[
https://issues.apache.org/jira/browse/CASSANDRA-17984?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17948707#comment-17948707
]
Brad Schoening commented on CASSANDRA-17984:
--------------------------------------------
[~mck] [~erickramirezau] it would be helpful to describe how the token
allocation algorithm works for initial allocation and re-allocation, clarify
what triggers re-allocation (adding or removing nodes), and how to verify if
allocation has achieved an even distribution. The one sentence
There are some good details [~blambov] provides in CASSANDRA-20230 and
[Token/vNode Allocation
Algorithm|https://docs.google.com/document/d/1ZXGPQC85jF78nk68J_O2LK0jBb4FgO-wN_XU8uKSzM8]
which would be useful to describe, perhaps in the Managing docs section.
We frequently see unbalanced allocations on datacenters and there can be
multiple reasons, including unbalanced allocations, partitions creating
hotspots, or blocked expired sstables. Improved documentation on allocations
would be great.
> WEBSITE - Improve docs on allocate_tokens_for_local_replication_factor
> ----------------------------------------------------------------------
>
> Key: CASSANDRA-17984
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17984
> Project: Apache Cassandra
> Issue Type: Task
> Components: Documentation/Website
> Reporter: Erick Ramirez
> Assignee: Erick Ramirez
> Priority: Normal
> Fix For: NA
>
>
> A user asked on Stack Overflow [how
> {{allocate_tokens_for_local_replication_factor}}
> works|https://stackoverflow.com/questions/74169734/] and that they were not
> able to find information on it.
> We have an opportunity to improve our docs so I've logged this ticket to get
> it done.
--
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
