Reading from /tmp no longer working

[su]

Since yesterday, reading a file copied into the cluster is no longer working.

What used to work:

blob = gcs_bucket.get_blob("dev/data.ndjson") -> works

blob.download_to_filename("/tmp/data-copy.ndjson") -> works

df = spark.read.json("/tmp/data-copy.ndjson") -> fails

When calling os.listdir('/tmp'), the file is listed as expected.

This worked yesterday. Has something changed?

[Debayan]

Hi @Sarah Usher​ , could you please provide the error you are receiving when it is getting failed?

[Anonymous]

Hi @Sarah Usher​ 

Hope all is well! Just wanted to check in if you were able to resolve your issue and would you be happy to share the solution or mark an answer as best? Else please let us know if you need more help. 

We'd love to hear from you.

Thanks!

[Evan_From_Bosto]

I encountered this same issue, and figured out a fix!

For some reason, it seems like only %sh cells can access the /tmp directory. So I just did...

%sh ch /tmp/<file> /dbfs/<desired-location> and then accessed it form there using Spark.