Package: pdns Version: 2.9.20-8 Severity: grave Justification: causes non-serious data loss
When a (super)master nameserver sends NOTIFY-packets to it's slave nameserver(s), those will queue an AXFR for the modified zone. However, if this AXFR fails (for example, because of a master nameserver getting overloaded with AXFR-requests when a lot of domains are updated at once), the slave nameserver won't retry the AXFR. In normal situations, this behaviour is no big issue: a few minutes/hours later the slaves will notice that their serials are out of date and an AXFR will be issued. However, if a supermaster notices his slaves about a /new/ zone and the initial AXFR fails, the slave nameservers will forget about the new zone and the nameserver configuration for this zone will become incomplete, which can cause very weird problems (including data loss). I think the slave nameservers should requeue an AXFR request after a failure and retry it after a few seconds/minutes. -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18-3-xen-686 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Versions of packages pdns depends on: ii pdns-recursor 3.1.4-1 PowerDNS recursor ii pdns-server 2.9.20-8 extremely powerful and versatile n pdns recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
