You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I would like to see pyCA communicate with systemd via notifications, so that we can leverage the systemd watchdog. Sometimes, some processes will hang themselves, which is unacceptable in a production environment.
If pyCA would use sdnotify, we could change the service type to notify and set a Watchdog that is about twice of the default check intervall and send a notification on every control loop iteration. A stuck worker would then be forcefully restarted by systemd, ensuring that the basic functionionality will work.
Of course this is much harder to do on the capture worker - I would leave this out for now.
The text was updated successfully, but these errors were encountered:
https://pypi.python.org/pypi/sdnotify
I would like to see pyCA communicate with systemd via notifications, so that we can leverage the systemd watchdog. Sometimes, some processes will hang themselves, which is unacceptable in a production environment.
If pyCA would use sdnotify, we could change the service type to
notify
and set a Watchdog that is about twice of the default check intervall and send a notification on every control loop iteration. A stuck worker would then be forcefully restarted by systemd, ensuring that the basic functionionality will work.Of course this is much harder to do on the capture worker - I would leave this out for now.
The text was updated successfully, but these errors were encountered: