traeak opened a new issue #3030: Riak cluster cleanup on cluster rebuid
URL: https://github.com/apache/trafficcontrol/issues/3030
 
 
   In traffic_ops_golang, whenever a riak cluster is torn down it isn't being 
stopped in a totally correct way according to the basho riak client package.
   
   When a cluster stops executing  we see in the error log per cluster server:
   
   ```
   riak error: 2018/10/03 07:41:43.841775 [ERROR] [connectionManager] error 
when closing connection in stop() write tcp 
localhost:53188->cluster_server:8087: i/o timeout
   ```
   
   The riak pooled cluster greatly mitigates this error but doesn't remove it.  
Previously a defer function was being called to stop cluster execution for each 
and every query, spamming the logs with the above message(s).
   
   The riak pooled cluster attaches a similar stop cluster function to the go 
garbage collector via (riak_services.go):
   
   ```
   runtime.SetFinalizer(sharedCluster, sharedCluster.Stop())
   ```
   
   Now this "error" only prints when the riak cluster machines are changed in 
TODB.
   
   This message in error.log can be recreated by using traffic portal to 
offline (or online) a riak cluster server and then perform an api query against 
anything ssl key related.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to