[
https://issues.apache.org/jira/browse/SOLR-8649?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15139447#comment-15139447
]
Steve Molloy commented on SOLR-8649:
------------------------------------
Another issue about launching Solr without a chroot to inexisting zNode.
Attached patch adds a parameter to automatically create the zNode if required.
> Fail fast on wrong ZK chroot
> ----------------------------
>
> Key: SOLR-8649
> URL: https://issues.apache.org/jira/browse/SOLR-8649
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Reporter: Shalin Shekhar Mangar
> Fix For: 5.5, Trunk
>
>
> A typical scenario is when a user sets up ZK with a chroot /solr, runs Solr
> and then restarts Solr without specifying the chroot. In the default
> legacyCloud mode, Solr will happily start and create all ZK nodes as well as
> collections found on the local cores.
> I've been bit many times by this and so have more than a few Solr users. In a
> private discussion, Hoss gave the following idea:
> * We add a command to bin/solr to "prepare" ZooKeeper that accepts the zk
> host string
> * The command creates the chroot if it does not exist
> * Touches /my-chroot/solr.key file
> * Writes the complete zk host in the solr.in.sh or solr.in.cmd file
> Once we do this, Solr will complain and fail fast if the solr.key file is not
> found in the given chroot. We could also write a fixed string in the
> /my-chroot instead of creating a /my-chroot/solr.key file.
> We can do these things automatically when using embedded ZooKeeper.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
