Did you manage to get any further with this?
On Fri, 6 Oct 2017 at 05:47, Manju A <[email protected]> wrote:
>
> > Hi Team,
> >
> >
> >
> > Using flume interceptor , I am reading messages from kafka with key and
> value pair. The key is represented by an integer variable called pk in
> below code and the value of message is represented by obj variable with
> Ctrl A delimiter.
> >
> >
> >
> > For inserting into hive table am doing this:-
> >
> > MutableRecord mutableRecordInput = new
> MutableRecord(pk,obj.split("\u0001"));
> >
> > MutableRecord mutableRecord = (MutableRecord)
> bucketIdResolver.attachBucketIdToRecord(mutableRecordInput);
> >
> > coordinator.insert(Collections.<String> emptyList(), mutableRecord);
> >
> >
> >
> >
> >
> > For updating the records in hive table with pk being key am doing this:-
> >
> > RecordIdentifier rd = new RecordIdentifier();
> >
> > MutableRecord mutableRecordInputup = new
> MutableRecord(pk,rd,obj.split("\u0001"));
> >
> > MutableRecord mutableRecordInputu = (MutableRecord)
> bucketIdResolver.attachBucketIdToRecord(mutableRecordInputup);
> >
> > coordinator.update(Collections.<String> emptyList(),
> mutableRecordInputu);
> >
> >
> >
> >
> >
> > The inserts work fine. However updates are causing data inconsistency
> and instead of updating, it is actually inserting for the first time and
> subsequent updates are updating this inserted row and hence is not working
> as expected. Please let me know if I am doing the updates in the right way.
> >
> > I am using flume 1.6 version and hive 2.1 version provided by mapr
> >
> > Thanks and Regards,
> >
> > Manjunath Anand
>
