[
https://issues.apache.org/jira/browse/PHOENIX-3976?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16154318#comment-16154318
]
Geoffrey Jacoby commented on PHOENIX-3976:
------------------------------------------
[~samarthjain] [~akshita.malhotra] - If a MapReduce task fails, then is
auto-retried by M/R and succeeds, would that throw the counters off?
> Validate Index ASYNC job complete when building off a data table snapshot
> -------------------------------------------------------------------------
>
> Key: PHOENIX-3976
> URL: https://issues.apache.org/jira/browse/PHOENIX-3976
> Project: Phoenix
> Issue Type: Improvement
> Reporter: Samarth Jain
> Assignee: Akshita Malhotra
>
> [~akshita.malhotra] had this good idea of validating whether an async index
> build job has completed successfully by comparing the
> PhoenixJobCounters.INPUT_RECORDS with the number of expected rows. This would
> be especially helpful when we are building the index using a data table
> snapshot. Since the data table snapshot won't be taking any writes, it should
> be correct and hopefully relatively easy (possibly chain of map reduce jobs
> with first job being a row count job run against the data table restored
> using snapshot?) to verify that the number of rows in the data table
> snapshot is equal to the PhoenixJobCounters.INPUT_RECORDS counter.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
