Nidhogg not removing taint after 2 attempts

[chomatdam]

Context

Version: 0.5.1

• Multus DS failing on a missing CNI config file and restarting
• Primary CNI adding its CNI config file
• Multus successfully wrapped this primary CNI, but Nidhogg not removing taint after Multus DS is in a ready state

Unexpected behavior

• two retries (resource version different, resource modified in between) then just stopped retrying / being requeued

Example for a failed attempt:

"error":"Operation cannot be fulfilled on nodes \"ip-172-15-2-127.us-gov-east-1.compute.internal\": the object has been modified; please apply your changes to the latest version and try again"

When successful, logs are showing it retried sometimes four times before succeeding and giving a

{"level":"info","ts":"2024-12-17T21:42:45Z","logger":"nidhogg","msg":"Node taints updated."}

The question here would be why after two retries, it stopped retrying? (caused a node to hang around in a not ready state for 5 hours with the startup taint not removed, a pod restart on Multus triggered Nidhogg again and fixed it temporarily)