question Re: I just wondering why should we take odd number(3,5,3,7) of Journal and zookeeper nodes in hadoop cluster...what is rule of (n-1)/ failure? in Support Questions https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108223#M71078 <P>@<A href="https://community.hortonworks.com/users/168/bleonhardi.html">Benjamin Leonhard</A></P><P>Thanks you ...</P> Fri, 01 Jul 2016 19:00:24 GMT g_rao9704 2016-07-01T19:00:24Z I just wondering why should we take odd number(3,5,3,7) of Journal and zookeeper nodes in hadoop cluster...what is rule of (n-1)/ failure? https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108221#M71076 Fri, 01 Jul 2016 17:34:49 GMT https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108221#M71076 g_rao9704 2016-07-01T17:34:49Z Re: I just wondering why should we take odd number(3,5,3,7) of Journal and zookeeper nodes in hadoop cluster...what is rule of (n-1)/ failure? https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108222#M71077 <P>The reason is that in a distributed transactional system with a Paxos or similar algorithm you need a quorum ( majority ). Essentially a transaction is committed once more than 50% of nodes say that the transaction is committed. </P><P>You could do 4 journalnodes / zookeeper nodes as well but you would not get any benefit over 3 nodes and you add additional overhead. 4 Nodes can only survive 1 failed node because 3 journal-nodes are a majority. But not 2. Therefore you need an uneven number. 3 nodes can survive 1 failure, 5 nodes can survive 2 failures, 7 nodes can survive 3 failures and so on ...</P><P><A href="https://en.wikipedia.org/wiki/Paxos_(computer_science)" target="_blank">https://en.wikipedia.org/wiki/Paxos_(computer_science)</A></P> Fri, 01 Jul 2016 18:17:25 GMT https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108222#M71077 bleonhardi 2016-07-01T18:17:25Z Re: I just wondering why should we take odd number(3,5,3,7) of Journal and zookeeper nodes in hadoop cluster...what is rule of (n-1)/ failure? https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108223#M71078 <P>@<A href="https://community.hortonworks.com/users/168/bleonhardi.html">Benjamin Leonhard</A></P><P>Thanks you ...</P> Fri, 01 Jul 2016 19:00:24 GMT https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108223#M71078 g_rao9704 2016-07-01T19:00:24Z Re: I just wondering why should we take odd number(3,5,3,7) of Journal and zookeeper nodes in hadoop cluster...what is rule of (n-1)/ failure? https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108224#M71079 <P>If 3 nodes can survive 1 failure , then how will quorum ( majority ) will meet with 2 running server?</P> Sat, 18 Nov 2017 16:52:32 GMT https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108224#M71079 agrawalreetika 2017-11-18T16:52:32Z Re: I just wondering why should we take odd number(3,5,3,7) of Journal and zookeeper nodes in hadoop cluster...what is rule of (n-1)/ failure? https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108225#M71080 <P>Additional details to Ben Leonhardi's response including the formula for Quorum Calculation: </P><P><A href="http://bytecontinnum.com/zookeeper-always-configured-odd-number-nodes/" target="_blank">http://bytecontinnum.com/zookeeper-always-configured-odd-number-nodes/</A></P> Tue, 16 Jan 2018 22:13:26 GMT https://community.cloudera.com/t5/Support-Questions/I-just-wondering-why-should-we-take-odd-number-3-5-3-7-of/m-p/108225#M71080 smdas 2018-01-16T22:13:26Z