applying c.k. davis qmail buffer patch to live qmail server
Hello --
I need to apply Christopher K. Davis' qmail buffer patch to my live production qmail server. (to solve the CNAME lookup failed temporarily problem) What are the steps I need to take ? This is a RH9 box, with a qmailrocks install (more or less). I imagine I need to stop some processes, apply the patch, recompile parts or all of qmail, and restart some processes... Anything more precise than that would be greatly appreciated, All best, --Sam |
Actually, it seems that this patch is already part of a patch pack that I installed originally with qmailrocks. So the problem is : why am I getting all this
delivery 19608: deferral: CNAME_lookup_failed_temporarily._(#4.4.3)/ for all my messages output by ezmlm ? --Sam |
That looks like a DNS error of some sort. The machine that is trying to send can't find the CNAME it is looking up. I would use the tool "dig" to see if the name has a resolution issue. If the domain you are trying to send to is named example.com, then this command will lookup the mailserver's name and IP -
Code:
dig -t mx example.com By the way, running anything on RH9 is an extremely poor choice. That OS is ancient, and hasn't been supported in 2+ years. All the software it installs is just as old. If you like Red Hat, then move up to Red Hat Enterprise 4 or 5, or get the free versions of it, CentOS or Whitebox. Fedora Core 6 is also popular. I doubt that is the source of your errors, but it certainly doesn't help any. Peace, JimBass |
Thank you for your reply.
dig doesn't seem to return consistent results : [root@yoruban qmail-send]# dig murez.com ; <<>> DiG 9.2.1 <<>> murez.com ;; global options: printcmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 7372 ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1 ;; QUESTION SECTION: ;pams.com. IN MX ;; ANSWER SECTION: pams.com. 7200 IN MX 10 MAIL.pams.com. ;; ADDITIONAL SECTION: MAIL.HEADSWELL.com. 2970 IN A 66.119.206.14 ;; Query time: 6 msec ;; SERVER: 192.168.0.1#53(192.168.0.1) ;; WHEN: Sun Apr 22 20:25:07 2007 ;; MSG SIZE rcvd: 78 [root@yoruban qmail-send]# dig murez.com ; <<>> DiG 9.2.1 <<>> murez.com ;; global options: printcmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 28858 ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0 ;; QUESTION SECTION: ;murez.com. IN A ;; ANSWER SECTION: murez.com. 172800 IN A 205.147.48.4 ;; Query time: 287 msec ;; SERVER: 192.168.0.1#53(192.168.0.1) ;; WHEN: Sun Apr 22 20:25:42 2007 ;; MSG SIZE rcvd: 43 that seems a bit weird. --Sam |
I get consistent results. You got different results because you didn't specify the -t mx part of the command. The first time you asked for a mail exchanger (mx), the second you asked for an A record (default).
Code:
jim@jimsworktop:~$ dig -t mx murez.com Code:
jim@ns2:~$ dig -t mx murez.com The output you gave included looking for things like ";; QUESTION SECTION: ;pams.com. IN MX", which makes no sense. You should not be asking for anything having to do with pams.com, as a trace shows the full path from the roots up to your mail exchangers: Code:
jim@jimsworktop:~$ dig -t mx murez.com +trace Peace, JimBass |
All times are GMT -5. The time now is 04:24 AM. |