Named will not start

[racker_mmontgom]

You need to remember that all the paths referenced in the error messages are relative to /var/named/run-root/.

You should have /var/named/run-root/etc/rndc.key in place. If not copy /etc/rndc.key into it's place. Also make sure that the key string found in /etc/named.conf matches that found in the rndc.key file

You should have /etc/rndc.key as a symlink to /var/named/run-root/etc/rndc.key

Matthew Montgomery

 

[wmgilligan]

Did this... still fails - but now - with no error messages!


Starting named: [FAILED]

 

[wmgilligan]

from messages log
====================
Aug 16 17:45:05 server1 named[11990]: starting BIND 9.2.5 -u named -c /etc/named.conf -u named -t /var/named/run-root
Aug 16 17:45:05 server1 named[11990]: using 1 CPU
Aug 16 17:45:05 server1 named[11990]: loading configuration from '/etc/named.conf'
Aug 16 17:45:05 server1 named[11990]: listening on IPv4 interface lo, 127.0.0.1#53
Aug 16 17:45:05 server1 named[11990]: listening on IPv4 interface eth0, 72.51.42.84#53
Aug 16 17:45:05 server1 named[11990]: could not configure root hints from 'named.ca': file not found
Aug 16 17:45:05 server1 named[11990]: loading configuration: file not found
Aug 16 17:45:05 server1 named[11990]: exiting (due to fatal error)
Aug 16 17:45:05 server1 named: named startup failed

 

[wmgilligan]

Got it running!
copied the missing file to the proper directory.
Now - it starts fine from command line - but Plesk bombs out with:

0: /usr/local/psa/admin/htdocs/server/restart_services.php:28 psaerror(string "Unable to make action: Unable to manage service by dnsmng: dnsmng: Service named failed to start")
========================
And the log files says:

Aug 16 17:51:40 server1 named: named startup succeeded
Aug 16 17:51:40 server1 named[12468]: running
Aug 16 17:51:40 server1 named[12468]: invalid command from 127.0.0.1#33672: bad auth
Aug 16 17:51:41 server1 named[12468]: invalid command from 127.0.0.1#33673: bad auth
Aug 16 17:51:42 server1 named[12468]: invalid command from 127.0.0.1#33674: bad auth

 

[racker_mmontgom]

This message states the problem...

Aug 16 17:45:05 server1 named[11990]: could not configure root hints from 'named.ca': file not found

Make sure you have a named.ca and other related files in /var/named/run-root

You should have most if not all of the following:

etc/localtime
etc/named.conf
etc/named.conf.default
etc/named.conf.lock
etc/named.conf.upgrader_saved
etc/rndc.conf
etc/rndc.conf.sample
var/named.broadcast
var/named.ca
var/named.ip6.local
var/named.local
var/named.root
var/named.zero
var/run/named/

 

[racker_mmontgom]

The error plesk is having has to do with not having the correct value for the rndc key. Make sure that your named.conf or rndc.conf matches the value of the key in rndc.key

Make sure that /etc/rndc.conf is also a symlink to /var/named/run-root/etc/rndc.conf

 

[wmgilligan]

THNK YOU THANK UYOU THANK YOU!

Got it running!

Had to match the rndc key, and the name as well!
But all good now.

One last question - asked by a previous poster - should I make sure up2date does not update this again? If so - how to do that?

Thanks again !

 

[racker_mmontgom]

If bind-chroot isn't already installed it should not be upgraded my up2date. To ensure 100% that it will not be do:

# up2date --config

up2date will prompt you for which config item to modify. Look for the number of "pkgSkipList", for RHEL3 it's 20, but yours may be different.

Type each package group to skip separated by a semi-colon " ; ". You will need to explicitly list any packages that previously exist in that field.


Example:

Current value:
20, pkgSkipList ['kernel*', 'spamassassin*']

New value:
kernel*;spamassassin*;bind-chroot*;

Matt Montgomery

 

[pixoul]

Re: bind-chroot

Originally posted by racker_mmontgom
You should be good to go at that point.

Matt Montgomery
Rackspace Managed Hosting

You rock, thanks!

 

[dtgorm]

Ok, I just had an OS reload performed and plesk 8.0

This morning, looks like yum updated and in there i seen bind updated also. Now all of a sudden, named no longer runs and will not start.. i also have the problem with plesk having the same error with

psaerror(string 'Unable to make action: Unable to manage service by dnsmng: dnsmng: Service named failed to start')

# /etc/rndc.key -> /var/named/chroot/etc/rndc.key

key "rndckey" {
algorithm hmac-md5;
secret "rGZdlnMebhlj08gHiR3txSY7QLFVC2WJpiWv86WQ3plenMVJpuDgFAmUS9Qw";
};


------------------
# rndc.conf -> /var/named/run-root/etc/rndc.conf
# Start of rndc.conf
key "rndc-key" {
algorithm hmac-md5;
secret "CeMgS23y0oWE20nyv0x40Q==";
};

options {
default-key "rndc-key";
default-server 127.0.0.1;
# default-port 953;
};
# End of rndc.conf

I have tried commenting out the key "rndckey" { and putting in key "rndc-key" { but still didnt seem to work.. any other ideas would be apprecaited.

Mar 10 14:23:01 ns1 named[22296]: starting BIND 9.2.4 -u named -c /etc/named.conf -u named -t /var/named/run-root
Mar 10 20:23:01 ns1 named[22296]: using 1 CPU
Mar 10 20:23:01 ns1 named[22296]: loading configuration from '/etc/named.conf'
Mar 10 20:23:01 ns1 named[22296]: listening on IPv4 interface lo, 127.0.0.1#53
Mar 10 20:23:01 ns1 named[22296]: listening on IPv4 interface eth0, 74.52.206.138#53
Mar 10 20:23:01 ns1 named[22296]: listening on IPv4 interface eth0:0, 74.52.206.139#53
Mar 10 20:23:01 ns1 named[22296]: listening on IPv4 interface eth0:1, 74.52.206.140#53
Mar 10 20:23:01 ns1 named[22296]: listening on IPv4 interface eth0:2, 74.52.206.141#53
Mar 10 20:23:01 ns1 named[22296]: listening on IPv4 interface eth0:3, 74.52.206.142#53
Mar 10 20:23:01 ns1 named[22296]: could not configure root hints from 'named.root': file not found
Mar 10 20:23:01 ns1 named[22296]: loading configuration: file not found
Mar 10 20:23:01 ns1 named[22296]: exiting (due to fatal error)
Mar 10 14:23:01 ns1 named: named startup failed