[
https://issues.apache.org/jira/browse/SOLR-5473?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14052880#comment-14052880
]
Mark Miller commented on SOLR-5473:
-----------------------------------
Let’s forget the code for a minute. I have to review that, but I don’t want to
just say the approach looks okay and you come back with a patch an hour later
and are ready to commit. I don’t doubt we can come to agreement on how to
handle the API.
I want to talk about cached clusterstate and watchers. Because, apparently,
that is really what this issue is, and the title of this issue is an irrelevant
implementation detail towards scaling to thousands of collections.
What are the limitations in the new system? What are the differences with the
old system?
If you are monitoring a cluster of many collections in the Solr admin UI and
nodes die and take out collections that are not hosted on the admin UI node,
when will you see this reflected by the cloud visualization tab? Ever? If there
are new limitations or changes, are they documented?
Also, if we are going to move to this different core way of doing things, do we
yet fully understand all of the changes and implications, or is this just build
something to scale to many collections and let the rest fall out as it may? I
honestly do not know yet, but if everything is understood, let’s get it all
listed in one place here and come to agreement on the path forward.
When changing such a core part of this system, and I know I’ve said it before,
it would be great to expand on the minimal tests we have for a lot of this
behavior. It’s part of paying for the change IMO - get your hands dirty on some
of the core tests or lack of tests and help fill out the skeleton SolrCloud
testing that has been put in place. We are short on unit tests especially.
And can we change the title of this issue to reflect it’s full scope?
> Make one state.json per collection
> ----------------------------------
>
> Key: SOLR-5473
> URL: https://issues.apache.org/jira/browse/SOLR-5473
> Project: Solr
> Issue Type: Sub-task
> Components: SolrCloud
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 5.0
>
> Attachments: SOLR-5473-74 .patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74.patch,
> SOLR-5473-74.patch, SOLR-5473-74.patch, SOLR-5473-74_POC.patch,
> SOLR-5473-configname-fix.patch, SOLR-5473.patch, SOLR-5473.patch,
> SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch,
> SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch,
> SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch,
> SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch, SOLR-5473.patch,
> SOLR-5473_undo.patch, ec2-23-20-119-52_solr.log, ec2-50-16-38-73_solr.log
>
>
> As defined in the parent issue, store the states of each collection under
> /collections/collectionname/state.json node
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
