jakirkham
commented
6 years ago Should add that sometimes when running this the value does change as it should, but it is infrequent.
Also sometimes a GeneratorExit exception is thrown and ignored resulting in a RuntimeError like so.
```python
Exception ignored in:
RuntimeError: generator ignored GeneratorExit
```
Occasionally a different error crops up suggesting the cluster did not get shutdown appropriately somehow. Error shown below. Sometimes there are longer versions of this error involving more StreamClosedErrors.
```python
Future exception was never retrieved
future: : Stream is closed',)>
Traceback (most recent call last):
File "/zopt/conda2/envs/test/lib/python3.6/site-packages/distributed/comm/tcp.py", line 179, in read
n_frames = yield stream.read_bytes(8)
File "/zopt/conda2/envs/test/lib/python3.6/site-packages/tornado/gen.py", line 1099, in run
value = future.result()
tornado.iostream.StreamClosedError: Stream is closed
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/zopt/conda2/envs/test/lib/python3.6/site-packages/tornado/gen.py", line 1107, in run
yielded = self.gen.throw(*exc_info)
File "/zopt/conda2/envs/test/lib/python3.6/site-packages/distributed/comm/tcp.py", line 200, in read
convert_stream_closed_error(self, e)
File "/zopt/conda2/envs/test/lib/python3.6/site-packages/distributed/comm/tcp.py", line 128, in convert_stream_closed_error
raise CommClosedError("in %s: %s" % (obj, exc))
distributed.comm.core.CommClosedError: in : Stream is closed
```
These two errors may or may not show up together or at all and may or may not correspond to the correct or incorrect output. All of this suggesting these are probably different issues.
mrocklin
Appears that when launching tasks with the Distributed Client, sometimes
Futures are held onto a little too long and their results may end up being reused for subsequent tasks that have the same name. A trivial example is shown below.Test script:
```python # test_dist_fut.py from dask.delayed import Delayed import distributed if __name__ == "__main__": c = distributed.Client() v = Delayed("v", {"v": 0}) v = v.persist() print(v.compute()) del v v = Delayed("v", {"v": 1}) v = v.persist() print(v.compute()) ```Running script:
```bash $ python test_dist_fut.py 0 0 ```Environment:
```yaml name: test channels: - conda-forge - defaults dependencies: - ca-certificates=2018.1.18=0 - click=6.7=py_1 - cloudpickle=0.5.2=py_0 - cytoolz=0.9.0.1=py36_0 - dask-core=0.17.2=py_0 - distributed=1.21.4=py36_0 - heapdict=1.0.0=py36_0 - msgpack-python=0.5.5=py36_0 - ncurses=5.9=10 - openssl=1.0.2n=0 - psutil=5.4.3=py36_0 - python=3.6.4=0 - pyyaml=3.12=py36_1 - readline=7.0=0 - six=1.11.0=py36_1 - sortedcontainers=1.5.9=py36_0 - sqlite=3.20.1=2 - tblib=1.3.2=py36_0 - tk=8.6.7=0 - toolz=0.9.0=py_0 - tornado=5.0.1=py36_1 - xz=5.2.3=0 - yaml=0.1.7=0 - zict=0.1.3=py_0 - zlib=1.2.11=0 ```