DNS lookup failures attribute to erroneous network traffic
Applies to
- StorageGRID
- NTP
Description
- StorageGRID nodes use the network time protocol (NTP) to synchronize the time within the grid. At least two servers at each site of the StorageGRID are specified as primary NTP time servers, one Admin Node, and a Main Gateway Node (not every Gateway Node should be set as a Primary NTP time source). These two servers must synchronize to at least 4 external "stratum one" time sources. All the other servers within StorageGRID will synchronize their time with the Primary NTP sources within the StorageGRID (that is, the two servers mentioned above).
- NTP, by design, will issue a reverse DNS lookup to resolve an IP address. A reverse DNS lookup is a query of the DNS for domain names when the IP address is known. DNS stores the IP addresses in the form of domain names as specially formatted names in the pointer (PTR) records within the infrastructure top-level domain.
- An issue exists where an NTP client will query the NTP sources listed in the
ntp.conf
file in the form of a reverse DNS lookup. If the DNS resolution is misconfigured, or the DNS PTR records are not properly configured, or it is just unable to resolve the addresses in thentp.conf
file (of a StorageGRID node), the NTP client will retry forever to resolve those addresses listed in thentp.conf
file. This adds needless network traffic over the wire and might bog down the DNS servers. This 'chatter' is a side effect of the way NTP clients resolve their time sources if DNS is not properly configured.