Bind suddenly starts responding clients with servfail

[Søren Andersen]

Hi Greg,

I'm glad what I'm not the only one having this issue. Currently i've not more information that are not already mention in this mail thread.

But do you have a link to the ticket you have created?

/Søren
________________________________
From: Greg Rivers <gcr+bind-users at tharned.org>
Sent: Friday, May 8, 2020 04:03
To: Søren Andersen <SOAN at stofa.dk>
Cc: bind-users at lists.isc.org <bind-users at lists.isc.org>
Subject: Re: Bind suddenly starts responding clients with servfail

[EXTERNAL MAIL]


On Monday, 27 April 2020 03:59:39 CDT Søren Andersen wrote:
> I'm running a few BIND servers, but lately one of my servers suddenly starts
> responding to clients with servfail for every request from the clients, and
> BIND doesn't respond to the rndc or statistics interface anymore.
>
> My logs for client-channel show me this:
> 25-Apr-2020 21:52:04.501 client @XX XX.37#2921 (google.dk): no more
> recursive clients (1000/900/1000): quota reached
>
> I've removed all the dns traffic from the server, and the quota is still
> reached after 6+ hours?
>
> Do you guys have some clue what all this is about? - Or any suggestions
> where to look for any further information?
>
> I'm running BIND 9.16.1 on CentOS 7:
>
I've had the very same thing happen twice in the past two weeks on different production recursive servers running BIND 9.16.2 on FreeBSD. I've opened a ticket with ISC, and they are looking into it. Can you share any additional information that might aid troubleshooting?

If anyone else experiences this, please report it.

--
Greg


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/bind-users/attachments/20200508/084ae90e/attachment.htm>