Colabtools: successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero

Created on 19 Aug 2019  路  4Comments  路  Source: googlecolab/colabtools

Bug report for Colab: http://colab.research.google.com/.

For questions about colab usage, please use stackoverflow.

  • Describe the current behavior: I am trying to build a UNet like Autoencoder for image segmentation. I am using Google Colab and Keras (TensorFlow backend) for the same. However, while training the model, the session crashes in the first epoch itself. I analyzed the colab-jupyter.log and observed that the root cause of crash is successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero

  • The web browser you are using (Chrome, Firefox, Safari, etc.): The issue occurred both on Chrome and Firefox.

  • Link to self-contained notebook that reproduces this issue
    (click the Share button, then Get Shareable Link): https://colab.research.google.com/drive/1wkfOmW82VCSSOGLjtEO0k3aHgYvxbeeV

Most helpful comment

Hi all,
If you found any solution please let me know. I am working with tensorflow and getting same error.

successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero

All 4 comments

Are there any updates to this issue yet?

Not yet. :(

The log line in question is an INFO log and does not in itself signify a problem.

The notebook in the original report is not self-contained (it depends on the contents of your Google Drive). If you are able to, please share a minimal and self-contained notebook that reproduces the issue.

Hi all,
If you found any solution please let me know. I am working with tensorflow and getting same error.

successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero

Was this page helpful?
0 / 5 - 0 ratings