[
https://issues.apache.org/jira/browse/CASSANALYTICS-128?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Josh McKenzie updated CASSANALYTICS-128:
----------------------------------------
Test and Documentation Plan: Unit tests for new flag.
Status: Patch Available (was: Open)
> Add WriterOption to allow bulk write to tables with Secondary Indexes
> ---------------------------------------------------------------------
>
> Key: CASSANALYTICS-128
> URL: https://issues.apache.org/jira/browse/CASSANALYTICS-128
> Project: Apache Cassandra Analytics
> Issue Type: Task
> Components: Writer
> Reporter: Josh McKenzie
> Assignee: Josh McKenzie
> Priority: Normal
> Time Spent: 10m
> Remaining Estimate: 0h
>
> We should allow users to bulk write to tables with 2i and/or SAI even before
> we have formal support for pre-writing (in the latter case) baked in. While
> CASSANALYTICS-31 will front-load the index writing portion of things, bulk
> writing tables w/secondary indexes on them is still technically correct in
> the current regime as the bulk import on the C* side creates those indexes.
> This is kind of a "small foot-gun" since one could theoretically bulk import
> a huge amount of data on a table with a legacy 2i and then have a
> nondeterministic window of time where they have a stale index being served.
> The "solution" to this would of course be to not start applications / users
> reading from that table before the index was done building.
> Either way - having options is good, especially when the architecture is
> eventually consistent and thus consistent w/the rest of the way C* works. :)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
