linux - stop - supervisor windows

Acción de supervisión si el servicio se reinicia

Question

Quiero crear un archivo cuando se reinicie el servicio.

rpm -qa | grep super
 supervisor-3.1.3-0.5.b1.el6.noarch

Mi archivo de configuración:

[program:tests]
command=/root/while.sh
directory=/root
user=root
autostart=true
autorestart=true

[eventlistener:tests_ls]
command=touch /tmp/32
events=PROCESS_STATE

Mi archivo de registro error

2016-06-27 10:48:15,794 ERRO pool tests_ls event buffer overflowed, discarding event 8
2016-06-27 10:48:15,794 INFO exited: tests_ls (exit status 0; not expected)
2016-06-27 10:48:16,796 ERRO pool tests_ls event buffer overflowed, discarding event 9
2016-06-27 10:48:16,796 INFO gave up: tests_ls entered FATAL state, too many start retries too quickly
2016-06-27 10:48:39,155 ERRO pool tests_ls event buffer overflowed, discarding event 10
2016-06-27 10:48:39,155 INFO exited: tests (terminated by SIGKILL; not expected)
2016-06-27 10:48:40,157 ERRO pool tests_ls event buffer overflowed, discarding event 11
2016-06-27 10:48:40,160 INFO spawned: ''tests'' with pid 26378
2016-06-27 10:48:41,163 ERRO pool tests_ls event buffer overflowed, discarding event 12


supervisorctl
 tests                            RUNNING   pid 26378, uptime 0:09:02
 tests_ls                         FATAL     Exited too quickly (process log may have details)

¿Qué hice mal?