How will the administrator rectify an HTTP Callout recursion issue after configuration?

How will the administrator rectify an HTTP Callout recursion issue after configuration?

• A. In the HTTP Callout policy, change the parameters to cip(CLIENT.IP.DST) && HTTP.REQ.HEADER('Request').EQ('Callout Request')
• B. Change the expression of the Responder policy to HTTP.REQ.HEADER('Request').EQ('Callout Request') NOT && SYS.HTTP_CALLOUT (myCallout).CONTAINS('IP Matched')
• C. In the HTTP Callout policy, change the parameters to cip(CLIENT.IP.SRC) && HTTP.REQ.HEADER('Request').EQ('Callout Request')
• D. Change the expression of the Responder policy to HTTP.REQ.HEADER('Request').EQ('Callout Request')

Answer: (B)

Recursion happens when the response from an HTTP Callout ends up being treated again as a new trigger by the policy, creating a loop. The fix is to stop the Responder from firing on traffic that originated from the callout itself. Choosing the Responder policy expression to require the request header equals "Callout Request" and, at the same time, negate any match from the HTTP callout (not SYS.HTTP_CALLOUT(myCallout).CONTAINS('IP Matched')) achieves this. In practice, you still match legitimate client requests, but if the flow result came from the HTTP Callout and includes an IP Matched indication, the NOT condition prevents the Responder from acting. This breaks the cycle by ensuring callout-generated traffic isn’t reprocessed by the Responder. The other options don’t address this loop condition: they either change the IP parameters for the callout or remove the exclusion that prevents the Responder from handling callout-originated traffic, which wouldn’t stop the recursion and could lead to missed decisions or further calls.

Recursion happens when the response from an HTTP Callout ends up being treated again as a new trigger by the policy, creating a loop. The fix is to stop the Responder from firing on traffic that originated from the callout itself.

Choosing the Responder policy expression to require the request header equals "Callout Request" and, at the same time, negate any match from the HTTP callout (not SYS.HTTP_CALLOUT(myCallout).CONTAINS('IP Matched')) achieves this. In practice, you still match legitimate client requests, but if the flow result came from the HTTP Callout and includes an IP Matched indication, the NOT condition prevents the Responder from acting. This breaks the cycle by ensuring callout-generated traffic isn’t reprocessed by the Responder.

The other options don’t address this loop condition: they either change the IP parameters for the callout or remove the exclusion that prevents the Responder from handling callout-originated traffic, which wouldn’t stop the recursion and could lead to missed decisions or further calls.