{
  "id": 130148,
  "title": "All attempts to get a Google authentication bearer token failed",
  "url": "/competitions/flower-classification-with-tpus/discussion/130148",
  "author_name": "Marcos Martins Marchetti",
  "post_date": "2020-02-12T12:40:10.167000",
  "votes": 3,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I have noticed this error message in the Log every time I commit. Is it normal or am I forgetting something?</p>",
  "messages": [
    {
      "id": 743974,
      "postDate": "2020-02-12T12:57:01.467Z",
      "content": "<p>Let me paste the error code:</p>\n\n<p><code>tensorflow/core/platform/cloud/google_auth_provider.cc:178] All attempts to get a Google authentication bearer token failed, returning an empty token. Retrieving token from files failed with \"Not found: Could not locate the credentials file.\". Retrieving token from GCE failed with \"Aborted: All 10 retry attempts failed. The last failure: Unavailable: Error executing an HTTP request: libcurl code 6 meaning 'Couldn't resolve host name', error details: Couldn't resolve host 'metadata'\".</code></p>",
      "rawMarkdown": "Let me paste the error code:\n\n`tensorflow/core/platform/cloud/google_auth_provider.cc:178] All attempts to get a Google authentication bearer token failed, returning an empty token. Retrieving token from files failed with \"Not found: Could not locate the credentials file.\". Retrieving token from GCE failed with \"Aborted: All 10 retry attempts failed. The last failure: Unavailable: Error executing an HTTP request: libcurl code 6 meaning 'Couldn't resolve host name', error details: Couldn't resolve host 'metadata'\".`",
      "votes": 3,
      "replies": [
        {
          "id": 744336,
          "postDate": "2020-02-12T18:47:12.927Z",
          "content": "<p>This error message is expected and harmless when running Tensorflow on Kaggle. (And yes, we should find a way of getting rid of it...)</p>\n\n<p>The full explanation is that Tensorflow is trying to contact the GCP 'metadata' server. That's useful if you are running on GCP and need to access your other GCP resources like GCS buckets for example. This feature is what makes access to private GCP resources \"just work\" inside a single GCP project.</p>",
          "rawMarkdown": "This error message is expected and harmless when running Tensorflow on Kaggle. (And yes, we should find a way of getting rid of it...)\n\nThe full explanation is that Tensorflow is trying to contact the GCP 'metadata' server. That's useful if you are running on GCP and need to access your other GCP resources like GCS buckets for example. This feature is what makes access to private GCP resources \"just work\" inside a single GCP project.",
          "votes": 4
        }
      ]
    },
    {
      "id": 743950,
      "postDate": "2020-02-12T12:40:10.167Z",
      "content": "<p>I have noticed this error message in the Log every time I commit. Is it normal or am I forgetting something?</p>",
      "rawMarkdown": "I have noticed this error message in the Log every time I commit. Is it normal or am I forgetting something?",
      "votes": 3
    },
    {
      "id": 2081700,
      "postDate": "2022-12-31T14:28:33.883Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 743974,
      "author_name": "Pawel Lubiński",
      "author_url": "",
      "post_date": "2020-02-12T12:57:01.467000",
      "content": "<p>Let me paste the error code:</p>\n\n<p><code>tensorflow/core/platform/cloud/google_auth_provider.cc:178] All attempts to get a Google authentication bearer token failed, returning an empty token. Retrieving token from files failed with \"Not found: Could not locate the credentials file.\". Retrieving token from GCE failed with \"Aborted: All 10 retry attempts failed. The last failure: Unavailable: Error executing an HTTP request: libcurl code 6 meaning 'Couldn't resolve host name', error details: Couldn't resolve host 'metadata'\".</code></p>",
      "votes": 3,
      "replies": [
        {
          "id": 744336,
          "author_name": "Martin Görner",
          "author_url": "",
          "post_date": "2020-02-12T18:47:12.927000",
          "content": "<p>This error message is expected and harmless when running Tensorflow on Kaggle. (And yes, we should find a way of getting rid of it...)</p>\n\n<p>The full explanation is that Tensorflow is trying to contact the GCP 'metadata' server. That's useful if you are running on GCP and need to access your other GCP resources like GCS buckets for example. This feature is what makes access to private GCP resources \"just work\" inside a single GCP project.</p>",
          "votes": 4,
          "replies": []
        }
      ]
    },
    {
      "id": 2081700,
      "author_name": "",
      "author_url": "",
      "post_date": "2022-12-31T14:28:33.883000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "743974": "Let me paste the error code:\n\n`tensorflow/core/platform/cloud/google_auth_provider.cc:178] All attempts to get a Google authentication bearer token failed, returning an empty token. Retrieving token from files failed with \"Not found: Could not locate the credentials file.\". Retrieving token from GCE failed with \"Aborted: All 10 retry attempts failed. The last failure: Unavailable: Error executing an HTTP request: libcurl code 6 meaning 'Couldn't resolve host name', error details: Couldn't resolve host 'metadata'\".`",
    "743950": "I have noticed this error message in the Log every time I commit. Is it normal or am I forgetting something?",
    "2081700": ""
  }
}