On Wednesday, September 21, 2016 at 8:10:16 AM UTC-7, Graham Clinch wrote: > >>> I have a DNS server (which is both forwarder and authoritative NS) and I > >>> see this odd behavior locally on the host: > >>> > >>> dig @localhost <name> # returns immediately with right response > >>> > >>> dig @<host-local-ip-on-eth0> <name> # returns sometimes, timesout most > >>> of the time > > [...] > > during this behavior, I saw lots of UDP packet loss on the host: > > > > netstat -s | egrep -A4 "Udp:" > > ... > > .. > > > > > > I tried similar local queries when traffic reduced (and when UDP pkt loss > > was zero) and both local queries succeeded. > > Which version of Bind are you running? This sounds like an issue I've > seen with prefetch in 9.10 before 9.10.4. > > https://kb.isc.org/article/AA-01315/0/prefetch-performance-in-BIND-9.10.html > > Graham
BIND 9.8.x. This behavior is seen when there are flood of NXDOMAIN queries sent to our nameserver running BIND 9.8.x As a short-term fix, we have added our nameservers behind a netscalar VIP and provided additional capacity to hadoop applications. I hope I'll have some time soon to get to the bottom of this problem (or just upgrade to BIND 9.10.x ? :) ) thanks Blr _______________________________________________ Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe from this list bind-users mailing list bind-users@lists.isc.org https://lists.isc.org/mailman/listinfo/bind-users
