receipients that didn't really opt in, and you're not processing their bounce requests?[22:50:44] *** Timmooo is now known as Timzzzz[22:51:02] <seekwill> Is your system on any RBLs yet? :)[22:51:14] <Aprogas> Not yet.[22:51:32] <seekwill> Lucky you have a very small list :)[22:51:51] *** marchelly has joined #postfix[22:54:56] *** busta has quit IRC[23:07:53] *** gerhard7 has quit IRC[23:10:00] *** matadore has joined #postfix[23:18:05] <matadore> Hello folks, I am trying to setup a pyramid clustered system, and am using some fallback_relays .... is it possible to create the same user/passwd and let it auth by itself? instead of adding 10diff ips to mynetworks = ?[23:18:40] <lunaphyte_> dunno what that means.[23:18:52] <lunaphyte_> what is "it"?[23:19:09] <lunaphyte_> and what is it authenticating against? and for what purpose?[23:20:12] *** sphenxes01 has joined #postfix[23:20:24] <seekwill> Why not just use a load balancer?[23:20:40] *** sphenxes has quit IRC[23:20:53] <matadore> seekwill: am currently using a load balance.[23:21:03] <matadore> seekwill: i got level 1, level 2 level 3[23:21:26] <matadore> seekwill: mail gets sent to via smtp auth to level 1... then level 1 servers fallback to level 2.[23:21:26] <seekwill> The pyramid cluster system is if one host/IP cannot send it, you try another host?[23:21:36] <matadore> yes[23:22:19] <seekwill> I don't believe you need to pass the auth once it gets into Postfix[23:22:24] <matadore> I am tryign to find a way to authenticate level1 -> level2 .. without editing level2 main.cf mynetworks = with every level1 ip[23:22:33] <seekwill> oh[23:23:01] <seekwill> Why do you need this?[23:23:17] <matadore> seekwill: just tedious to change and rearrange[23:23:28] <seekwill> Why do you need a clustered system like that?[23:23:55] <seekwill> I don't think most people need fallback_relay[23:24:13] <matadore> am not trying to get smart with ya but why else would someone build a cluster system?[23:24:40] <seekwill> Well, people who do usually know what they're doing :)[23:24:45] <matadore> hehe[23:24:55] <seekwill> So why do _you_ want it?[23:25:16] <matadore> to make sure mail is delivered in a timely manner[23:25:32] <seekwill> You don't need a cluster for that[23:25:46] <matadore> why not[23:25:54] <seekwill> I don't even think fallback_relay is even meant for that use case[23:26:09] <matadore> if mail gets deferred i want it to reach at a given time[23:26:13] <seekwill> Why do you need fallback_relay? How does that help you?[23:26:25] <matadore> fallback relay would allow that if 1 box fails it will get deliveried eventually by another in a timely manner[23:26:47] <seekwill> Your'e talking like a yahoo 421 deferral?[23:27:00] <matadore> yea some of that[23:27:27] <seekwill> iirc, (and you'll need to get a postfix expert to help you) fallback relay only comes into play when postfix cannot connect to the server[23:27:35] <adaptr> !fallback_relay[23:27:35] <knoba> adaptr: "fallback_relay" : a configuration parameter in the main.cf: Optional list of relay hosts for destinations that can't be found or that are unreachable.[23:27:41] <seekwill> Trying to push your message to Yahoo when they told you to wait is going to get you blacklisted[23:27:56] <adaptr> note: this is NOT "retrying", which postfix always does in case the destination is unavailable[23:28:15] <seekwill> I'm getting good at Postfix :)[23:28:29] <seekwill> I should use it one day..[23:28:33] <adaptr> just you stick to Exchange buddy, it's what you're good at[23:28:38] <seekwill> okie[23:29:13] *** sbeam has quit IRC[23:29:39] <matadore> so how to avoid listing multiple ips in mynetworks for relaying without building a vpn?[23:29:51] <lunaphyte_> !tell matadore smtpauth[23:29:51] <knoba> matadore: "smtpauth" : a feature that authenticates trusted users for submitting email to postfix. See !sasl.[23:30:14] <lunaphyte_> mynetworks should be empty.[23:30:15] <seekwill> matadore: Your cluster solution isn't going to address your problem. You mentioned auth, which makes me assume transactional messages. Those messages SHOULD get in without deferrals. If you do get deferrals, Yahoo can be doing that to everyone, so just wait.[23:30:47] <matadore> seekwill: if i got 100k yahoos they need to go out... i think i found it.. sender dependent auth[23:30:50] <seekwill> matadore: If you're sending bulk messages, load balance across several IPs[23:31:07] <seekwill> So you're bulk sending?[23:31:20] <seekwill> Yeah, again, you definitely don't want to retry using different servers.[23:31:22] <matadore> seekwill: yea am load balancing by levels... level 1 relays to level 2 relays to level 3.[23:31:33] <seekwill> Spammy :)[23:32:02] <seekwill> oh[23:32:03] <matadore> http://www.geakeit.co.uk/wp-content/uploads/2010/12/deliverability-pyramid1.jpg[23:32:03] <matadore> :P[23:32:18] <matadore> http://www.geakeit.co.uk/2010/12/15/performance-tuning-postfix-for-sending-bulk-email/[23:32:22] <seekwill> Wait, I thought it was the other way[23:32:41] <seekwill> I can't read that now, but thanks for the link[23:33:07] <matadore> cool if you take a look at the pyramid thats basically what am trying to do[23:33:19] <adaptr> matadore: which idiot told you that ?[23:33:38] <seekwill> oh...[23:33:42] <seekwill> This doesn't sound good at all[23:34:20] <seekwill> I would suggest not taking this guy's advice..[23:34:46] <matadore> seekwill: what do you suggest then[23:34:48] <adaptr> "fallback_array" ?[23:34:54] <adaptr> nslookup ?[23:34:57] <lunaphyte_> Host www.geakeit.co.uk not found: 3(NXDOMAIN)[23:34:58] <seekwill> matadore: How much volume are you talking about?[23:35:27] <matadore> seekwill: 300k every so often[23:35:37] <seekwill> willmbp:~ will$ dig +short www.geakeit.co.uk[23:35:38] <seekwill> 109.169.74.168[23:35:46] <lunaphyte_> hopefully none, since even basic dns seems to be a challenge.[23:35:48] <seekwill> lunaphyte_: Are you using Level3 Public DNS???[23:36:17] <seekwill> matadore: How often is that?[23:36:25] <matadore> seekwill: every 3-4 days[23:36:51] <seekwill> matadore: Commercial, or non-profit all-volunteer-run organization?[23:37:51] <matadore> seekwill: what are you mail police? if i say commercial, you would say get a enterprise system.. if i say non profit.. whats the budget method[23:38:03] <seekwill> Nevermind[23:38:06] <matadore> seekwill: this will be for different projects[23:38:25] <adaptr> matadore: lose the defensive attitude. if you think our advice is unhelpful, go elsewhere.[23:38:59] <matadore> am defensive because i know how this operates... if someone asks about bulk, they get scrutinize and label spamma[23:39:26] <lunaphyte_> and how will you being defensive change that in any way?[23:40:07] <matadore> probably wont but would save some fluff talk[23:40:29] <seekwill> High volume email (transactional and bulk) is sort of my speciality... but who cares![23:41:48] *** twobitsprite has left #postfix[23:43:15] <adaptr> not us, we just love you because you're furry[23:44:46] <matadore> Is there any particular reading that you guys recommend for a bulk system if this pyrami fallback model is the wrong way? am already running multiple instances, load balancing and throttling per domains[23:44:52] *** e-anima has quit IRC[23:45:08] <adaptr> how soon do you want to send 300K messages ? one hour ?[23:45:12] <adaptr> does it matter ?[23:45:22] <adaptr> it's not a particularly large volume[23:45:29] <matadore> adaptr: 1 hr or so yes[23:45:57] <adaptr> so... 100/sec or so. a single postfix will do that sleeping[23:45:58] <matadore> or atleast a scalable system.. i might be doing bigger systems so i just want to get my method correct and make it scalable from 100k to 500k whatever[23:46:39] <adaptr> have you tested any of this ? do you have hard data if you need "load balancers" and multiple instances ?[23:46:42] <matadore> adaptr: yea but if most are a certain domain.. lets say yahoo 60/hr .. its gonna take a while[23:47:03] <adaptr> if you send volume to the big players, you need to get whitelisted[23:47:10] <adaptr> or suffer delays[23:47:34] <matadore> k so whitelisting is a must if bulk ok[23:49:31] <matadore> thanks alot fellas, like arnold "Ill be back" :)[23:53:40] <Corey> thumbs: Only three files? :-)[23:53:43] <Corey> That's either too many or too few.[23:53:58] <thumbs> Corey: it's a stupid question.[23:54:00] <Corey> The httpd.conf, the .htaccess file, and then you've got potentially php dependancies, etc.[23:54:03] <Corey> Oh, most definitely.[23:54:34] *** Section1 has quit IRC[23:54:56] <Corey> thumbs: Good call.[23:55:04] <thumbs> yeah, let's stop this[23:55:15] <Corey> thumbs: I'd also never want to work for a place that used THAT as a screening question.[23:55:23] <thumbs> agreed[23:55:41] <thumbs> Corey: as for your answer - it was incomplete, too.[23:55:49] <seekwill> thumbs: WORK FOR US![23:55:58] <Corey> thumbs: Of course, because I don't bloody know what the "right" answer to this is. :-)[23:56:03] <Corey> I mean hell, robots.txt could qualify.[23:56:10] <Corey> index.html as well.[23:56:18] <Corey> Any includes in httpd.conf[23:56:31] <thumbs> rewritemaps[23:56:37] <thumbs> rewriterules[23:56:47] <thumbs> redirects[23:56:58] <Corey> thumbs: Right. That's a lot more than three.[23:57:06] <thumbs> handler[23:57:11] <thumbs> external filter[23:57:39] <thumbs> but I'm gonna stop here.[23:57:45] <thumbs> seekwill: who's us?[23:58:04] <seekwill> US[23:58:12] <seekwill> Govt