1. After removing the health check, I started the container and successfully called the health check endpoint from inside the container. See Figure 1.
Plaintext
[root@admin-66bbbb996f-4vfj5 /]# curl http://localhost:81/status
pool: www
process manager: static
start time: 22/Jun/2021:13:17:53 +0800
start since: 40
accepted conn: 1
listen queue: 0
max listen queue: 0
listen queue len: 128
idle processes: 19
active processes: 1
total processes: 20
max active processes: 1
max children reached: 0
slow requests: 0
[root@admin-66bbbb996f-4vfj5 /]#
2. After adding the health check and upgrading, the workload remained stuck in the Upgrading state. See Figure 2.
3. I checked all Pods in the current workload. The newly created container was in the ContainerCreating state. See Figure 3.
4. The health check configuration used during the upgrade is shown below. See Figure 4.
5. After setting all health checks to none, the upgrade succeeded. See Figure 5.
6. I force-deleted all Pods in the current workload. See Figure 6.
7. After adding the health check back, I upgraded again, and the upgrade succeeded.