Note: This is a beta release of Red Hat Bugzilla 5.0. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Also email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback here.
Bug 163525 - AD DNS records break named startup zone check
Summary: AD DNS records break named startup zone check
Alias: None
Product: Fedora
Classification: Fedora
Component: bind
Version: 4
Hardware: i386
OS: Linux
Target Milestone: ---
Assignee: Jason Vas Dias
QA Contact: Ben Levenson
Depends On:
TreeView+ depends on / blocked
Reported: 2005-07-18 16:05 UTC by James J. Moore
Modified: 2007-11-30 22:11 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2005-07-18 17:18:39 UTC

Attachments (Terms of Use)

Description James J. Moore 2005-07-18 16:05:04 UTC
Description of problem: 
   The 'named' daemon fails to start when MS AD records of the form 
'' have been added to a master forward zone file.  
During the run of the initscript, named-checkconf returns a non-zero error 
code when checking forward zones having MS Active Directory gc._msdcs 'A' 
records.  The named initscript does not give one the choice of passing up this 
name check, even if 'check-names ignore' is set in the named.conf file.  As a 
result, the initscript bails before attempting to start named.  
Version-Release number of selected component (if applicable): 
How reproducible: 
Steps to Reproduce: 
1.  Allow MS AD Domain Controller to add dynamic DNS records to zonefile 
mastered by BIND. 
2.  Restart BIND on the primary DNS server for that zone: /sbin/service named 
Actual results: 
Starting named: 
Error in named configuration: 
zone loaded serial 2 
data/db.thebank:96: bad owner name (check-names) 
data/db.thebank:97: bad owner name (check-names) 
data/db.thebank:98: bad owner name (check-names) 
data/db.thebank:99: bad owner name (check-names) 
zone loaded serial 19564 
data/db.ciassociates:27: bad owner name 
zone loaded serial 160 
zone loaded serial 4 
data/db.cnbpost:27: bad owner name (check-names) 
zone loaded serial 23 
[ snipped ] 
Expected results: 
   BIND starts normally 
Additional info:

Comment 1 James J. Moore 2005-07-18 16:08:41 UTC
Never mind this one!  Turns out that an unrelated permissions error was really 
causing the server to crash.  Once that was eliminated, it started up fine. 

Comment 2 Jason Vas Dias 2005-07-18 17:18:39 UTC
Yes, to support DDNS updateable zones, you need to either locate the zone
files in the 'slaves/' subdirectory, or 
  # chown named:named $ROOTDIR/var/named; chmod g+w $ROOTDIR/var/named
and, if SELinux is enabled in Enforcing mode:
  # setsebool named_write_master_zones 1
(see man named_selinux) .

The errors messages you see above are from the bind-9.3.1 "check-names" code -
perhaps because the files were not correctly updated because of the permissions
problem ? 
But bind-9.3.1 has new behaviour with name checking, that you can configure with
the named.conf 'options{... check-names ... }' option:
  check-names ( master | slave | response )( warn | fail | ignore ); ]
(see the BIND9 Administrators Reference Manual:

Note You need to log in before you can comment on or make changes to this bug.