[
https://issues.apache.org/jira/browse/CASSANDRA-6127?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13815379#comment-13815379
]
Brandon Williams commented on CASSANDRA-6127:
---------------------------------------------
At this point, I think we should:
* see if the flapping happens with vnodes (maybe Quentin already knows from his
last test)
* see if the flapping happens without vnodes but the same number of nodes
Because if sum() in ArrivalWindow is burning the most CPU in the Gossiper task
(note: not bottlenecking, each call was at most ~3ms, there were just lots of
them) then that means that the problem is no longer tied to vnodes (if it ever
was, since sum is per-node, not per-token) and we should probably open a new
ticket (can't start a cluster of size >=X all at once, or similar) and discuss
there. We know that clusters much larger than any discussed on this ticket
exist, but I don't think any of them have all rebooted at once.
> vnodes don't scale to hundreds of nodes
> ---------------------------------------
>
> Key: CASSANDRA-6127
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6127
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Environment: Any cluster that has vnodes and consists of hundreds of
> physical nodes.
> Reporter: Tupshin Harper
> Assignee: Jonathan Ellis
> Attachments: 6000vnodes.patch, AdjustableGossipPeriod.patch,
> delayEstimatorUntilStatisticallyValid.patch
>
>
> There are a lot of gossip-related issues related to very wide clusters that
> also have vnodes enabled. Let's use this ticket as a master in case there are
> sub-tickets.
> The most obvious symptom I've seen is with 1000 nodes in EC2 with m1.xlarge
> instances. Each node configured with 32 vnodes.
> Without vnodes, cluster spins up fine and is ready to handle requests within
> 30 minutes or less.
> With vnodes, nodes are reporting constant up/down flapping messages with no
> external load on the cluster. After a couple of hours, they were still
> flapping, had very high cpu load, and the cluster never looked like it was
> going to stabilize or be useful for traffic.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
