[
https://issues.apache.org/jira/browse/SOLR-18438?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18113835#comment-18113835
]
Eric Pugh commented on SOLR-18438:
----------------------------------
This cleans up the admin ui and removes another reason for CLUSTERSTATUS to
have liveNodes parameter.
> Unify retrieving "live node" information in Solr Admin to a single API
> ----------------------------------------------------------------------
>
> Key: SOLR-18438
> URL: https://issues.apache.org/jira/browse/SOLR-18438
> Project: Solr
> Issue Type: Improvement
> Components: Admin UI, v2 API
> Affects Versions: 10.0
> Reporter: Eric Pugh
> Priority: Major
>
> Admin UI get's live node data via two different mechanims:
> * Collections.status() → GET /api/cluster → reads .cluster.live_nodes, a
> plain JSON array — used by the main "Nodes" table view.
> - Zookeeper.liveNodes → GET admin/zookeeper?path=/live_nodes — a generic
> ZK-tree-browsing endpoint, returning a jstree-style node structure
> (data.tree[0].children[c].text) that initGraph() manually walks to build a
> live_nodes[name] = true lookup map — used by a separate "graph" visualization
> mode.
> One more thing that we found is that the liveNodes is driven directly off of
> clusterstate.json, and that specific file may be deprecated or not used?
>
> /api/cluster/nodes is the V2 openapi that we SHOULD be using.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
