Skip to main content
Question

PyCelonis - "500 Server Error: Internal Server Error"

  • November 30, 2022
  • 3 replies
  • 17 views
F
Level 1
Forum|alt.badge.img+4

Hi all,

I am currently trying to import data into the ML workbench using pycelonis get_data_frame. However, I am receiving the error "500 Server Error: Internal Server Error". It seems that my instance has no permissions for POST/GET in the API. I have added all thinkable permissions to the App key, used my personal admin user key etc., unfortunately, all without success. Does anyone have any clue what the issue could be? The same set-up works perfectly fine for another Celonis instance that we have commissioned.

Much appreciated and kind regards,

Florian

    3 replies

    F
    Level 1
    Forum|alt.badge.img+4
    • florian.schop
    • Author
    • Level 1
    • November 30, 2022

    pycelonis_error

    D
    Level 0

    Hello Florain,

     

    FIrst question: which version of PyCelonis are you using? 2.0.0 is quite unusable and unstable, if so downgrade to newest old version 1.7.4. by running command !PIP INSTALL PYCELONIS==1.7.2. or just run it without ! within the terminal.

     

    Second question: is there need to use that certain function? Can you bring more details from where exactly do you want to get the data? Is that from Analysis component, backend table or other source?

    Probably I can assume you're trying to get it from the Data Model table.

     

    Third question: how much data are your try to retrive? I can suggest using chunksize option, it's needed if you're trying to get more then 1gb size table. Timeout mentioned in the error message means that probably the response was to big to tackle :)

     

    df = datamodel.get_data_frame(q, chunksize=10000)

    df.head()

     

    Best Regards,

    Mateusz Dudek

     

     

    F
    Level 1
    Forum|alt.badge.img+4
    • florian.schop
    • Author
    • Level 1
    • December 2, 2022

    Hi Mateusz,

     

    Thank you very much for your response,

     

    First question: ML Workbench is running at 1.7.2. I tried multiple versions locally. The issue persists in both environments.

     

    Second question: The data resides in a Celonis data model. I am able to query the model, table and its fields, but unable to pull the data itself. Please find a screenshot attached with the console outputs of the queries as well as the error,

     

    Third question: I tried with and without chunksizing. The issue persists with tables <1k rows. So I assume that is not the underlying issue.

     

    As mentioned I tried this with an APP key as well as my personal admin key. Are there any further settings with regards to API and being able to actually pull data, which I am missing? I assume the admin personal key should already have all those permissions.

     

    Much appreciated and kind regards,

    Florianpycelonis_example_snap

    Community Code of ConductAccessibility statement