[
https://issues.apache.org/jira/browse/CASSANDRA-18546?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17726949#comment-17726949
]
Stefan Miklosovic commented on CASSANDRA-18546:
-----------------------------------------------
trunk
j11
https://app.circleci.com/pipelines/github/instaclustr/cassandra/2322/workflows/1565ea8d-47d7-4bd4-b31e-125b2351bc77
j8
https://app.circleci.com/pipelines/github/instaclustr/cassandra/2322/workflows/ac14b081-811c-432e-8f34-964b8a387338
branch https://github.com/instaclustr/cassandra/tree/CASSANDRA-18546-trunk
4.1
j11
https://app.circleci.com/pipelines/github/instaclustr/cassandra/2321/workflows/bb3c4b46-dee5-4d25-9f6f-e24a26f5a3c8
j8
https://app.circleci.com/pipelines/github/instaclustr/cassandra/2321/workflows/92f5397d-14e0-4725-8b0b-45fe77d147ff
branch https://github.com/instaclustr/cassandra/tree/CASSANDRA-18546-4.1
4.0
j11
https://app.circleci.com/pipelines/github/instaclustr/cassandra/2318/workflows/d456f4a1-0fed-4397-9ccd-8406e3260728
j8
https://app.circleci.com/pipelines/github/instaclustr/cassandra/2318/workflows/bb938792-21be-40d6-a0dc-331a74a6dc1e
branch https://github.com/instaclustr/cassandra/tree/CASSANDRA-18546-4.0
> Remove Gossiper#makeRandomGossipDigest
> --------------------------------------
>
> Key: CASSANDRA-18546
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18546
> Project: Cassandra
> Issue Type: Improvement
> Components: Cluster/Gossip
> Reporter: Cameron Zemek
> Assignee: Stefan Miklosovic
> Priority: Normal
> Fix For: 4.0.x, 4.1.x, 5.x
>
>
> In going through trying to understand the Gossiper code I come across:
> {code:java}
> /**
> * The gossip digest is built based on randomization
> * rather than just looping through the collection of live endpoints.
> *
> * @param gDigests list of Gossip Digests.
> */
> private void makeRandomGossipDigest(List<GossipDigest> gDigests) {code}
> But I couldn't see what purpose randomization had. In fact in 3.11 it will
> call:
> {code:java}
> doSort(gDigestList); {code}
> On the receiving end, negating the purpose of the randomization.
>
> In discussion with [~stefan.miklosovic] he found this ticket CASSANDRA-14174
> So it seems to me this randomization may have been to allow for limited sizes
> of SYN messages. But this feature doesn't exist and as such by randomizing it
> is:
> * creating more garbage
> * using more CPU (sure its mostly trival; see next point)
> * more time spent on unnecessary functionality on the *single threaded*
> gossip stage.
> * complicating the code and making it more difficult to understand
> In fact there is a bug in the implementation:
> {code:java}
> int generation = 0;
> int maxVersion = 0; // local epstate will be part of
> endpointStateMap
> List<InetAddress> endpoints = new
> ArrayList<InetAddress>(endpointStateMap.keySet());
> Collections.shuffle(endpoints, random);
> for (InetAddress endpoint : endpoints)
> {
> epState = endpointStateMap.get(endpoint);
> if (epState != null)
> {
> generation = epState.getHeartBeatState().getGeneration();
> maxVersion = getMaxEndpointStateVersion(epState);
> }
> gDigests.add(new GossipDigest(endpoint, generation, maxVersion));
> } {code}
> If epState is null and we already had a non-null epState, then the next
> digest will use the generation and maxVersion of the previous iterated
> epState.
>
> Here is change to remove this randomization and fix the above bug,
> [https://github.com/apache/cassandra/pull/2357/commits/1ba422ab5de35f7057c7621ec3607dcbca19768c]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
