Red Hat Bugzilla – Bug 923113
Potential crash during startup/reload if global configuration options are set
Last modified: 2013-11-21 07:11:05 EST
This bug is created as a clone of upstream ticket: https://fedorahosted.org/bind-dyndb-ldap/ticket/110 * Define some global DNS option, e.g. `idnsAllowSyncPTR` attribute in global configuration object `cn=dns, dc=example, dc=com`. * Start or reload BIND. * Keep your fingers crossed, this problem not so often on RHEL 6.4. * Check logs. The error is more likely in latest builds from master (e1122fde894946d5cf4aa209df4c7060bb877d0a). `/var/log/messages`: {{{ 13-Mar-2013 19:50:06.718 examining set of settings 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 setting 'dyn_update' was found in set 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 examining set of settings 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 setting 'dyn_update' was found in set 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 [settings.c : 419: setting_update_from_ldap_entry] check failed: ignore 13-Mar-2013 19:50:06.718 zone 168.192.IN-ADDR.ARPA/IN: loaded; checking validity 13-Mar-2013 19:50:06.718 examining set of settings 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 setting 'sync_ptr' was found in set 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 examining set of settings 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 setting 'sync_ptr' was found in set 'LDAP idnsConfig object for database ipa' 13-Mar-2013 19:50:06.718 task.c:1678: REQUIRE(task->state == task_state_running) failed, back trace 13-Mar-2013 19:50:06.718 exiting (due to assertion failure) }}}
Workaround: Remove all global configuration options from LDAP. IPA command: $ ipa dnsconfig-mod --forwarder='' --forward-policy='' --allow-sync-ptr='' --zone-refresh='' Global forwarders and zone refresh parameters can be configured per-server in /etc/named.conf. Zone refresh was superseded by persistent search in IPA version 3.0 and higher and should not be enabled. 'Sync PTR' feature can be configured per server in /etc/named.conf and also per-zone in LDAP (via ipa dnszone-mod).
BIND will crash if following conditions are met at the same time: * Persistent search is enabled (default from IPA 3.0) * Configuration of PTR synchronization feature has different value in global configuration and in local /etc/named.conf. I.e. PTR synchronization is globally enabled and locally disabled or vice versa. Simpler workaround: * Configure PTR synchronization in /etc/named.conf or per-zone as described in comment #1. * Remove configuration for PTR synchronization from global configuration: $ ipa dnsconfig-mod --allow-sync-ptr='' OR * Disable persistent search.
This issue is fixed by http://rhn.redhat.com/errata/RHBA-2013-0739.html . Why is the status still MODIFIED?
(In reply to comment #15) > This issue is fixed by http://rhn.redhat.com/errata/RHBA-2013-0739.html . > Why is the status still MODIFIED? Because the bug was not added to the list of bugs fixed in an errata? If this is the case, we should close this bug&fill fixed in field.
Ah, I see it now, this Bugzilla is for 6.5 - but I still think that Fixed In Version should be filled.
Tested using: ipa-server-3.0.0-33.el6.x86_64 bind-dyndb-ldap-2.3-4.el6.x86_64 # ipa dnsconfig-mod --allow-sync-ptr=TRUE Allow PTR sync: TRUE # service named restart Stopping named: .[ OK ] Starting named: [ OK ] Not seeing the behaviour reported above. From https://bugzilla.redhat.com/show_bug.cgi?id=928429#c10, verifying Sanity Only.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1636.html