Ahhhh... finally an answer ! Thanks !
Yes, they are on a slave server. The contents of the files are that of a zone
file, but some are whole and some are incomplete.
Your explanation jives with what could be happening here. I know that
there have
been ocassions when customers need both name servers updated immediately (even
though prior to the change we shortened the TTLs), and I have deleted
the slave
servers zone files and rndc reloaded or restarted the daemon to force an
immediate update and thus re-creation of the zone files. It does seem like I
have far more temp files than I would have created with this behavior,
but this
would explain some of them for sure.
Since this server is slave to maybe 70 or 80 or so masters which are
managedc by
as many differnet admins, there is no telling what is happening at any given
time or during any given zone transfer.
...and I guess the consensus is that they can be safely deleted.
Thanks,
Mike
Quoting "Gregory K. Ruiz-Ade" <[EMAIL PROTECTED]>:
On Aug 3, 2006, at 6:18 PM, Joshua Penix wrote:
I'm sure that both find and tmpwatch can take care of the symptom,
but I suspect Mike was more interested in the cause. I'm
interested as well, since I have one machine which exhibits the
same behavior Mike is seeing andI don't believe it's something that
BIND typically does. Yes I could just clean up the symptoms, but
if the behavior is indicative of some sort of misconfiguration or
bug, I'd much prefer to fix that.
Having run multiple BIND servers in environments large and small, I
can safely say that these are left over from aborted zone transfers
(axfr) attempts. At least, that's the only time I've ever seen them
get generated.
If you are running a slave server and can afford a bit of downtime,
you'll see these files get created if you blow away the local copies
of the zones and re-start named. Zones are copied to temporary files
with those naming patterns until the entire axfr is complete and
verified. Then, the temp file is copied to the filename specified in
named.conf for that zone. If there is an error, and/or the axfr
never completes, it's possible for these temp files to be left
unreaped.
As proof of this theory, have you looked at the contents?
Also, this assumes that my memory of the beginning of this thread is
correct, and these files are on a slave server.
Gregory
making a couple assumptions
--
Gregory K. Ruiz-Ade <[EMAIL PROTECTED]>
OpenPGP Key ID: EAF4844B keyserver: pgpkeys.mit.edu
--
************************************************************
Michael J. McCafferty
Principal, Security Engineer
M5 Hosting
858-576-7325 Voice
http://www.m5hosting.com
************************************************************
--
KPLUG-List@kernel-panic.org
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
