Fix Ansible \"Could Not Resolve Hostname\"

The Error

fatal: [myserver]: UNREACHABLE! => {
    "msg": "Failed to connect to the host via ssh: ssh: Could not resolve hostname myserver:
    Name or service not known"
}

This means the control node's DNS can't resolve the hostname in your inventory to an IP address.

Quick Diagnosis

# Test from the Ansible control node (not your laptop)
ssh myserver
# ssh: Could not resolve hostname myserver: Name or service not known

# Try ping
ping -c 1 myserver
# ping: myserver: Name or service not known

# Try with FQDN
ping -c 1 myserver.example.com
# PING myserver.example.com (192.168.1.10) 56(84) bytes of data.

If FQDN works but short name doesn't, the issue is DNS search domain configuration.

5 Solutions

Solution 1: Use ansible_host with IP Addresses (Most Reliable)

# inventory.ini
[webservers]
web1 ansible_host=192.168.1.10
web2 ansible_host=192.168.1.11
db1  ansible_host=192.168.1.20

# inventory.yml
all:
  children:
    webservers:
      hosts:
        web1:
          ansible_host: 192.168.1.10
        web2:
          ansible_host: 192.168.1.11

This decouples the inventory name from DNS — the most reliable approach.

Solution 2: Use FQDNs in Inventory

[webservers]
web1.example.com
web2.example.com

Solution 3: Add Entries to /etc/hosts

# On the Ansible control node
echo "192.168.1.10 web1 web1.example.com" | sudo tee -a /etc/hosts
echo "192.168.1.11 web2 web2.example.com" | sudo tee -a /etc/hosts

Or automate with Ansible (on a working host):

- name: Populate /etc/hosts
  ansible.builtin.lineinfile:
    path: /etc/hosts
    line: "{{ hostvars[item].ansible_host }} {{ item }}"
    state: present
  loop: "{{ groups['all'] }}"
  when: hostvars[item].ansible_host is defined

Solution 4: Fix DNS Configuration

# Check current DNS server
cat /etc/resolv.conf
# nameserver 8.8.8.8

# Test DNS resolution
nslookup myserver
dig myserver
host myserver

# Test with specific DNS server
dig @10.0.0.1 myserver.example.com

# Add search domain
echo "search example.com" | sudo tee -a /etc/resolv.conf

Solution 5: Fix DNS Search Domain

If short hostnames should resolve via a domain suffix:

# /etc/resolv.conf
nameserver 10.0.0.1
search example.com internal.company.com

Now myserver resolves as myserver.example.com.

Common Causes

Cause	Symptom	Fix
Typo in inventory	One host fails	Check spelling
DNS server down	All hosts fail	Check `/etc/resolv.conf`
VPN not connected	Private hosts fail	Connect VPN first
Wrong inventory file	Wrong hosts targeted	`ansible-playbook -i correct_inventory`
Cloud instance rebuilt	IP changed	Update `ansible_host`
Split DNS (VPN)	Internal names fail	Check DNS over VPN

Debugging Steps

1. Verify Which Inventory Ansible Uses

ansible-config dump | grep DEFAULT_HOST_LIST
# DEFAULT_HOST_LIST = ['/etc/ansible/hosts']

# Override with -i
ansible all -i my_inventory.ini --list-hosts

2. Test SSH Directly

# Test with the exact command Ansible would use
ssh -o StrictHostKeyChecking=no -o ConnectTimeout=10 deploy@web1

3. Increase Verbosity

ansible web1 -m ping -i inventory -vvvv
# Shows the exact SSH command and DNS resolution attempt

4. Check for Stale DNS Cache

# Flush DNS cache
# systemd-resolved
sudo resolvectl flush-caches

# nscd
sudo nscd -i hosts

# dnsmasq
sudo systemctl restart dnsmasq

VPN and Network Issues

# Check if VPN is connected
ip route | grep tun
ip route | grep vpn

# Check if you can reach the network
ping -c 1 192.168.1.1  # Gateway

# Check firewall isn't blocking DNS
sudo ss -ulnp | grep :53

Ansible Configuration Options

# ansible.cfg
[defaults]
# Timeout for SSH connection
timeout = 30

# Retry failed hosts
retries = 3

[ssh_connection]
# SSH timeout
ssh_args = -o ConnectTimeout=30 -o ServerAliveInterval=15

Conclusion

"Could not resolve hostname" means DNS lookup failed on the control node. The most reliable fix is using ansible_host with IP addresses in your inventory. For DNS-based resolution, verify /etc/resolv.conf, check VPN connectivity for private networks, and test with dig/nslookup from the control node. Use ansible -vvvv to see the exact SSH command and pinpoint where resolution fails.