[
https://issues.apache.org/jira/browse/SOLR-14702?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17170461#comment-17170461
]
Tomas Eduardo Fernandez Lobbe commented on SOLR-14702:
------------------------------------------------------
bq. They are replicas. If the "leader" fails the state of the cluster is very
different for an end-user managing the cluster compared to when a leader goes
down in SolrCloud.
I don't fully understand what you mean here. In non-SolrCloud case, when the
leader fails the followers won't have the latest index for some time, is that
it? In SolrCloud, the replication may be delayed too (TLOG/PULL replicas). Yes,
in the SolrCloud case, if the leader goes down there are leader elections and
such, but about the fact that one replica "leads" and the other "follow", it's
the same in both architecture IMO
> Remove Master and Slave from Code Base and Docs
> -----------------------------------------------
>
> Key: SOLR-14702
> URL: https://issues.apache.org/jira/browse/SOLR-14702
> Project: Solr
> Issue Type: Improvement
> Security Level: Public(Default Security Level. Issues are Public)
> Affects Versions: master (9.0)
> Reporter: Marcus Eagan
> Priority: Critical
> Time Spent: 6h 50m
> Remaining Estimate: 0h
>
> Every time I read _master_ and _slave_, I get pissed.
> I think about the last and only time I remember visiting my maternal great
> grandpa in Alabama at four years old. He was a sharecropper before WWI, where
> he lost his legs, and then he was back to being a sharecropper somehow after
> the war. Crazy, I know. I don't know if the world still called his job
> sharecropping in 1993, but he was basically a slave—in America. He lived in
> the same shack that his father, and his grandfather (born a slave) lived in
> down in Alabama. Believe it or not, my dad's (born in 1926) grandfather was
> actually born a slave, freed shortly after birth by his owner father. I never
> met him, though. He died in the 40s.
> Anyway, I cannot police all terms in the repo and do not wish to. This
> master/slave shit is archaic and misleading on technical grounds. Thankfully,
> there's only a handful of files in code and documentation that still talk
> about masters and slaves. We should replace all of them.
> There are so many ways to reword it. In fact, unless anyone else objects or
> wants to do the grunt work to help my stress levels, I will open the pull
> request myself in effort to make this project and community more inviting to
> people of all backgrounds and histories. We can have leader/follower, or
> primary/secondary, but none of this Master/Slave nonsense. I'm sick of the
> garbage.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
