Comparison
Startup probevsTermination grace period
Startup probe
the application takes ninety seconds to warm up, and this is what stops the health checker killing it at thirty.
A health check that runs only during start-up, suspending the other probes until the process reports ready for the first time. It exists to solve one specific bug: a slow-starting application in a restart loop because its liveness check began before it was ever alive. Without it the only fix is a liveness delay long enough to also delay detection of genuine hangs for the rest of the workload's life.
Full entry →Termination grace period
you get thirty seconds after the signal to finish, and at second thirty-one you are killed with no negotiation.
The window between the polite termination signal and the unconditional kill. It has to exceed the longest piece of work a container can legitimately be doing, which for a request handler is short and for a batch consumer can be minutes. Every deployment being exactly as slow as this number is the classic sign that nothing is handling the signal and everything is waiting to be killed.
Full entry →