Assessing Programmatic Usage of X-Request-Id
125 views
Skip to first unread message
Dennis Patterson (Cerner)
Dec 14, 2021, 11:42:28 AM12/14/21
to Cerner FHIR Developers
General development principles support assigning a unique id per transaction for logging/support purposes. Further, the FHIR spec provides some recommendations around how this is accomplished.
This post focuses in on the X-Request-Id header, which is included on all of our FHIR HTTP responses. While the spec makes a provision for cases where a client may want to supply a value, the server ultimately is allowed to assign its own value in the response.
The ask is to reach out if 1) you have a use-case to supply an X-Request-Id with request to the FHIR server and 2) if you have expectations around the response X-Request-Id matching what was sent in the request.
No doubt many of you are using X-Request-Id to interact with us for troubleshooting purposes, but this post is specifically asking to if anybody has logic to consume the X-Request-Id response header with any expectations of a predictable value.
Thank you!
This post focuses in on the X-Request-Id header, which is included on all of our FHIR HTTP responses. While the spec makes a provision for cases where a client may want to supply a value, the server ultimately is allowed to assign its own value in the response.
The ask is to reach out if 1) you have a use-case to supply an X-Request-Id with request to the FHIR server and 2) if you have expectations around the response X-Request-Id matching what was sent in the request.
No doubt many of you are using X-Request-Id to interact with us for troubleshooting purposes, but this post is specifically asking to if anybody has logic to consume the X-Request-Id response header with any expectations of a predictable value.
Thank you!
Reply all
Reply to author
Forward
0 new messages