[
https://issues.apache.org/jira/browse/SOLR-11438?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Erick Erickson updated SOLR-11438:
----------------------------------
Attachment: SOLR-11438.patch
Current patch, I think it's ready.
I reworked a couple of things. It seems easier if I split out the replication
trackers into a "rollup" and "leader" trackers, these are different roles.
Additionally, rather than accumulate errors and count them, just using a simple
counter in the LeaderReplicationTracker does the trick.
Adding in the delete-by-id was a pain. Not all requests are forwarded to all
replicas.
I successfully beasted this test prior to modifications 1,000 times, so I'm
pretty sure removing the BadApple annotation is OK.
I intend to beast this test another 1,000 times on my Mac Pro and a server. If
all that goes well I'll commit later today.
precommit and tests pass.
> Solr should return rf when min_rf is specified for deletes as well as adds
> --------------------------------------------------------------------------
>
> Key: SOLR-11438
> URL: https://issues.apache.org/jira/browse/SOLR-11438
> Project: Solr
> Issue Type: Improvement
> Security Level: Public(Default Security Level. Issues are Public)
> Affects Versions: 6.6.1, 7.0, master (8.0)
> Reporter: Erick Erickson
> Assignee: Erick Erickson
> Attachments: SOLR-11438.patch, SOLR-11438.patch
>
>
> When we add documents and specify min_rf we get back an rf parameter in the
> response which is the number of replicas that successfully received the add.
> However, for delete-by-id or delete-by-query we do not return this data. Is
> there any harm in it?
> Assigning to myself to track, anyone else who wants it feel free.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
