updates (how 2 stop?)

chris cariffe chriss at well.com
Wed Jul 7 21:06:02 UTC 1999


Well we just did this test and I saw it happen.  After about 10 or so
update failures in a row it stopped answering request and needed to be
stopped and started.  this is on an Auspex 4.1.4 sunos box.  any ideas?

-chris

On Wed, 7 Jul 1999, Jim Reid wrote:

> >>>>> "chris" == chris cariffe <chriss at well.com> writes:
> 
>     chris> That's what I see in the logs.  It isn't allowing updates
>     chris> but it is logging.  The logging is fine as it is a good
>     chris> security measure.  However I was told when this client came
>     chris> up and tried to do an update the BIND process on the server
>     chris> hung.  Is this a known problem or should I test this
>     chris> myself?  I didn't witness this myself but was told this by
>     chris> my client.
> 
> Your client is badly misinformed. A BIND8 name server does not hang if
> it gets an unapproved update packet. If it did, anyone could stop the
> Internet by sending unauthorised update packets to the root name
> servers. That would be an amazingly simple and really nasty denial of
> service attack. Perhaps it's the box that *sends* the unapproved
> update packet that hangs?
> 
> 
> 
> 



More information about the bind-users mailing list