DHCP issues can result in node boot failures. This procedure helps investigate and resolve such issues.
kubectl
is installed.Encryption of compute node logs is not enabled, so the passwords may be passed in clear text.
Log in to a non-compute node (NCN) as root.
Check that the DHCP service is running.
kubectl get pods -A | grep kea
Example output:
services cray-dhcp-kea-554698bb69-r9wwt 3/3 Running 0 13h
Start a tcpdump
session on the NCN.
The following example sends tcpdump
data to stdout
.
tcpdump
Obtain the DHCP pod’s ID.
PODID=$(kubectl get pods --no-headers -o wide | grep cray-dhcp | awk '{print $1}')
Enter the DHCP pod using its ID.
kubectl exec -it $PODID /bin/sh
Start a tcpdump
session from within the DHCP pod.
Open another terminal to perform the following tasks:
Issue a DHCP discover request from the NCN using nmap
.
Analyze the NCN tcpdump
data in order to ensure that the DHCP discover request is visible.
Go back to the original terminal to analyze the DHCP pod’s tcpdump
data in order to ensure that the DHCP discover request is visible inside the pod.
Troubleshooting Information:
If the DHCP Discover request is not visible on the NCN, it may be due to a firewall issue. If the DHCP Discover request is not visible inside the pod, double check if the request was issued over the correct interface for the Node Management Network (NMN). If it was, it could indicate a firewall issue.