单节点Cassandra的行为类似于多节点Cassandra [英] Single node Cassandra behaving like multinode Cassandra

问题描述

我有一个单节点Cassandra群集，命中 ReadTimeOutException ，并且在服务器中观察到以下日志，这对我来说似乎很奇怪，

I have a single node Cassandra cluster which hits to ReadTimeOutException and I observe the following logs in server which seems weird to me,

ERROR [SharedPool-Worker-91] 2018-05-29 12:09:53,023 ErrorMessage.java:338 - Unexpected exception during request
java.lang.RuntimeException: org.apache.cassandra.exceptions.ReadTimeoutException: Operation timed out - received only 1 responses.
       at org.apache.cassandra.auth.CassandraRoleManager.getRole(CassandraRoleManager.java:489) ~[apache-cassandra-3.0.8.jar:3.0.8]
       at org.apache.cassandra.auth.CassandraRoleManager.getRoles(CassandraRoleManager.java:269) ~[apache-cassandra-3.0.8.jar:3.0.8]
       at org.apache.cassandra.auth.RolesCache.getRoles(RolesCache.java:66) ~[apache-cassandra-3.0.8.jar:3.0.8]
       at org.apache.cassandra.auth.Roles.hasSuperuserStatus(Roles.java:51) ~[apache-cassandra-3.0.8.jar:3.0.8]
       at org.apache.cassandra.auth.AuthenticatedUser.isSuper(AuthenticatedUser.java:71) ~[apache-cassandra-3.0.8.jar:3.0.8]

它说操作超时-仅收到1个响应。，在单节点中，为什么这样说多个回应？

It says Operation timed out - received only 1 responses., In Single node why Its saying expecting more than one response? Could some explain pls.

注意：我为此 system_auth 启用了其他策略密钥空间

NOTE : I have enabled different strategy for this system_auth keyspace

CREATE KEYSPACE system_auth WITH replication = {'class': 'NetworkTopologyStrategy', 'datacenterproc': '1'}  AND durable_writes = true;

并将一致性级别设置为 LOCAL_QUORUM

and set consistency level as LOCAL_QUORUM

Cassandra服务器版本：3.0.8

这会是原因吗？

推荐答案

GC超过超时时间（例如您在其他问题中发布的9秒GC）会导致内部身份验证超时。它很可能收到了响应，有了GC，然后注册为超时。

GCs longer than the timeout (like the 9 second GCs you posted in other questions) can cause internal auth timeouts. Most likely it received the response, had the GC, then registered as timeout.

这篇关于单节点Cassandra的行为类似于多节点Cassandra的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！