Hi Frank, On Wed, 1 Jun 2022 12:48:01 +0200 "frank+p...@tembo.be" <frank+p...@tembo.be> wrote:
> Hi Jan, > > I completely understand NDAs and myself (and numerous other PowerDNS > Certified Consultants on this list) are happy to sign them, as part > of a professional engagement. Please reach out to me off-list to > discuss your options. > > However, this also means that on this list, we can't help you much... > > As per your questions: first we need to know what happens. The trace > should tell us. Options to look at: (yes this list is long and some > won't apply, but please reread the first sentence of this mail) > > * https://doc.powerdns.com/recursor/settings.html#network-timeout I'm getting the feeling this is the culprit. We looked at the nsspeeds the recursor reported and saw times between 0.6 and 1.2s for the root servers just after a restart of de recursor. I now get reports that the admins see 2s responce times in their environment, making the default of 1.5s timeout a tad close to the times observed in a clean environment and to low for reported times. We instructed to set the network-timeout to 3000 or even 5000. > * > https://doc.powerdns.com/recursor/settings.html#dont-throttle-netmasks We already set this to 0.0.0.0/0,::/0 and kept seeing the issues. The local admins are keeping an eye on the environment with the new settings and will report back. -- Jan Huijsmans b...@koffie.nu ... cannot activate /dev/brain, no response from main coffee server _______________________________________________ Pdns-users mailing list Pdns-users@mailman.powerdns.com https://mailman.powerdns.com/mailman/listinfo/pdns-users
