Networking not found after reboot of VIDM cluster

Products

• VMware Aria Suite

Issue/Introduction

• Issue can occur after a reboot the cluster as part of the process of addressing an issue where the root partition full as per kb vIDM root partition is full on one node of the cluster 
• When viewing the VIDM nodes in vCenter web session a red banner is displayed stating: "NO NETWORKING DETECTED. PLEASE LOGIN AND RUN THE COMMAND..."
• Running the command systemctl status systemd-networkd.service shows the error /etc/systemd/network/10-eth0.network:9: Failed to parse dns server address, ignoring: ##.###.###.##,##.###.###.##

Environment

VIDM 3.3.x

Cause

The  DNS entries are in the format of DNS = ##.###.###.##,##.###.###.## in /etc/systemd/network/10-eth0.network file . When each DNS entry should be on different line in this file.  An example of the /etc/systemd/network/10-eth0.network where the DNS is incorrectly on the same line is below:

[Match]

Name=eth0

[Network]

DHCP=no

Address=##.###.###.##/24

Gateway=##.###.###.###

DNS=##.###.###.##,##.###.###.##

LinkLocalAddressing=no

IPv6AcceptRA=no

 

Resolution

Note: It is always recommended to take snapshots before making changes as per kb Snapshot Creation in VMware Aria Operations

1. Sign into ssh session to primary node as root user
2. Take a backup of the file cp /etc/systemd/network/10-eth0.network /etc/systemd/network/10-eth0.network.bckup 
3. Edit the file vi /etc/systemd/network/10-eth0.network
4. Update the file so that the DNS entries are on different lines resembling the following structure

   [Match]
   
   Name=eth0
   
   [Network]
   
   DHCP=no
   
   Address=##.###.###.##/24
   
   Gateway=##.###.###.###
   
   DNS=##.###.###.##
   DNS= ##.###.###.##
   
   LinkLocalAddressing=no
   
   IPv6AcceptRA=no

5. Restart the network services systemctl restart systemd-networkd
6. Repeat the steps for each VIDM node in the cluster

Note: If you are still experiencing the red banner NO NETWORKING DETECTED. PLEASE LOGIN AND RUN THE COMMAND.. when accessing the web console in vCenter, and you no longer receive the DNS parse error when running the command systemctl status systemd-networkd.service but you have the additional symptom where running the command ip a in the VIDM nodes on root ssh session you get <NO-CARRIER,BROADCAST,MULTICAST,UP> please review kb NSX UI becomes unavailable with Error code 101