See issue 815340 for context.

This service isn't getting restarted on push-to-prod, meaning it doesn't get updated on pushes. This caused the metrics fix to never reach the two servers in question.


When I manually tried to restart the service, it failed when stopping, but then started OK:

chromeos-test@cros-bighd-0003:~$ sudo service mysql_stats restart
stop: Job failed while stopping
mysql_stats start/running, process 54758
chromeos-test@cros-bighd-0003:~$ ps aux | grep mysql_stats
root      54758  8.2  0.0  92340 27100 ?        Ss   10:58   0:00 sudo -u chromeos-test /usr/local/autotest/site_utils/stats/mysql_stats.py
chromeo+  54761  8.5  0.0 174512 32676 ?        Sl   10:58   0:00 /usr/bin/python /usr/local/autotest/site_utils/stats/mysql_stats.py
chromeo+  54775  0.0  0.0  23756  2236 pts/7    S+   10:58   0:00 grep --color mysql_stats