Bo Zhang created IOTDB-1503: ------------------------------- Summary: 1 node crash causes whole cluster cannot work. Key: IOTDB-1503 URL: https://issues.apache.org/jira/browse/IOTDB-1503 Project: Apache IoTDB Issue Type: Sub-task Components: Cluster Reporter: Bo Zhang Fix For: master branch
sometimes, 1 node crashing can causes whole cluster cannot work. Reason: When 1 nodes crashed, the related RPC connnection will failed. But sometimes, in such case, some nodes can not connect to the crashed nodes and then put the bad client back to client pool . This causes later communication can not work well. -- This message was sent by Atlassian Jira (v8.3.4#803005)