The original stop_ganesha() function unmounted mountpoints and attempted to kill the Ganesha daemon only once. If the pkill operation failed, the Ganesha daemon could remain hanging, leading to Ganesha CI containers becoming "unhealthy" and requiring Jenkins nodes to be restarted.
This commit enhances the stop_ganesha() function to handle failed pkill attempts. If the initial pkill fails, a new Ganesha daemon is started, which should terminate the hanging one. This change aims to ensure proper cleanup of resources and prevent Ganesha CI containers from entering an "unhealthy" state, thus avoiding unnecessary Jenkins nodes restarts.
The original
stop_ganesha()function unmounted mountpoints and attempted to kill the Ganesha daemon only once. If thepkilloperation failed, the Ganesha daemon could remain hanging, leading to Ganesha CI containers becoming "unhealthy" and requiring Jenkins nodes to be restarted.This commit enhances the
stop_ganesha()function to handle failedpkillattempts. If the initialpkillfails, a new Ganesha daemon is started, which should terminate the hanging one. This change aims to ensure proper cleanup of resources and prevent Ganesha CI containers from entering an "unhealthy" state, thus avoiding unnecessary Jenkins nodes restarts.