[
https://issues.apache.org/jira/browse/SOLR-1255?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12730125#action_12730125
]
Hoss Man commented on SOLR-1255:
--------------------------------
bq. "no" is my answer I cannot find a usecase for the same .
A single handler can be configured to be both a master an a slave right?
the only scenario i can think of is if you wanted some slaves (for queries) to
get every snapshot created on commit, but other slaves (for offsite backups)
should only get snapshots created on every optimize (which might be something
your cron to happen during downtime, nights or weekends perhaps) ... so you
might want an "/replication" instance of the handler on your master server with
<str name="replicateAfter">commit</str> that all of your query slaves pull
from, but you might also want a "/backup" instance with <str
name="replicateAfter">optimize</str> that slaves in a remote data center pull
from.
Is that something you can do right now with two instances of the replication
handler? or would something in the implementation cause that to break anyway
(in which case we should definitely prevent two instances from being registered)
> An attempt to visit the replication admin page when its not a defined handler
> should display an approp message
> --------------------------------------------------------------------------------------------------------------
>
> Key: SOLR-1255
> URL: https://issues.apache.org/jira/browse/SOLR-1255
> Project: Solr
> Issue Type: Bug
> Reporter: Mark Miller
> Assignee: Noble Paul
> Priority: Trivial
> Fix For: 1.4
>
> Attachments: SOLR-1255.patch
>
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
