Dask hangs indefinitely when repartioning after perist.

[kjleftin]

Describe the issue:

A certain order of persist, repartition, and checking for npartitions causes Dask to hang indefinitely.

Minimal Complete Verifiable Example:

import pandas as pd
import dask.dataframe as dd
from dask.distributed import Client

client = Client()

data = {'col1': [1, 2, 3], 'col2': ['A', 'B', 'C']}
df = pd.DataFrame(data)
ddf = dd.from_pandas(df, npartitions=1)

ddf = ddf.persist() # Remove this persist, and the code works fine.
ddf = ddf.repartition(partition_size='250MB') 
print(ddf.npartitions)
print(ddf.compute())

Anything else we need to know?:

Environment:

• Dask version: 2025.7.0
• Python version: Reproduced on 3.10 and 3.11.
• Operating System: Reproduced on colab and local machine (MacOSX)
• Install method (conda, pip, source): On Colab, using preinstalled deps. Also reproduced locally where deps are installed using pip.