> ## Documentation Index
> Fetch the complete documentation index at: https://docs.open-metadata.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Run the GCS Connector Externally

> Use YAML to configure GCS storage ingestion for metadata visibility into cloud buckets.

export const CodePanel = ({children, fileName = 'config.yaml', showLineNumbers = false}) => {
  const codePanelRef = useRef(null);
  const codeContentRef = useRef(null);
  const isProgrammaticScroll = useRef(false);
  const hoverTimeout = useRef(null);
  useEffect(() => {
    let tries = 0;
    const wrapLines = () => {
      const root = codeContentRef.current;
      if (!root) return;
      const pres = Array.from(root.querySelectorAll('pre'));
      if (!pres.length) {
        if (tries++ < 20) requestAnimationFrame(wrapLines);
        return;
      }
      let globalLine = 1;
      pres.forEach(pre => {
        const code = pre.querySelector('code') || pre;
        if (!code || code.dataset.wrapped === 'true') return;
        const raw = code.textContent || '';
        let lines = raw.split('\n');
        while (lines[0] === '') lines.shift();
        while (lines[lines.length - 1] === '') lines.pop();
        code.innerHTML = lines.map(line => {
          const ln = globalLine++;
          const num = showLineNumbers ? `<span class="line-number">${ln}</span>` : '';
          const safe = line.replace(/</g, '&lt;').replace(/>/g, '&gt;') || ' ';
          return `<span class="code-line" data-line="${ln}">${num}${safe}</span>`;
        }).join('');
        code.dataset.wrapped = 'true';
      });
    };
    wrapLines();
  }, [children, showLineNumbers]);
  useEffect(() => {
    const panel = codePanelRef.current;
    const content = codeContentRef.current;
    if (!panel || !content) return;
    const waitForLines = () => {
      const codeLines = content.querySelectorAll('.code-line');
      if (!codeLines.length) {
        requestAnimationFrame(waitForLines);
        return;
      }
      setupHighlighting(codeLines);
    };
    const setupHighlighting = codeLines => {
      const layout = panel.closest('.split-layout');
      const sections = layout.querySelectorAll('.content-section');
      const parseLines = str => {
        if (!str) return [];
        const out = [];
        str.split(',').forEach(p => {
          if (p.includes('-')) {
            const [s, e] = p.split('-').map(Number);
            for (let i = s; i <= e; i++) out.push(i);
          } else {
            const n = Number(p);
            if (!isNaN(n)) out.push(n);
          }
        });
        return out;
      };
      const clearHighlight = () => {
        codeLines.forEach(l => l.classList.remove('highlighted'));
      };
      const highlight = lines => {
        clearHighlight();
        lines.forEach(n => {
          const el = content.querySelector(`.code-line[data-line="${n}"]`);
          if (el) el.classList.add('highlighted');
        });
      };
      const scrollToLines = lines => {
        if (!lines.length) return;
        const first = lines[0];
        const targetLine = lines.length > 1 ? first : lines[0];
        const el = content.querySelector(`.code-line[data-line="${targetLine}"]`);
        if (!el) return;
        isProgrammaticScroll.current = true;
        const containerRect = content.getBoundingClientRect();
        const elRect = el.getBoundingClientRect();
        const offset = elRect.top - containerRect.top + content.scrollTop;
        const TOP_PADDING = 16;
        content.scrollTo({
          top: Math.max(offset - TOP_PADDING, 0),
          behavior: 'smooth'
        });
        setTimeout(() => {
          isProgrammaticScroll.current = false;
        }, 200);
      };
      const activate = (section, scroll) => {
        if (section.classList.contains('active')) return;
        sections.forEach(s => s.classList.remove('active'));
        section.classList.add('active');
        const lines = parseLines(section.dataset.lines);
        highlight(lines);
        if (scroll) scrollToLines(lines);
      };
      const observer = new IntersectionObserver(entries => {
        if (isProgrammaticScroll.current) return;
        entries.forEach(e => {
          if (e.isIntersecting) activate(e.target, false);
        });
      }, {
        threshold: 0.3,
        rootMargin: '-80px 0px -40% 0px'
      });
      sections.forEach(section => {
        observer.observe(section);
        section.addEventListener('click', () => activate(section, true));
        section.addEventListener('mouseenter', () => {
          clearTimeout(hoverTimeout.current);
          hoverTimeout.current = setTimeout(() => activate(section, true), 80);
        });
      });
      if (sections[0]) activate(sections[0], false);
    };
    waitForLines();
  }, []);
  const handleCopy = e => {
    const btn = e.currentTarget;
    const codeLines = codeContentRef.current?.querySelectorAll('.code-line');
    if (!codeLines || codeLines.length === 0) return;
    const text = Array.from(codeLines).map(line => {
      const clone = line.cloneNode(true);
      const lineNumber = clone.querySelector('.line-number');
      if (lineNumber) lineNumber.remove();
      return clone.textContent;
    }).join('\n');
    if (!text) return;
    navigator.clipboard.writeText(text).then(() => {
      btn.dataset.copied = 'true';
      setTimeout(() => btn.dataset.copied = 'false', 1500);
    });
  };
  return <div className="code-panel" ref={codePanelRef}>
      <div className="code-header">
        {fileName}
        <button className="copy-btn" aria-label="Copy full code" data-copied="false" onClick={handleCopy}>
          <svg className="icon-copy" viewBox="0 0 15 16" fill="currentColor">
            <path d="M10.113 3.124H2.205C1.463 3.124.86 3.655.86 4.31v10.005c0 .654.603 1.186 1.345 1.186h7.908c.742 0 1.345-.532 1.345-1.186V4.31c0-.655-.606-1.186-1.345-1.186Z" />
            <path d="M13.138.5H5.229c-.742 0-1.344.531-1.344 1.186 0 .23.209.414.47.414s.47-.184.47-.414c0-.197.182-.357.404-.357h7.909c.223 0 .404.16.404.357V11.69c0 .196-.181.356-.404.356-.262 0-.47.184-.47.415 0 .23.208.415.47.415.742 0 1.344-.532 1.344-1.186V1.686C14.482 1.03 13.88.5 13.138.5Z" />
          </svg>

          <svg className="icon-check" viewBox="0 0 20 20" fill="currentColor">
            <path fillRule="evenodd" d="M16.707 5.293a1 1 0 010 1.414l-7.25 7.25a1 1 0 01-1.414 0l-3.25-3.25a1 1 0 011.414-1.414l2.543 2.543 6.543-6.543a1 1 0 011.414 0z" clipRule="evenodd" />
          </svg>
        </button>
      </div>

      <div className="code-content" ref={codeContentRef}>
        {children}
      </div>
    </div>;
};

export const ContentSection = ({id, title, lines, children}) => <div className="content-section" data-content-id={id} data-lines={lines}>
    {title && <h4>{title}</h4>}
    {children}
  </div>;

export const ContentPanel = ({children}) => <div className="content-panel">{children}</div>;

export const CodePreview = ({children}) => {
  const [instanceId] = useState(() => `preview-${Math.random().toString(36).slice(2)}`);
  useEffect(() => {
    const nav = document.querySelector('nav') || document.querySelector('header') || document.querySelector('[class*="nav"]');
    if (nav) {
      document.documentElement.style.setProperty('--navbar-height', `${nav.offsetHeight}px`);
    }
  }, []);
  return <div className="split-layout" data-preview-id={instanceId}>
      {children}
    </div>;
};

export const ConnectorDetailsHeader = ({name, icon, stage, availableFeatures, unavailableFeatures = [], availableFeaturesCollate = []}) => {
  const showSubHeading = availableFeatures?.length > 0 || unavailableFeatures?.length > 0 || availableFeaturesCollate?.length > 0;
  const totalAvailableFeatures = [...availableFeatures || [], ...availableFeaturesCollate || []];
  return <div className="container">
      <div className="Heading">
        <div className="flex items-center gap-3">
          {icon && <div className="IconContainer">
              <img src={icon} alt={name} noZoom className="ConnectorIcon" />
            </div>}
          <h1 className="ConnectorName">{name}</h1>
          <span className={`StageBadge ${stage === 'PROD' ? 'prod' : 'beta'}`}>
            {stage}
          </span>
        </div>
      </div>
      {showSubHeading && <div className="SubHeading">
          <div className="FeaturesHeading">Feature List</div>
          <div className="FeaturesList">
            {totalAvailableFeatures.map(feature => <div className="FeatureTag AvailableFeature" key={feature}>
                ✓ {feature}
              </div>)}
            {unavailableFeatures.map(feature => <div className="FeatureTag UnavailableFeature" key={feature}>
                ✕ {feature}
              </div>)}
          </div>
        </div>}
    </div>;
};

<ConnectorDetailsHeader icon="/public/images/connectors/gcs.webp" name="GCS" stage="PROD" availableFeatures={["Metadata"]} unavailableFeatures={[]} />

This page contains the setup guide and reference information for the GCS connector.
Configure and schedule GCS metadata workflows from the CLI:

* [Requirements](#requirements)
* [Metadata Ingestion](#metadata-ingestion)

## How to Run the Connector Externally

To run the Ingestion via the UI you'll need to use the OpenMetadata Ingestion Container, which comes shipped with
custom Airflow plugins to handle the workflow deployment.

If, instead, you want to manage your workflows externally on your preferred orchestrator, you can check
the following docs to run the Ingestion Framework **anywhere**.

<Columns cols={2}>
  <Card title="External Schedulers" href="/v1.12.x/deployment/ingestion">
    Get more information about running the Ingestion Framework Externally
  </Card>
</Columns>

## Requirements

To run the GCS ingestion, you will need to install:

```bash theme={null}
pip3 install "openmetadata-ingestion[datalake-gcs]"
```

<Card title="OpenMetadata 1.0 or later" href="/v1.12.x/deployment">
  To deploy OpenMetadata, check the Deployment guides.
</Card>

We need the following permissions in GCP:

### GCS Permissions

For all the buckets that we want to ingest, we need to provide the following:

* `storage.buckets.get`
* `storage.buckets.list`
* `storage.objects.get`
* `storage.objects.list`

### OpenMetadata Manifest

In any other connector, extracting metadata happens automatically. In this case, we will be able to extract high-level
metadata from buckets, but in order to understand their internal structure we need users to provide an `openmetadata.json`
file at the bucket root.
`Supported File Formats: [ "csv",  "tsv", "avro", "parquet", "json", "json.gz", "json.zip" ]`
You can learn more about this [here](/v1.12.x/connectors/storage). Keep reading for an example on the shape of the manifest file.

## OpenMetadata Manifest

Our manifest file is defined as a [JSON Schema](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/metadataIngestion/storage/containerMetadataConfig.json),
and can look like this:

<CodePreview>
  <ContentPanel>
    <ContentSection id={1} title="Entries" lines="1-3">
      **Entries**: We need to add a list of `entries`. Each inner JSON structure will be ingested as a child container of the top-level one. In this case, we will be ingesting 7 children.
    </ContentSection>

    <ContentSection id={2} title="Simple Container" lines="4-9">
      **Simple Container**: The simplest container we can have would be structured, but without partitions. Note that we still need to bring information about:

      * **dataPath**: Where we can find the data. This should be a path relative to the top-level container.
      * **structureFormat**: What is the format of the data we are going to find. This information will be used to read the data.
      * **separator**: Optionally, for delimiter-separated formats such as CSV, you can specify the separator to use when reading the file. If you don't, we will use `,` for CSV and `/t` for TSV files.

      After ingesting this container, we will bring in the schema of the data in the `dataPath`.
    </ContentSection>

    <ContentSection id={3} title="Partitioned Container" lines="10-14">
      **Partitioned Container**: We can ingest partitioned data without bringing in any further details.

      By informing the `isPartitioned` field as `true`, we'll flag the container as `Partitioned`. We will be reading the source files schemas', but won't add any other information.
    </ContentSection>

    <ContentSection id={4} title="Single-Partition Container" lines="15-23">
      **Single-Partition Container**: We can bring partition information by specifying the `partitionColumns`. Their definition is based on the [JSON Schema](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/entity/data/table.json#L232) definition for table columns. The minimum required information is the `name` and `dataType`.

      When passing `partitionColumns`, these values will be added to the schema, on top of the inferred information from the files.
    </ContentSection>

    <ContentSection id={5} title="Multiple-Partition Container" lines="24-33">
      **Multiple-Partition Container**: We can add multiple columns as partitions.

      Note how in the example we even bring our custom `displayName` for the column `dataTypeDisplay` for its type.

      Again, this information will be added on top of the inferred schema from the data files.
    </ContentSection>

    <ContentSection id={6} title="Automated Container Ingestion" lines="34-37">
      **Automated Container Ingestion**: Registering all the data paths one by one can be a time consuming job, to make the automated structure container ingestion you can provide the depth at which all the data is available.

      For example, suppose following is the file hierarchy within your bucket:

      ```
      # prefix/depth1/depth2/depth3
      athena_service/my_database_a/my_schema_a/table_a/date=01-01-2025/data.parquet
      athena_service/my_database_a/my_schema_a/table_a/date=02-01-2025/data.parquet
      athena_service/my_database_a/my_schema_a/table_b/date=01-01-2025/data.parquet
      athena_service/my_database_a/my_schema_a/table_b/date=02-01-2025/data.parquet

      athena_service/my_database_b/my_schema_a/table_a/date=01-01-2025/data.parquet
      athena_service/my_database_b/my_schema_a/table_a/date=02-01-2025/data.parquet
      athena_service/my_database_b/my_schema_a/table_b/date=01-01-2025/data.parquet
      athena_service/my_database_b/my_schema_a/table_b/date=02-01-2025/data.parquet
      ```

      All table folders containing actual data are at depth 3. When you specify `depth: 3` in the manifest entry, all following paths will get registered as containers in OpenMetadata with this single entry:

      ```
      athena_service/my_database_a/my_schema_a/table_a
      athena_service/my_database_a/my_schema_a/table_b
      athena_service/my_database_b/my_schema_a/table_a
      athena_service/my_database_b/my_schema_a/table_b
      ```

      This saves effort - 1 entry instead of 4 individual entries.
    </ContentSection>

    <ContentSection id={7} title="Unstructured Container" lines="38-45">
      **Unstructured Container**: OpenMetadata supports ingesting unstructured files like images, PDFs, etc. We support fetching the file names, size, and tags associated with such files.

      * To ingest a **single unstructured file**: specify the full path of the file in `dataPath`
      * To ingest **specific file types** (e.g., `pdf` & `png`): provide the folder name in `dataPath` and list of extensions in `unstructuredFormats`
      * To ingest **all unstructured files** regardless of type: provide the folder name in `dataPath` and `["*"]` in `unstructuredFormats`
    </ContentSection>
  </ContentPanel>

  <CodePanel fileName="openmetadata.json">
    ```json theme={null}
    {
      "entries": [
        {
          "dataPath": "transactions/",
          "structureFormat": "csv",
          "separator": ","
        },
        {
          "dataPath": "orders/",
          "structureFormat": "parquet",
          "isPartitioned": true
        },
        {
          "dataPath": "users/",
          "structureFormat": "parquet",
          "isPartitioned": true,
          "partitionColumns": [
            {
              "name": "signup_date",
              "dataType": "DATE"
            }
          ]
        },
        {
          "dataPath": "events/",
          "structureFormat": "parquet",
          "isPartitioned": true,
          "partitionColumns": [
            {
              "name": "event_date",
              "dataType": "DATE"
            },
            {
              "name": "region",
              "dataType": "STRING"
            }
          ]
        },
        {
          "depth": 3,
          "structureFormat": "parquet"
        },
        {
          "dataPath": "reports/report.pdf"
        },
        {
          "dataPath": "documents/",
          "unstructuredFormats": ["pdf", "png", "jpg"]
        }
      ]
    }
    ```
  </CodePanel>
</CodePreview>

### Global Manifest

You can also manage a **single** manifest file to centralize the ingestion process for any container, named `openmetadata_storage_manifest.json`.

<CodePreview>
  <ContentPanel>
    <ContentSection id={1} title="Existing Fields" lines="1-15">
      The fields shown above (`dataPath`, `structureFormat`, `isPartitioned`, etc.) are still valid and work the same way in the global manifest.
    </ContentSection>

    <ContentSection id={2} title="Container Name" lines="1-15">
      **Container Name**: Since we are using a single manifest for all your containers, the field `containerName` will help us identify which container (or Bucket in S3, etc.) contains the presented information.

      Each entry in the global manifest must include a `containerName` to specify which bucket or container it belongs to.
    </ContentSection>
  </ContentPanel>

  <CodePanel fileName="openmetadata_storage_manifest.json">
    ```json theme={null}
    {
      "entries": [
        {
          "containerName": "my-s3-bucket-1",
          "dataPath": "transactions/",
          "structureFormat": "csv",
          "separator": ","
        },
        {
          "containerName": "my-s3-bucket-1",
          "dataPath": "orders/",
          "structureFormat": "parquet",
          "isPartitioned": true
        },
        {
          "containerName": "my-s3-bucket-2",
          "dataPath": "users/",
          "structureFormat": "parquet",
          "isPartitioned": true,
          "partitionColumns": [
            {
              "name": "signup_date",
              "dataType": "DATE"
            }
          ]
        },
        {
          "containerName": "my-s3-bucket-2",
          "dataPath": "events/",
          "structureFormat": "json"
        }
      ]
    }
    ```
  </CodePanel>
</CodePreview>

You can also keep local manifests `openmetadata.json` in each container, but if possible, we will always try to pick up the global manifest during the ingestion.

## Metadata Ingestion

All connectors are defined as JSON Schemas.
[Here](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/entity/services/connections/storage/GCSConnection.json)
you can find the structure to create a connection to Athena.
In order to create and run a Metadata Ingestion workflow, we will follow
the steps to create a YAML configuration able to connect to the source,
process the Entities if needed, and reach the OpenMetadata server.
The workflow is modeled around the following
[JSON Schema](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/metadataIngestion/workflow.json)

### 1. Define the YAML Config

This is a sample config for Athena:

<CodePreview>
  <ContentPanel>
    <ContentSection id={1} title="gcpConfig:" lines="8-10">
      **gcpConfig:**
      **1.** Passing the raw credential values provided by GCP. This requires us to provide the following information, all provided by GCP:

      **gcpConfig:**

      * **type**: Credentials Type is the type of the account, for a service account the value of this field is `service_account`. To fetch this key, look for the value associated with the `type` key in the service account key file.
      * **projectId**: A project ID is a unique string used to differentiate your project from all others in Google Cloud. To fetch this key, look for the value associated with the `project_id` key in the service account key file. You can also pass multiple project id to ingest metadata from different BigQuery projects into one service.
      * **privateKeyId**: This is a unique identifier for the private key associated with the service account. To fetch this key, look for the value associated with the `private_key_id` key in the service account file.
      * **privateKey**: This is the private key associated with the service account that is used to authenticate and authorize access to BigQuery. To fetch this key, look for the value associated with the `private_key` key in the service account file.
      * **clientEmail**: This is the email address associated with the service account. To fetch this key, look for the value associated with the `client_email` key in the service account key file.
      * **clientId**: This is a unique identifier for the service account. To fetch this key, look for the value associated with the `client_id` key in the service account key  file.
      * **authUri**: This is the URI for the authorization server. To fetch this key, look for the value associated with the `auth_uri` key in the service account key file. The default value to Auth URI is [https://accounts.google.com/o/oauth2/auth](https://accounts.google.com/o/oauth2/auth).
      * **tokenUri**: The Google Cloud Token URI is a specific endpoint used to obtain an OAuth 2.0 access token from the Google Cloud IAM service. This token allows you to authenticate and access various Google Cloud resources and APIs that require authorization. To fetch this key, look for the value associated with the `token_uri` key in the service account credentials file. Default Value to Token URI is [https://oauth2.googleapis.com/token](https://oauth2.googleapis.com/token).
      * **authProviderX509CertUrl**: This is the URL of the certificate that verifies the authenticity of the authorization server. To fetch this key, look for the value associated with the `auth_provider_x509_cert_url` key in the service account key file. The Default value for Auth Provider X509Cert URL is [https://www.googleapis.com/oauth2/v1/certs](https://www.googleapis.com/oauth2/v1/certs)
      * **clientX509CertUrl**: This is the URL of the certificate that verifies the authenticity of the service account. To fetch this key, look for the value associated with the `client_x509_cert_url` key in the service account key  file.

      **2.**  Passing a local file path that contains the credentials:

      * **gcpCredentialsPath**
      * If you prefer to pass the credentials file, you can do so as follows:

      ```yaml theme={null}
      source:
        type: gcs
        serviceName: local_gcs
        serviceConnection:
          config:
            type: GCS
            credentials:
              gcpConfig:
              path: <path to file>
      ```

      * If you want to use [ADC authentication](https://cloud.google.com/docs/authentication#adc) for GCP you can just leave
        the GCP credentials empty. This is why they are not marked as required.

      ```yaml theme={null}
      ...
      source:
        type: gcs
        serviceName: local_gcs
        serviceConnection:
          config:
            type: GCS
          credentials:
            gcpConfig: {}
      ...
      ```
    </ContentSection>

    <ContentSection id={2} title="Connection Options (Optional)" lines="12">
      **Connection Options (Optional)**: Enter the details for any additional connection options that can be sent to storage service during the connection. These details must be added as Key-Value pairs.
    </ContentSection>

    <ContentSection id={3} title="Connection Arguments (Optional)" lines="14">
      **Connection Arguments (Optional)**: Enter the details for any additional connection arguments such as security or protocol configs that can be sent to storage service during the connection. These details must be added as Key-Value pairs.
    </ContentSection>
  </ContentPanel>

  <CodePanel fileName="connector_config.yaml">
    ```yaml theme={null}
    source:
      type: gcs
      serviceName: "<service name>"
      serviceConnection:
        config:
          type: GCS
    credentials:
            gcpConfig:
    # taxonomyLocation: us
          # taxonomyProjectID: ["project-id-1", "project-id-2"]
          # usageLocation: us
    # connectionOptions:
          #   key: value
    # connectionArguments:
          #   key: value
    ```

    ```yaml theme={null}
      sourceConfig:
        config:
          type: StorageMetadata
          # containerFilterPattern:
          #   includes:
          #     - container1
          #     - container2
          #   excludes:
          #     - container3
          #     - container4
          # storageMetadataConfigSource:
          ## For S3
          #   securityConfig:
          #     awsAccessKeyId: ...
          #     awsSecretAccessKey: ...
          #     awsRegion: ...
          #   prefixConfig:
          #     containerName: om-glue-test
          #     objectPrefix: <optional prefix>
          ## For HTTP
          #   manifestHttpPath: http://...
          ## For Local
          #   manifestFilePath: /path/to/openmetadata_storage_manifest.json
    ```

    ```yaml theme={null}
    sink:
      type: metadata-rest
      config: {}
    ```

    ```yaml theme={null}
    workflowConfig:
      loggerLevel: INFO  # DEBUG, INFO, WARNING or ERROR
      openMetadataServerConfig:
        hostPort: "http://localhost:8585/api"
        authProvider: openmetadata
        securityConfig:
          jwtToken: "{bot_jwt_token}"
        ## Store the service Connection information
        storeServiceConnection: true  # false
        ## Secrets Manager Configuration
        # secretsManagerProvider: aws, azure or noop
        # secretsManagerLoader: airflow or env
        ## If SSL, fill the following
        # verifySSL: validate  # or ignore
        # sslConfig:
        #   caCertificate: /local/path/to/certificate
    # ingestionPipelineFQN: <service name>.<ingestion name> ## e.g., "my_redshift.metadata"
    ```
  </CodePanel>
</CodePreview>