刘珍 created IOTDB-1779:
-------------------------
Summary: The network of one node is unreachable，write blocking
problem for three node three replica cluster
Key: IOTDB-1779
URL: https://issues.apache.org/jira/browse/IOTDB-1779
Project: Apache IoTDB
Issue Type: Improvement
Components: Core/Cluster
Reporter: 刘珍
Attachments: 3节点3副本_拔网线_测试结果.docx, image-2021-09-29-15-49-33-180.png,
image-2021-09-29-15-49-44-405.png, image-2021-09-29-15-51-24-108.png,
拔网线_v12_config.properties
master 87e1ae444523c07b66b7bde1ec63d6b1924e3fca
Three node three replica cluster，benchmark connects a node to write sequential
data，remove the network cable from one node，Iotdb has the following problems：
（3节点3副本集群，benchmark连接其中1个节点写入顺序数据，拔掉其他节点的网线火停止其他节点的网络服务--每次操作只影响1个节点，iotdb有如下几个问题）：
1. Symptom of pulling out a network cable（拔网线的表现）
问题1：Benchmark got exception (benchmark写入异常）：
org.apache.iotdb.rpc.StatementExecutionException: {color:#DE350B}701: Request
timed out.{color}
org.apache.iotdb.rpc.StatementExecutionException: 702:{color:#DE350B}* Leader
cannot be found.*{color}
cli 执行insert ：
问题2：cli执行select耗时120秒：
!image-2021-09-29-15-49-33-180.png!
问题3：node status 耗时120多秒
!image-2021-09-29-15-49-44-405.png!
问题4：插上网线，网络恢复后
Benchmark依然报错：
org.apache.iotdb.rpc.IoTDBConnectionException: Fail to reconnect to server.
Please check server status
停止网络服务（systemctl stop network）
benchmark写入报同样的错误：
org.apache.iotdb.rpc.IoTDBConnectionException: Fail to reconnect to server.
Please check server status
但是命令行方式停止网络服务期间，cli 执行insert，select正常：
insert into root.net.dev(col) values(now(),3);
insert into root.net.dev(col) values(now(),4);
insert into root.net.dev(col) values(now(),5);
select * from root.net.dev;
!image-2021-09-29-15-51-24-108.png!
详细结果见附件。
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
