[
https://issues.apache.org/jira/browse/PHOENIX-585?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gabriel Reid resolved PHOENIX-585.
----------------------------------
Resolution: Fixed
Bulk resolve of closed issues imported from GitHub. This status was reached by
first re-opening all closed imported issues and then resolving them in bulk.
> Late-replaying index writes from the WAL.
> -----------------------------------------
>
> Key: PHOENIX-585
> URL: https://issues.apache.org/jira/browse/PHOENIX-585
> Project: Phoenix
> Issue Type: Task
> Reporter: Jason Yates
>
> When WAL recovery occurs, we need to ensure that we can write
> to the index table. If the index region we were targeting lived
> on the same server as the primary region, it will likely be involved
> in the same log splitting and recovery process. This means, when we
> replay the writes into the primary table, the index table/region
> we would write to is probably not up yet. In 0.96 this is fixed by
> allowing regions to take writes before reads during WAL replay. We
> are only on 0.94 (and its a huge effort to backport that feature),
> so instead we can get around the issue by buffering the failed writes
> and replaying them later (in the postOpen hook).
--
This message was sent by Atlassian JIRA
(v6.2#6252)
