| |
| |
| |
|
Page: 1 2 3
Comments:
<0> Remember, a virtual alias MUST point to a real mailbox.
<1> rob0: yes, and it points to two real mailboxen
<0> ok, humor me, add the address to your virtual_mailbox_maps and try again. :)
<1> rob0: lo and behold, it delivered the mail to foo@foxybanana.com's Maildir and completely ignored the alias
<0> virtual_alias_maps = pgsql:/etc/postfix/pgsql_aliases.cf
<0> and your virtual_alias_maps returns the right result?
<1> you mean when using postmap -q?
<0> yes
<1> yes
<2> it was a bad test
<2> it invokes that wonky alias again
<0> postconf.5.html#receive_override_options
<2> my fault
<1> haroldp: why is that a wonky alias? because it has two recipients or because it points to a mailbox named the same?
<2> cause it 1) points to itself and also 2) has a second address
<2> either of those things along should be ok
<2> dunno how it will behave in the presence of both
<0> MysticOne: receive_override_options is the problem, see the documentation.
<1> haroldp: the explanation I had before (From someone in this channel, I don't remember who) was that aliases and mailboxen were considered separately, and after the alias was matched it would then compare the recipient mailboxen and go where it was supposed to go
<1> rob0: THAT fixed it!
<3> now...
<1> ohhhh...
<1> I know what it was
<3> Does anyone know if it's possible to integrate a pre-processing mail formatter like formail in postfix before delivery or on delivery ?
<0> Rikardo: maybe a pipe(8) transport?
<1> okay, here's what the problem is ... and it's something that I forgot about
<3> hum, well, not quite rob0
<1> originally, on the old box, I had amavis set up as a content filter, but on the new box I didn't set that up because I wanted to avoid it
<4> hello
<3> because pipe will invoke the process, however formail stays up until it finds a mail in which it can add the said header.
<4> i recive emails to my server... but when i like to read them ... i am not able to... they disapear after reciving... why is that ?
<3> or process the mail in the manner specified.
<1> without no_address_mappings, while using amavis, you get two copies of each message
<1> so I had to stop it from comparing the alias twice when using amavis
<2> see, this is why I don't believe in "it worked before". Cause it always worked before under totaly different circumstances, on a different box, different setup, etc.. :)
<1> but when there's no more amavis, then I guess that option prevents it from evaluating the aliases at all
<3> My main battle is appending charsets to mails. A whole lot of clients don't do it, so I have to force it here... :/
<0> sdk: I ate them ... sorry.
<0> Just a few bytes.
<1> haroldp: I totally forgot about originally having amavis, but I didn't think it would've made a difference with the aliases
<4> rob0 ok
<4> but it looks ok
<4> Feb 27 00:23:34 codeplanet amavis[31012]: (31012-09) P***ed CLEAN, [193.189.160.132] <15135273@users.siol.net> -> <info@rsteam.com>, Message-ID: <2395686.1140992228722.JavaMail.root@fep1>, Hits: 4.917, 3891 ms
<4> Feb 27 00:23:34 codeplanet postfix/lmtp[565]: 31D16274B5A: to=<info@rsteam.com>, relay=127.0.0.1[127.0.0.1], delay=4, status=sent (250 2.6.0 Ok, id=31012-09, from MTA: 250 Ok: queued as 31B75274B61)
<4> Feb 27 00:23:34 codeplanet postfix/qmgr[517]: 31D16274B5A: removed
<4> u see
<0> Rikardo: MAYBE a policy service, or even header_checks.
<4> ups sorry for flood
<3> rob0 header_checks doesn't just remove by regexp ?
<0> header_checks can PREPEND too.
<0> But, you'd have to match something in the message.
<0> sdk: your LMTP relay (Cyrus?) on localhost accepted that message.
<1> okay, thank you all, and now I shall go to the store and buy food so my wife can make dinner tonight
<4> rob0 well i get this
<4> New mail for admin@codeplanet has arrived:
<4> but when i check email
<4> there is nothing
<3> rob0 and I can... just negate
<0> <== not going to try to help with a Cyrus problem.
<3> By negating the regex to locate the specified header, I can match it :)
<4> ok
<4> rob0
<0> No, unfortunately negative matching is not possible.
<4> it works now
<3> it ain't ?
<3> :/
<3> sdk it's a mistery :)
<4> i don't know how is it possible that premissions of files in var/spool/mail changed ?
<0> Not if you're looking for a missing header.
<4> i mean they haven't been writable
<4> that was the problem
<3> I can explain that sdk
<4> Rikardo please do
<3> Your main problem is the hunk o' meat between the chair and the keyboard :>
<3> heh j/k
<0> Header checks cannot determine that a particular header is not there.
<0> For that you need a content filter.
<3> procmail might be able to do it.
<3> or maybe not.
<3> :)
<0> procmail can do it for local deliveries, but not for relay (outbound) mail.
<3> i don't need it for outbound.
<3> my outbound mail is correctly formatted. Just the inbound mail gets around without the correct charset causing my mail client to display it not so well.
<3> rob0 sorry to disturb you again, but, is it possible to negate regexes in procmail ? :)
<0> Not sure.
<3> yes it does.
<3> I've checked it against a few howto's.
<5> is a email sent as a reply any different as far as the server is concerned to a new mail?
<6> dan2003_: no
<6> dan2003_: typically just additional header lines like referencing tags
<5> i ask this becasue im getting 554 relay access denied erros when i try to reply to mail/ but if i send a new mail it goes fine/ and i find this a bit odd
<5> ive got two servers set up at two different domains and both are exhibting this behavior
<6> do you smtp auth?
<5> yes
<6> which client?
<5> sylpheed/kmail/outlook all the same
<6> your client does not auth if " 554 relay access denied" happens
<5> forward casues the same problem as well
<6> " 554 relay access denied" comes from your relay host or the final recipient host?
<5> relay host
<5> theres nothing in the destination logs
<6> run smtpd with -v
<6> looks like smtp does not happen
<0> nothing in logs?
<0> Wait, you're saying that the $relayhost is rejecting you? You would have to discuss that with whoever runs that MTA.
<5> i dont seem to have smtpd ;/ in the logs on the relay i get an authenticated tls conection followed by the relay access denied
<0> You run the $relayhost?
<5> yup/ both ends in thiss test case
<5> Feb 27 00:14:28 sqs postfix/smtpd[2961]: TLS connection established from cpc1-ware3-0-0-cust291.lutn.cable.ntl.com[86.13.233.36]: TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)
<5> Feb 27 00:14:29 sqs postfix/smtpd[2961]: NOQUEUE: reject: RCPT from cpc1-ware3-0-0-cust291.lutn.cable.ntl.com[86.13.233.36]: 554 <daniel.squires@i-mark.uk.com>: Relay access denied; from=<mango@sqs.homedns.org> to=<daniel.squires@i-mark.uk.com> proto=ESMTP helo=<localhost.localdomain>
<0> BASIC_CONFIGURATION_README.html#relay_from
<0> 86.13.233.36 is not in $mynetworks for sqs.
<5> so why am i able to send new mail? but not replys to?
<5> ie if i replt to mailfrom dan@i-maark.uk.com/ it goes fine
<5> sorry. if i send new it goes fine
<5> if i reply to that address it fails
<0> Show a lot of one of those going through.
<0> log
<5> 2 secs
<6> from log shown i do not see a " authenticated tls conection"
<6> just TLS
<6> but no AUTH
<6> "helo=<localhost.localdomain>" is likely to cause problems with foreign MTAs
<5> ive confused 2 problems here.. sorry.. this doesnt solve my problem.. but fromt this server the problem ws i hadnt got my login detilas for sending.. its frome the other server i knwo the problem exists.. just going to get some logs for both scenarios
<5> im lost now! it seems to be working between these two hosts! a freind came to me with the issue to start with/ and i couldnt believe wot he was telling me when he said new mails would go and replys wouldnt. so i tried it for my self and reproduced the problem. i cant now tho it would seem. i will need to check the destination addess in question which i cnat do right now.
<7> i've googled for a good arite-up in migrating from sendmail to postfix, but havent found something informative, could anyone point me in the right direction?
<0> !basic
<8> rob0: 'basic' : http://www.postfix.org/BASIC_CONFIGURATION_README.html : a good starting place for Postfix beginners, many common questions are answered here.
<7> danke
<7> thats a basic config redame, which i will read, i was hoping for leads to a document stating possible issues when moving from sendmail to postfix
<0> How experienced are you with sendmail? If very, I think you'll be pleasantly surprised at how easy Postfix is.
<9> anyone here have spam******in setup to work with Postfix? SA is letting through a lot of stuff that is obviously spam and not marking it.
<10> i use spam******in with amavisd-new
<10> i have postfix turn away mail based on helo checks and a dnsrbl that contains dhcp space, if it p***es those two tests spam******ins attacks it
<7> rob0, i'd say im fairly experienced with sendmail, but i'm not one to jump into something (sendmail teaches you that) :) and had read a nice migration article that was even based on a slackware system which unfortunately i didn't bookmark like an idiot and can't find now
<9> lockdown: yeah, I use amavisd-new too.
<11> i am finding the migration to cyrus more intimidating than the migration to postfix
<0> The one way in which sendmail is probably more powerful is in pre-queue content filtering.
<0> (But I get by on some heavily spammed domains without content filtering at all.)
<10> JeffH, train the bayenesian filter, you might want to check out rulesemporium.com and uribl.com
<7> rob0, this is actually for a friends box, and i tried introducing sendmail to her, but obviously postfix is easier to maintain
<9> Strykar: I've tried to train it and it doesn't seem to make a difference.
<12> how can I find out if there is a smart host setting from a filter?
<12> run postconf as a shell call?
<12> ok how about binding to the ethernet device using a local name so I can have seperate bindings for localhost and ethernet
<13> does anyone know what this means? : Feb 27 00:58:31 testy master[5995]: fatal: master_spawn: exec /usr/libexec/postfix/trivial-rewritebounce: No such file or directory
<14> /usr/libexec/postfix/trivial-rewritebounce: No such file or directory <-- pretty direct to me
<15> Hello. How can I allow anonymous connections from localhost but require SASL auth from everyone else?
<16> cilkay: It's a matter of order. First allow connections from localhost (permit_mynetworks or check_client_access) then use permit_sasl_authenticated
<15> ChrisH: I have: smtpd_recipient_restrictions = permit_mynetworks, permit_sasl_authenticated, reject_unauth_destination
<15> I'm trying to find out what "mynetworks" represents here
Return to
#postfix
or
Go to some related
logs:
extract_url
#python
perl url exists -php
dpkg --configure -a automatix error
qtparted opening as ntfs failed operation not supported
ERROR 1045 (28000): Access denied for user fresh install windows
gl117.deb
#iptables
red_flame
canadgroup
|
|