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
Let's assume my-graphite.internal.address resolves to 192.168.0.1
Machine 192.168.0.1 goes down and graphite gets restarted on 192.168.0.2
Statsd fails to connect to 192.168.0.1, logging the error but attempting to reconnect again to 192.168.0.1. However it does not terminate the TCP connection properly, leaving it orphaned in FIN_WAIT2. This keeps happening until all sockets are blocked and no other service on this machine can use any socket.
Expected Behaviour
Statsd should resolve the DNS name on every connection attempt and not save the IP address. This way, when the IP address changes, resolution will happen on the next reconnect attempt.
If reconnecting is not possible, statsd should terminate the connection properly, not leaving it orphaned.
Problem
graphiteHost=my-graphite.internal.address:1704
my-graphite.internal.address
resolves to192.168.0.1
192.168.0.1
goes down and graphite gets restarted on192.168.0.2
192.168.0.1
, logging the error but attempting to reconnect again to192.168.0.1
. However it does not terminate the TCP connection properly, leaving it orphaned inFIN_WAIT2
. This keeps happening until all sockets are blocked and no other service on this machine can use any socket.Expected Behaviour
Relevant Logs
The text was updated successfully, but these errors were encountered: