如何在Shell中编写if条件判断curl请求是否成功？及执行异常排查

Hey there! Let's work through your two curl and Shell scripting questions step by step.

When verifying if a curl request worked, you have two key angles to cover:

• Whether the curl command itself ran without execution errors (like network drops, DNS failures, or invalid URLs)
• Whether the HTTP response returns the expected success status code (most commonly 200)

Checking curl's Exit Code (Command Execution Success) #

Curl returns an exit code of 0 if the transfer completes without technical errors—even if the HTTP status is 4xx or 5xx. You can use this directly in an if statement:

if curl -s -o /dev/null "https://your-target-url.com"; then
    echo "Curl command executed successfully (transfer completed)"
else
    echo "Curl failed (network error, DNS issue, etc.)"
fi

The -s flag silences noisy progress output, and -o /dev/null discards the response body so we only focus on execution success.

Checking for a Specific HTTP Status Code (e.g., 200) #

If you need to confirm the HTTP response is actually successful (status 200), use curl's -w flag to print the status code, then validate it:

HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" "https://your-target-url.com")
if [[ "$HTTP_STATUS" -eq 200 ]]; then
    echo "Request succeeded with HTTP 200"
else
    echo "Request returned status code: $HTTP_STATUS"
fi

Or combine it into a single if check with grep for brevity:

if curl -s -o /dev/null -w "%{http_code}" "https://your-target-url.com" | grep -q "200"; then
    echo "HTTP 200 success"
else
    echo "Non-200 response or curl failure"
fi

If curl returns 200 when you run it manually, but a script or specific execution flow outputs "Fail", here are the most likely causes and debug steps:

• Mismatched command parameters: Double-check that the curl command in your script/specific run is identical to the one you test manually. Even small differences—like an extra proxy flag (-x), a missing header (-H), or a typo in the URL—can lead to unexpected behavior.
• Incorrect success logic in the script: The "Fail" message is almost certainly coming from your own conditional code, not curl itself. For example:

  # This script checks for specific content, not just HTTP 200
  RESPONSE_BODY=$(curl -s "https://your-url.com")
  if [[ "$RESPONSE_BODY" != "Expected Success Message" ]]; then
      echo "Fail"
  fi
  

  Here, even with a 200 status, if the response body doesn't match what the script expects, it will output "Fail".
• Environment differences: When running in a script (especially via cron, sudo, or a different user), the environment might differ from your interactive shell. Verify:
  • Curl version: Run curl --version both manually and in the script to confirm they're the same.
  • Proxy settings: Your interactive shell might have http_proxy set, but the script doesn't (or vice versa).
  • Permissions: If the script runs as a different user, it might lack access to SSL certificates or other resources needed for the request.
• Alias or shell override: If you've set an alias for curl in your interactive shell (e.g., alias curl='curl -v'), the script might use the raw curl binary without the alias. Test this by using the full path to curl (/usr/bin/curl) in both places.

Quick Debugging Trick #

Add debug output to your script to see exactly what's happening:

set -x  # Enables debug mode to print every command executed
curl -s -o /dev/null -w "%{http_code}" "https://your-url.com"
# Insert your existing logic here
set +x  # Disable debug mode

This will show you the exact curl command being run, the HTTP status code returned, and where the "Fail" trigger is coming from.

内容的提问来源于stack exchange，提问作者Moose