[Pdns-announce] PowerDNS Recursor instability

bert hubert bert.hubert at netherlabs.nl
Thu Jan 28 21:26:12 UTC 2010


More information can be found on the dns-operations mailing list:
https://lists.dns-oarc.net/pipermail/dns-operations/2010-January/004891.html

It now appears that a third large DNS deployment saw a five-fold increase of
DNS traffic world wide at exactly the times mentioned in the posting to the
dns-operations mailing list.

Could PowerDNS Recursor operators with graphs check out if they saw anything
anomalous on the 28th of January 03:40:44 CET (28th of January 02:40:44 UTC,
27th of January 21:40:44 EST, plus or minus 2 seconds)

Thanks!

On Thu, Jan 28, 2010 at 05:38:59PM +0100, bert hubert wrote:
> To all PowerDNS Recursor users,
> 
> We've received two simultaneous reports of stability problems in the
> PowerDNS Recursor. There is reason to believe these problems (crashes) are
> caused by third parties.
> 
> Action needed
> -------------
> All operators are urged to keep a close eye on their installations, and
> possibly to make sure that a crashed recursor is restarted swiftly.
> 
> A way to do this is to run the Recursor like this:
> 
> # while true; do /usr/sbin/pdns_recursor --daemon=no; done
> 
> This command should preferably be run inside 'screen', or something else
> that preserves the console connection (nohup, for example).
> 
> Further things you can do
> -------------------------
> If you observe stability issues, please email me privately.
> 
> We will keep you updated as the situation develops.
> 
> Kind regards,
> 
> Bert Hubert



More information about the Pdns-announce mailing list