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
When we started using a multi rack cassandra cluster, we populated the cassandra seed DNS (A) record with one IP from each rack. We instantly saw network connections continuously climbing from the erlang hosts to the seed hosts. We reverted the DNS record to have only one IP and the connections stopped increasing.
The text was updated successfully, but these errors were encountered:
Did you put a DNS hostname in the bootstrap_ips field? I've never tested that, I always use qualified IP. I'll experiment on my side to see if I can reproduce.
We have to use DNS hostnames since updating the IPs would require restarting the Erlang service and the restart of the service is really expensive. Also, the connection leak might be caused by #47 Going to review this with the author of #47
When we started using a multi rack cassandra cluster, we populated the cassandra seed DNS (A) record with one IP from each rack. We instantly saw network connections continuously climbing from the erlang hosts to the seed hosts. We reverted the DNS record to have only one IP and the connections stopped increasing.
The text was updated successfully, but these errors were encountered: