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
In order to monitor the health of each connection from the application side, it would be great to store the last successful pong timestamp (in redis, for instance), which we could then build alerts on top of depending on our needs.
For some context, I was thinking about how the node-slack-sdk library implements a client-level ping/pong dance and memorizes that timestamp here.
After browsing through relax source code, though, I realize that you are probably relying on the lower-level gorilla websocket connection to handle that ping/pong directly at the websocket level, right?
Even if that's confirmed, it would be indeed great to be able to propagate that information to Rails etc (via Redis), for custom monitoring.
Just some thoughts :-) Keep up the good work!
The text was updated successfully, but these errors were encountered:
I just saw startPingPump so now understand that you are doing something similar to node-slack-sdk. Storing the last timestamp into redis could be useful (but then I do realize there is probably more work to be done, e.g. making sure reconnections attempt are propagated too). Well - just thoughts, again!
In order to monitor the health of each connection from the application side, it would be great to store the last successful pong timestamp (in redis, for instance), which we could then build alerts on top of depending on our needs.
For some context, I was thinking about how the node-slack-sdk library implements a client-level ping/pong dance and memorizes that timestamp here.
After browsing through relax source code, though, I realize that you are probably relying on the lower-level gorilla websocket connection to handle that ping/pong directly at the websocket level, right?
Even if that's confirmed, it would be indeed great to be able to propagate that information to Rails etc (via Redis), for custom monitoring.
Just some thoughts :-) Keep up the good work!
The text was updated successfully, but these errors were encountered: