[
https://issues.apache.org/jira/browse/NIFI-5542?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16598679#comment-16598679
]
ASF GitHub Bot commented on NIFI-5542:
--------------------------------------
Github user pepov commented on the issue:
https://github.com/apache/nifi/pull/2970
Nitpicking but now we do 2 lookups. First we look up the group to get the
identifier then we lookup the same group using that identifier. I beleive it
would be enough to store the group name property and look it up in the
populateNodes method only, but this is nitpicking, since there is no
performance or memory penalty in it, so overall this looks good to me.
> Add support for node groups to FileAccessPolicyProvider
> -------------------------------------------------------
>
> Key: NIFI-5542
> URL: https://issues.apache.org/jira/browse/NIFI-5542
> Project: Apache NiFi
> Issue Type: Improvement
> Reporter: Andrew Christianson
> Priority: Major
>
> Currently in FileAccessPolicyProvider, it is possible to specify a set of
> node identities, which are given access to /proxy. This works well for static
> clusters, but does not work so well for dynamic clusters (scaling up/down #
> of nodes) because we don't know in advance what the node identities will be
> or how many there will be.
> In order to support dynamic sets of node identities, add support for
> specifying a "Node Group," for which all identities in the group will be
> granted access to /proxy. A UserGroupProvider can then be implemented to
> gather node identities dynamically from the cluster environment.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
