DNS Overview
DNS performs a crucial role in enabling user access
to network resources so that users need not remember IP addresses
and individual computers need not store a huge volume of domain
names mapped to IP addresses. DNS employs a client/server model;
a DNS server resolves a query for a DNS client by looking up the
domain in its cache and if necessary sending queries to other servers
until it can respond to the client with the corresponding IP address.
The DNS structure of domain names is hierarchical; the top-level
domain (TLD) in a domain name can be a generic TLD (gTLD): com,
edu, gov, int, mil, net, or org (gov and mil are for the United
States only) or a country code (ccTLD), such as au (Australia) or
us (United States). ccTLDs are generally reserved for countries
and dependent territories.
A fully qualified domain name (FQDN) includes at a minimum a
host name, a second-level domain, and a TLD to completely specify
the location of the host in the DNS structure. For example, www.paloaltonetworks.com
is an FQDN.
Wherever a Palo Alto Networks firewall uses an FQDN in the user
interface or CLI, the firewall must resolve that FQDN using DNS.
Depending on where the FQDN query originates, the firewall determines
which DNS settings to use to resolve the query.
A DNS record of an FQDN includes a time-to-live (TTL) value,
and by default the firewall refreshes each FQDN in its cache based
on that individual TTL provided the DNS server, as long as the TTL
is greater than or equal to the
Minimum FQDN Refresh
Time you configure on the firewall, or the default setting
of 30 seconds if you don’t configure a minimum. Refreshing an FQDN
based on its TTL value is especially helpful for securing access
to cloud platform services, which often require frequent FQDN refreshes
to ensure highly available services. For example, cloud environments
that support autoscaling depend on FQDN resolutions for dynamically
scaling services up and down, and fast resolutions of FQDNs are
critical in such time-sensitive environments.
By configuring a minimum FQDN refresh time, you limit how small
a TTL value the firewall honors. If your IP addresses don’t change
very often you may want to set a higher Minimum FQDN Refresh Time
so that the firewall doesn’t refresh entries unnecessarily. The
firewall uses the higher of the DNS TTL time and the configured
Minimum FQDN Refresh Time.
For example, two FQDNs have the following TTL values. The Minimum
FQDN Refresh Time overrides smaller (faster) TTL values.
| TTL | If Minimum FQDN Refresh = 26 | Actual Refresh Time |
The FQDN refresh timer starts when the firewall receives a DNS
response from the DNS server or DNS proxy object that is resolving
the FQDN.
Additionally, you can set a
stale timeout to
configure how long the firewall continues to use stale (expired)
FQDN resolutions in the event of an unreachable DNS Server. At the
end of the stale timeout period, if the DNS server is still unreachable,
the stale FQDN entries become unresolved (the firewall removes stale
FQDN entries).
The following firewall tasks are related to DNS:
Configure your firewall with at least one DNS server
so it can resolve hostnames. Configure primary and secondary DNS
servers or a DNS Proxy object that specifies such servers, as shown
in
Use Case 1: Firewall Requires DNS Resolution.
Enable
Passive
DNS Monitoring, which allows the firewall to automatically
share domain-to-IP address mappings based on your network traffic
with Palo Alto Networks. The Palo Alto Networks threat research
team uses this information to gain insight into malware propagation
and evasion techniques that abuse the DNS system.
Configure
an Interface as a DHCP Server. This enables the firewall
to act as a DHCP Server and sends DNS information to its DHCP clients
so the provisioned DHCP clients can reach their respective DNS servers.
