OpenShift Container Platform 4 Master Nodes not always getting the same IP after restart when using dhcpd

Solution Verified - Updated -

Issue

  • Our OpenShift Container Platform 4 Master Nodes running Red Hat CoreOS are regularly getting a new IP assigned by the DHCP server (dhcpd).
  • The lease file of the DHCP server shows the same MAC address but a different "UID" for each entry:

    $ cat /var/lib/dhcpd/dhcpd.leases
    [..]
    lease 10.228.108.221 {
      starts 5 2020/10/30 14:42:33;
      ends 5 2020/10/30 14:52:33;
      tstp 5 2020/10/30 14:52:33;
      cltt 5 2020/10/30 14:42:33;
      binding state free;
      hardware ethernet 00:50:56:ab:bd:ef;
      uid "\377V\224\275\270\000\004/Q\343\243\025<E\026\211\244\022`\261Y4)";
    }
    lease 10.228.108.226 {
      starts 1 2021/01/11 14:30:37;
      ends 1 2021/01/11 14:40:37;
      tstp 1 2021/01/11 14:40:37;
      cltt 1 2021/01/11 14:30:37;
      binding state free;
      hardware ethernet 00:50:56:ab:bd:ef;
      uid "\001\000PV\224\275\270";
    }
    

Environment

  • Red Hat OpenShift Container Platform (OCP) 4
  • DHCP server running dhcpd

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In