shunping commented on issue #33815:
URL: https://github.com/apache/beam/issues/33815#issuecomment-2844087710
I can reproduce this with a simplified pipeline in Python as well.
```python
import apache_beam as beam
from apache_beam.options.pipeline_options import PipelineOptions
from apache_beam.transforms.periodicsequence import PeriodicImpulse
from apache_beam.utils.timestamp import Timestamp
options = PipelineOptions([
"--streaming",
"--job_server_timeout=600",
"--environment_type=LOOPBACK",
#"--runner=PrismRunner", "--prism_beam_version_override=v2.64.0",
"--runner=PortableRunner",
"--job_endpoint=localhost:8073",
])
class InitCount(beam.DoFn):
def process(self, element):
return [1]
class PlusOne(beam.DoFn):
def process(self, element):
print(element)
return [element + 1]
INT32_MAX_TIMESTAMP = Timestamp(micros=2147483647 * 1000)
duration = 5
## windowing before flatten
with beam.Pipeline(options=options) as p:
unboundedSource = p | "s2" >> PeriodicImpulse(
start_timestamp=INT32_MAX_TIMESTAMP, fire_interval=duration)
c = unboundedSource | "i2" >> beam.ParDo(InitCount())
_ = c | beam.Reshuffle() | beam.ParDo(PlusOne())
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
