lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hoss Man (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SOLR-7099) bin/solr -cloud mode should launch a local ZK in its own process using zkcli's runzk option (instead of embedded in the first Solr process)
Date Wed, 11 Feb 2015 22:43:12 GMT

    [ https://issues.apache.org/jira/browse/SOLR-7099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14317133#comment-14317133
] 

Hoss Man commented on SOLR-7099:
--------------------------------

bq. I think it might make sense to have a more generic name. That 1. hides the implementation
detail of running zk for anyone who doesn't want/need to know. 2. Gives us the freedom to
replace the configuration manager (zk) with something else

I disagree -- right now, for a high quality production installation of Solr it's very important
to understand that ZooKeeper is involved, and to understand the importance of having multiple
ZK nodes.  If/when we replace (or add an option to substitute something new for) ZooKeeper,
then it will almost certaily still be important to understand how that new thing works and
how to have it working reliably.

it's one thing to add a convince option that says "here's a simple command line to setup a
single node ZK instance" but we shouldn't hide the fact that it's zk, or that it's a single
node -- it should not be magic.  and if we name this command line option/script something
agnostic of the fact that it's launching a zk node, then the user will only ever think of
it as magic, and never understnad why they have to run it (or what the importance of having
multiple "magic" (aka: zk) nodes configured to talk to eachother is).

> bin/solr -cloud mode should launch a local ZK in its own process using zkcli's runzk
option (instead of embedded in the first Solr process)
> -------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: SOLR-7099
>                 URL: https://issues.apache.org/jira/browse/SOLR-7099
>             Project: Solr
>          Issue Type: Improvement
>          Components: SolrCloud
>            Reporter: Timothy Potter
>
> Embedded ZK is great for unit testing and quick examples, but as soon as someone wants
to restart their cluster, embedded mode causes a lot of issues, esp. if you restart the node
that embeds ZK. Of course we don't want users to have to install ZooKeeper just to get started
with Solr either. 
> Thankfully, ZkCLI already includes a way to launch ZooKeeper in its own process but still
within the Solr directory structure. We can hide the details and complexity of working with
ZK in the bin/solr script. The solution to this should still make it very clear that this
is for getting started / examples and not to be used in production.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message