topic Re: Python job run error messages are unreadable in Data Engineering

Python job run error messages are unreadable

stinodego — Fri, 18 Nov 2022 08:48:49 GMT

This has been going on for some time now; all errors look like this (note the weird `[0;34m` marks everywhere). How can we fix this?

We're not doing anything crazy, this is just the latest runtime with pretty much the simplest possible hello world program.

Re: Python job run error messages are unreadable

Anonymous — Sat, 19 Nov 2022 08:10:31 GMT

Could you give your scenario, your code and which dbr runtime you executed that code?

Re: Python job run error messages are unreadable

Anonymous — Mon, 21 Nov 2022 04:03:55 GMT

Could you try on the other web browsers? Or clear cookies, check library installed or not?

Re: Python job run error messages are unreadable

stinodego — Mon, 21 Nov 2022 05:47:55 GMT

Tried this, but no change. Anyone looking at these logs sees the same thing.

Re: Python job run error messages are unreadable

UmaMahesh1 — Sat, 26 Nov 2022 18:00:45 GMT

Hi @Stijn de Gooijer

Could you please tell the DBR runtime you executed that code on.

Assuming that you are doing basic stuff and not importing any other python packages. If not, please paste what all packages are imported.

Cheers.

Re: Python job run error messages are unreadable

stinodego — Sat, 26 Nov 2022 18:29:48 GMT

DBR is 11.3 LTS (non-ML). Not sure about packages, I'll try to work out a minimum example.

I was hoping one of you had seen this before and knows what's going on.

Re: Python job run error messages are unreadable

Anonymous — Mon, 28 Nov 2022 03:02:57 GMT

@Stijn de Gooijer did you try downgrade or upgrade new version python in your cluster? base on ur screenshoot look like bug python version 3.9

Re: Python job run error messages are unreadable

Md_Tamim_Hosen — Mon, 28 Nov 2022 04:09:00 GMT

yeah

Re: Python job run error messages are unreadable

VaibB — Fri, 02 Dec 2022 20:03:34 GMT

Have you tried detaching and reattaching the notebook? Or Cluster restart?

Did you check you are not importing any specific library someone else with the right access might have installed some library with install to all clusters as checked.