Hemali, would this be a reasonable workaround for your problem?
/cc +Kenneth Knowles - In case there is an alternative
workaround to BEAM-6855.
/cc +Cosmin Arad
On Thu, Feb 18, 2021 at 1:27 PM Brian Hulette wrote:
> I added JvmInitializer [1] to do some one-time initialization per JVM
>
+1 to removing from the codebase, and if it becomes of interest again,
porting to cyvcf2. But most genomics workflows are not using Beam at the
moment.
On Tue, Feb 23, 2021 at 1:12 AM Chamikara Jayalath
wrote:
> Given that we don't support Python 2 anymore, it sounds like this is just
> broken
Yes we have checkpointing enabled in our cluster.
Pipeline runs fine when I do a restart without making any code changes but
after adding a DoFn class (that too not used in pipeline), we have observed
pipeline still reads from kafka, does windowing and distinct but does not write
to S3.
I
On Tue, Feb 23, 2021 at 1:07 AM Jan Lukavský wrote:
> First, +1 to the conclusion of this thread.
>
> One note regarding the composite transforms and triggers *inside* those
> transforms - I think that propagating the triggering from input PCollection
> might be even dangerous and composite
First, +1 to the conclusion of this thread.
One note regarding the composite transforms and triggers *inside* those
transforms - I think that propagating the triggering from input
PCollection might be even dangerous and composite PTransforms that would
be sensitive to the change of triggering
