hi kant,
"FULL OUTER JOIN" job will generate retract message, so toRetractStream is
required to guarantee the correctness.
I think it's better to use StreamExecutionEnvrionment.execute, because you
have converted the Table to DataStream.
kant kodali 于2020年1月19日周日 上午11:59写道：
> Hi Godfrey,
>
> I w
Hi Godfrey,
I was just clicking the run button on my IDE and it doesn't really show me
errors so I used command line fink run and that shows me what the
error is. It tells me I need to change to toRetractStream() and both
StreamExecutionEnvrionment and StreamTableEnvrionment .execute seems to
wor
I think this exception is not because the hadoop version isn't high enough.
It seems that the "s3" URI scheme could not be recognized by
`S3FileSystemFactory`. So it fallbacks to
the `HadoopFsFactory`.
Could you share the debug level jobmanager/taskmanger logs so that we could
confirm whether the
Hi Godfrey,
Thanks a lot for your response. I just tried it with env.execute("simple
job") but I still get the same error message.
Kant
On Sat, Jan 18, 2020 at 6:26 PM godfrey he wrote:
> hi kant,
>
> > 1) The Documentation says full outer join is supported however the below
> code just exits
hi kant,
> 1) The Documentation says full outer join is supported however the below
code just exits with value 1. No error message.
if you have converted Table to DataStream, please execute it
with StreamExecutionEnvironment ( call env.execute("simple job") )
> 2) If I am using a blink planner sh
Hi All,
Does Flink 1.9 support create or replace views syntax in raw SQL? like
spark streaming does?
Thanks!
[Note that this question is better suited for the user mailing list than
dev.]
In general using ListState and MapState is recommended rather than
using ValueState> or ValueState>.
Some of the state backends are able to optimize common access patterns for
ListState and MapState in ways that are no
Hi Anuj,
I think that there may be a fundamental misunderstanding about the role of
a schema registry in Kafka. So let me first clarify that.
In each Avro/Parquet file, all records have the same schema. The schema is
stored within the file, such that we can always retrieve the writer schema
for th
Thanks, Arvid.
I do not fully understand the above approach,
so currently, I am thinking to go with the envelope approach that you
suggested.
One more question I have if I do not want to keep schema in my consumer
project even its a single envelope schema. I want it to be fetched from the
schema
(Readded user mailing list)
Hi Anuj,
since I'd still recommend going with distinct sources/sinks, let me try to
solve your issues in this mail. If that doesn't work out, I'd address your
concerns about the envelope format later.
In Flink, you can have several subtopologies in the same applicatio
What about using an accumulator? Does it work for you needs?
Il Sab 18 Gen 2020, 10:03 Soheil Pourbafrani ha
scritto:
> Hi,
>
> I'm using Flink to insert some processed records into the database. I need
> to have some aggregated information about records inserted into the
> database so far. For
Hi All,
1) The Documentation says full outer join is supported however the below
code just exits with value 1. No error message.
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.
Hi,
I'm using Flink to insert some processed records into the database. I need
to have some aggregated information about records inserted into the
database so far. For example, for a specific column value, I need to know
how many records have been inserted. Can I use the Flink Matrics to provide
t
13 matches
Mail list logo
