I am using Yandex and OpenDNS resolvers as my DNS filtering and for each I have a caching server.
I have seen many times that the result is a timeout like:
2021/03/16 22:48:46 Error in request ACL script for //www.gravatar.com: GoError: read udp 127.0.0.1:54495->127.0.0.1:53: i/o timeout
... other timeouts
2021/03/16 22:56:06 Error in request ACL script for //mx-hadoop.s3.ap-northeast-2.amazonaws.com: GoError: read udp 127.0.0.1:49580->127.0.0.1:53: i/o timeout
The github.com/miekg/dns has an option to use exchange with context which can be added with a reasonable timeout like 10-15 seconds for specific cases.
One of these cases is low ttl and another one can be a retry for a dns lookup in case of timeout error.
I am testing the first option now.
I am using Yandex and OpenDNS resolvers as my DNS filtering and for each I have a caching server.
I have seen many times that the result is a timeout like:
The
github.com/miekg/dnshas an option to use exchange with context which can be added with a reasonable timeout like 10-15 seconds for specific cases.One of these cases is low ttl and another one can be a retry for a dns lookup in case of timeout error.
I am testing the first option now.