noreentry comments

Results 5 comments of


                                            noreentry

KilledWorker (exceeded 95% memory budget) with new optimizer

1. with memory_limit='30GB' it has 15Gb per thread and still runs out of memory. 2. removing `sort=True` in `groupby('ts2', sort=True).aggregate({` solved the problem (but `sort=True` is default for groupby in...

KilledWorker (exceeded 95% memory budget) with new optimizer

so if df1 is sorted by index ts2 than `df2=df1.groupby('ts2').aggregate({...` should be sorted by ts2 also?

StreamClosedError when calling client.shutdown()

There is a case where a df is persisted using `.persist()` on workers, and when `.close()` or `.scale(1)` is called, some workers exit earlier than others. As a result, I...

dd.concat([]).reset_index().set_index() raises IndexError

It works in - Dask: 2025.11.0 (installed from conda-forge) - Python: 3.13.9

df indexing 'The index and array have different numbers of blocks' error

some simplified scenario of creating errored parq ```python #%% import dask import dask.dataframe as dd import dask.distributed import fastparquet import numpy as np import os import pandas as pd import...