Bug: agentd service restart can hang in stop-sigterm for ~2m

t-748·WorkTask·
·
·
Created2 weeks ago·Updated2 weeks ago·pipeline runs →

Dependencies

Description

Edit

During deployer-driven unit update, agentd logged graceful shutdown and ExitSuccess, but systemd kept unit in deactivating/stop-sigterm until timeout, then SIGKILLed old process tree before starting new binary. This causes delayed rollouts and transient downtime. Need to investigate wrapper/child process behavior on SIGTERM and systemd stop completion semantics.

Timeline (0)

No activity yet.