如何跨组织获取所有运行中应用？如何访问Cloud Foundry的Doppler URL？

Great questions! Let's break this down step by step for both manual CLI usage and API access, plus how to leverage Doppler for real-time staging event monitoring.

If you need to gather this data manually without scripting, follow these steps:

• First, log in without specifying an organization or space—this lets you access all orgs you have permissions for:

  cf login --skip-ssl-validation -a <CF_API_URL> -u <USERNAME> -p <PASSWORD>
  

• List all organizations:

  cf orgs
  

• For each organization, switch your target org, list its spaces, then switch to each space to view running apps:

  # Example workflow for a single org/space
  cf target -o <ORG_NAME>
  cf spaces
  cf target -s <SPACE_NAME>
  # Filter to only show running apps
  cf apps | grep "started"
  # Get full details for a specific app
  cf app <APP_NAME>
  

To avoid repetitive manual work, use a simple bash script to automate the full traversal:

#!/bin/bash
# Fetch all orgs (trim header/footer from cf orgs output)
ORGS=$(cf orgs | tail -n +4 | head -n -1)
for ORG in $ORGS; do
  echo "=== Scanning Organization: $ORG ==="
  cf target -o "$ORG"
  # Fetch all spaces in the org
  SPACES=$(cf spaces | tail -n +4 | head -n -1)
  for SPACE in $SPACES; do
    echo "--- Scanning Space: $SPACE ---"
    cf target -s "$SPACE"
    # List running apps with basic info
    cf apps | grep "started"
    # Uncomment below to get full details for each running app
    # APP_NAMES=$(cf apps | grep "started" | awk '{print $1}')
    # for APP in $APP_NAMES; do
    #   echo "===== App Details: $APP ====="
    #   cf app "$APP"
    # done
  done
done

Cloud Foundry provides a robust REST API for programmatic access, which is far more scalable as organizations/spaces change. Here's the workflow:

• Step 1: Retrieve an access token
  Authenticate to get a bearer token for API requests:

  curl -k -X POST "<CF_API_URL>/oauth/token" \
    -H "Content-Type: application/x-www-form-urlencoded" \
    -d "grant_type=password&username=<USERNAME>&password=<PASSWORD>&client_id=cf"
  

  Extract the access_token field from the JSON response.
• Step 2: Fetch all organizations

  curl -k -H "Authorization: Bearer <ACCESS_TOKEN>" "<CF_API_URL>/v2/organizations"
  

  Note the metadata.guid for each organization—this is needed to fetch its spaces.
• Step 3: Fetch spaces for each organization
  For each org GUID, get its associated spaces:

  curl -k -H "Authorization: Bearer <ACCESS_TOKEN>" "<CF_API_URL>/v2/organizations/<ORG_GUID>/spaces"
  

• Step 4: Fetch running apps for each space
  Filter apps to only include those in a STARTED state:

  curl -k -H "Authorization: Bearer <ACCESS_TOKEN>" "<CF_API_URL>/v2/spaces/<SPACE_GUID>/apps?q=state:STARTED"
  

  The response includes full app details like memory limits, routes, buildpack info, and metadata. You can also use the v3 API (/v3/apps) for newer features, but v2 is widely supported for basic app queries.

Doppler is Cloud Foundry's component for streaming logs and real-time events. You'll use WSS (WebSocket Secure) (not HTTPS) to maintain a persistent connection for live updates. Here's how to set this up:

• Step 1: Get the Doppler endpoint
  Retrieve the Doppler URL from the CF API info endpoint:

  curl -k "<CF_API_URL>/v2/info"
  

  Look for the doppler_logging_endpoint field—it will look like wss://doppler.your-cf-domain.com:443.
• Step 2: Connect to the Doppler Firehose
  To monitor global staging events, subscribe to the Firehose (requires admin-level permissions like firehose.admin or cloud_controller.admin). Use a WebSocket client like wscat (install via npm install -g wscat) to connect:

  wscat -k "Authorization: Bearer <ACCESS_TOKEN>" -c "wss://doppler.your-cf-domain.com:443/firehose?subscription_id=my-staging-monitor"
  

• Step 3: Filter for Staging Completed Events
  The Firehose streams all platform events. Look for messages where the event_type is StagingCompleted—these contain details about newly staged apps, including app GUID, organization/space metadata, staging status, and buildpack results. You can parse these messages programmatically (e.g., with a Node.js or Python WebSocket client) to automatically fetch new app details whenever staging finishes.

Key Notes: #

• Ensure your user account has the necessary permissions to access all organizations/spaces and the Firehose.
• For production use, implement reconnection logic for the Doppler WebSocket connection (since connections can drop unexpectedly).
• The cf logs command is for individual app logs—use the Firehose for global event monitoring.

内容的提问来源于stack exchange，提问作者overexchange