简体繁体 English

Cassandra闲话协议和phi_threshold如何工作？

[英]How does cassandra gossip protocol and phi_threshold works?

原文 2016-10-04 18:39:31 6 1 cassandra/ gossip

Current setting, cassandra 2.2.5, gossip is 1 second default and phi threshold value is 8. The problem, I am facing is spikes in hints. 当前设置为cassandra 2.2.5，八卦默认值为1秒，phi阈值为8。我面临的问题是提示中的峰值。 And one of the reason hints goes up is when node is marked down (gossip has not communicated for phi threshold value). 提示上升的原因之一是节点被标记为向下（闲话未传达phi阈值）。

I read one article, where it say phi threshold value of 8 corresponds to 18 seconds, it will be few seconds here or there. 我读了一篇文章，说phi阈值8对应于18秒，这里或那里将是几秒钟。 Now I need to understand what is the reason, what is blocking gossip to communicate for 18 seconds. 现在，我需要了解什么是原因，什么阻止八卦沟通18秒钟。 What is the checklist that need to be satisfied for gossip to communicate? 八卦沟通需要满足的清单是什么？

1 个解决方案

Re: "How does cassandra gossip protocol and phi_threshold works?": Phi is approximated as: phi = (tnow - tLast) / mean and a node is marked down when phi > phi_threshold / 0.434 . 回复：“ cassandra闲话协议和phi_threshold如何工作？”：将phi 近似为： phi = (tnow - tLast) / mean并且当 phi > phi_threshold / 0.434 时，节点被标记为向下。 For your settings (and assuming a mean of 1 [as in the node usually receives the heartbeat 1 second apart]) a node will be marked down if we didn't receive any heartbeats from it for 8 / 0.434 = 18.42 seconds. 对于您的设置（并假设平均值为1 [在节点中通常每隔1秒就接收一次心跳]），如果我们在8 / 0.434 = 18.42秒内未从节点收到任何心跳，则该节点将被标记为down。 The paper documenting the algorithm can be found here . 有关该算法的文献可在此处找到。
Re: "What is the checklist that needs to be satisfied for gossip to communicate?": to me there are a few things: 回复：“八卦沟通需要满足的清单是什么？”：对我来说，有几件事：
- the network: the gossip messages being dropped or the gossip port (7000/7001) being blocked; 网络：八卦消息被丢弃或八卦端口（7000/7001）被阻止；
- the nodes themselves: the nodes is busy/unresponsive (ie doing GC, doing some heavy load operation) so they don't get to send any/much gossip messages. 节点本身：节点忙/无响应（即执行GC，执行一些繁重的操作），因此它们不会发送任何/很多八卦消息。