Hi Buntu, Apparently, the necessary classes related to Flume client are already part of the phoenix client. Hence, you wouldn't need to build a Flume client artifact additionally. In the current implementation, the UPSERT query is based on an existing table only . However, if you would like to have support for dynamic columns , please do raise a JIRA . We can definitely add features based on your feedback.
Regards Ravi On Wed, Jun 10, 2015 at 10:15 PM, Buntu Dev <[email protected]> wrote: > I plan on using Apache Phoenix on CDH and installed the parcel > (4.3.0-clabs-phoenix-1.0.0-all). Does the Flume client still needs to be > built for the client jar as mentioned in the install and setup steps here: > http://phoenix.apache.org/flume.html > > I'm also looking for some examples of ingesting events to Phoenix via > Flume. Any pointers on creating a table and dynamically adding column > families and columns during ingestion will be really helpful. > > > Thanks! >
