From owner-freebsd-questions Fri Mar 14 2:13:22 2003 Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C273537B401 for ; Fri, 14 Mar 2003 02:13:16 -0800 (PST) Received: from gregale.wcom.co.uk (gregale.wcom.co.uk [193.131.254.138]) by mx1.FreeBSD.org (Postfix) with ESMTP id A8FE343F93 for ; Fri, 14 Mar 2003 02:13:15 -0800 (PST) (envelope-from Philip.Payne@uk.uu.net) Received: from cyclone ([170.127.79.100] helo=cyclone.wcom.co.uk) by gregale.wcom.co.uk with esmtp (Exim 4.12) id 18tmBT-0000c5-00; Fri, 14 Mar 2003 10:13:03 +0000 Received: from borasco (borasco [170.127.79.99]) by cyclone.wcom.co.uk (4.0.1.37) with ESMTP id KAA13538; Fri, 14 Mar 2003 10:12:50 GMT Received: from ukcamgate1.cbg.uk.corp.eu.uu.net ([62.191.1.65]) by borasco with esmtp (Exim 4.10.11) id 18tmB6-0000Tr-00; Fri, 14 Mar 2003 10:12:40 +0000 Received: by ukcamgate1.cbg.uk.corp.eu.uu.net with Internet Mail Service (5.5.2653.19) id ; Fri, 14 Mar 2003 10:13:04 -0000 Message-ID: <36D04A8168B2D41182250008C7E6F8780374F58F@ukcamexch2.cbg.uk.corp.eu.uu.net> From: Philip Payne To: "'Janine C.Buorditez'" Cc: questions@freebsd.org Subject: RE: [URGENT] Bad MX record; very bad. Date: Fri, 14 Mar 2003 10:13:13 -0000 X-Mailer: Internet Mail Service (5.5.2653.19) MIME-Version: 1.0 (Generated by NET-TEL Mailguard SMTP version 4.0.1.40) Content-Type: text/plain; charset="iso-8859-1" X-Spam-Score: -7.1 (-------) X-Scanner: exiscan for exim4 (http://duncanthrax.net/exiscan/) *18tmBT-0000c5-00*WNbqVwEh1wM* Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Hi, I don't know how much you know about DNS so if I aim too low then ignore me otherwise read on for a full explanation. I strongly suspect you're suffering from the fact that your old address is simply cached on various resolvers around the internet and you've just got to wait until it times out but I'll show how to check. > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> > $ORIGIN terrabionic.com. > $TTL 86400 > > @ IN SOA ns1.terrabionic.com. > johann.ninja.terrabionic.com. ( > 2003011901 ; Serial > 10800 ; Refresh > 3600 ; Retry > 604800 ; Expire > 86400 ) ; Minimum TTL > > IN A 213.187.181.68 > IN NS ns1.terrabionic.com. > IN NS ns2.terrabionic.com. > > www IN CNAME terrabionic.com. > > ninja IN A 213.187.181.68 > ninja6 IN AAAA 3ffe:4008:1b::1200 > ns1 IN A 213.187.181.68 > ns2 IN A 209.98.239.41 > > IN MX ninja.terrabionic.com. > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> You should have dig on your freebsd machine right so you can check this out. I notice on my own ISP's resolver that --- happyclown>dig @158.43.128.1 ninja.terrabionic.com a ; <<>> DiG 8.3 <<>> @158.43.128.1 ninja.terrabionic.com a ; (1 server found) ;; res options: init recurs defnam dnsrch ;; got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 4 ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0 ;; QUERY SECTION: ;; ninja.terrabionic.com, type = A, class = IN ;; ANSWER SECTION: ninja.terrabionic.com. 10h44m25s IN A 217.13.29.51 ;; Total query time: 1 msec ;; FROM: happyclown.cbg.eng.emea.uu.net to SERVER: 158.43.128.1 158.43.128.1 ;; WHEN: Fri Mar 14 09:40:18 2003 ;; MSG SIZE sent: 39 rcvd: 55 --- gives me a result that ninja is 217.13.29.51 (your old address) but the important thing to note is the 10h44m25s... this is the time left that this resolver will cache that record for. Until that time has passed this resolver simply will not bother even to check whether a new record exists. The way to check if the correct record will be propagated to this resolver when the record times out is to query the authoritative nameservers for the domain. You can tell what the authoritative nameservers are by doing: --- happyclown>dig @a.gtld-servers.net terrabionic.com ns ; <<>> DiG 8.3 <<>> @a.gtld-servers.net terrabionic.com ns ; (1 server found) ;; res options: init recurs defnam dnsrch ;; got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 4 ;; flags: qr rd; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 2 ;; QUERY SECTION: ;; terrabionic.com, type = NS, class = IN ;; ANSWER SECTION: terrabionic.com. 2D IN NS ns1.terrabionic.com. terrabionic.com. 2D IN NS ns2.terrabionic.com. ;; ADDITIONAL SECTION: ns1.terrabionic.com. 2D IN A 213.187.181.68 ns2.terrabionic.com. 2D IN A 209.98.239.41 ;; Total query time: 87 msec ;; FROM: happyclown.cbg.eng.emea.uu.net to SERVER: a.gtld-servers.net 192.5.6.30 ;; WHEN: Fri Mar 14 09:49:26 2003 ;; MSG SIZE sent: 33 rcvd: 101 --- So, ns1 & ns2 should be carrying records for terrabionic.com . If I query one of those name servers I should get an authoritative answer: --- happyclown>dig @213.187.181.68 terrabionic.com mx ; <<>> DiG 8.3 <<>> @213.187.181.68 terrabionic.com mx ; (1 server found) ;; res options: init recurs defnam dnsrch ;; got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 4 ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 3 ;; QUERY SECTION: ;; terrabionic.com, type = MX, class = IN ;; ANSWER SECTION: terrabionic.com. 10h39m14s IN MX 10 ninja.terrabionic.com. ;; AUTHORITY SECTION: terrabionic.com. 10h30m27s IN NS ns1.terrabionic.com. terrabionic.com. 10h30m27s IN NS ns2.terrabionic.com. ;; ADDITIONAL SECTION: ninja.terrabionic.com. 10h30m27s IN A 217.13.29.51 ns1.terrabionic.com. 10h30m27s IN A 217.13.29.51 ns2.terrabionic.com. 1d10h30m27s IN A 209.98.239.41 ;; Total query time: 63 msec ;; FROM: happyclown.cbg.eng.emea.uu.net to SERVER: 213.187.181.68 213.187.181.68 ;; WHEN: Fri Mar 14 09:54:17 2003 ;; MSG SIZE sent: 33 rcvd: 139 --- Notice the flags section... if this name server was carrying an actual zone for this domain you would get an "aa" flag. I don't so it's probably secondaried to this machine and unfortunately the ninja.terrabionic.com still reads as the old address.... so... solution: 1) Update the serial number in your zonefile so you ensure the secondary picks up the new zone and hopefully the new address for ninja as you seem to have this specified correctly in the zonefile from the email. 2) Also, the ordering of entries in your zonefile is bad, in it's original order you're specifying an mx record for ns2.terrabionic.com instead of the whole domain terrabionic.com so it should read: --- @ IN SOA ns1.terrabionic.com. johann.ninja.terrabionic.com. ( 2003011901 ; Serial 10800 ; Refresh 3600 ; Retry 604800 ; Expire 86400 ) ; Minimum TTL IN A 213.187.181.68 IN NS ns1.terrabionic.com. IN NS ns2.terrabionic.com. IN MX ninja.terrabionic.com. www IN CNAME terrabionic.com. ninja IN A 213.187.181.68 ninja6 IN AAAA 3ffe:4008:1b::1200 ns1 IN A 213.187.181.68 ns2 IN A 209.98.239.41 --- To avoid problems like this in future, the day or two before any serious changes like this, drop your TTL in the SOA record to 15 mins (900 secs). --- @ IN SOA ns1.terrabionic.com. johann.ninja.terrabionic.com. ( 2003011901 ; Serial 10800 ; Refresh 3600 ; Retry 604800 ; Expire 900 ) ; 15 min TTL --- This TTL controls how long name resolvers cache records for so, this new TTL time propagates around the net over the next 24 hours and starts the caches re-checking for records every 15 mins instead of every 24 hours... then when you make your important change, like updating your mailserver IP address, you reset the SOA back to 24 hours. This propagates quickly (due to the 15 mins you set the day or two before) but immediately resets the caches back to sensible behaviour. Hope that helps. Anyone else wants to comment feel free, I'm more familiar with DNS from BIND 8 than 9 so any changes in behaviour point it out. Phil. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message