lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark Miller (Commented) (JIRA)" <>
Subject [jira] [Commented] (SOLR-2808) Node Recovery
Date Fri, 02 Dec 2011 03:04:42 GMT


Mark Miller commented on SOLR-2808:

To update what's done here:

I've taken what Yonik has done in SOLR-2700 and implemented basic replication recovery.

When a new replica comes up, or a node goes down and comes back, recovery is triggered. The
recovery process issues a commit to the leader, starts buffering updates to the transaction
log, and does a replication. Then it replays the buffered updates.

This gives us working recovery. Eventually, we will want to use realtime-get to do a faster
recovery when the node has not been down for long - that's really just an optimization though.

I'm still slaving away on getting good tests for this, but its up and working which is pretty
> Node Recovery
> -------------
>                 Key: SOLR-2808
>                 URL:
>             Project: Solr
>          Issue Type: Sub-task
>          Components: SolrCloud
>            Reporter: Yonik Seeley
>             Fix For: 4.0
>         Attachments:
> Node recovery encompasses everything involved in getting a node to the "active" state
after coming (back) up.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message