BIND 9.4-ESV-R2 acting as master and also allowing recursive queries.
It was just a disaster recovery exercise, so the load was extremely light.
It happened repeatedly at the time, but I could not duplicate the problem
on the busy production server.
Matus posted that it could be journal data, which is believable, but I did
not know the dumpdb was supposed to include history -- I thought it was
supposed to be a fully digested and cherent snapshot.
I have not had time to revist this, but I still do want to find out more.
Thanks.
--
Gordon A. Lang
----- Original Message -----
From: "Cathy Almond" <cat...@isc.org>
To: <bind-users@lists.isc.org>
Sent: Wednesday, September 01, 2010 6:52 AM
Subject: Re: discrepancy with rndc dumpdb -zones
Hi Gordon,
We've not seen this before (and it doesn't sound like anyone else has
either). What version of BIND is it? Has it reappeared since? Is this
a particularly heavily loaded/busy server? Does it have recursive cache
as well as authoritative zones?
Kind regards,
Cathy
Gordon A. Lang wrote:
After several successful "update delete ..." nsupdate sends to the master
DNS server, verified with dig, the "rndc dumpdb -zones" command produced
named_dump.db file still showing the deleted records. This was
repeatable
and persistent (over the half hour time period) until I performed a hard
restart of named.
Has anyone else seen this sort of thing?
Can anyone explain this?
_______________________________________________
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users
_______________________________________________
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users
