Tracking your origin's name, IP, and port
Last updated October 30, 2018
Being able to track information related to your origin can be helpful in troubleshooting errors and making sure requests are processed as expected. Fastly provides three VCL variables that allow you to see and track origin information:
You can create VCL Snippets that use these variables to capture information about an origin and then set up remote log streaming to save that information.
Retrieve the origin information
Create a regular VCL Snippet to retrieve the origin information:
- Log in to the Fastly web interface and click the Configure link.
- From the service menu, select the appropriate service.
- Click the Edit configuration button and then select Clone active. The Domains page appears.
- Click the VCL snippets link. The VCL snippets page appears.
- Click Create snippet. The Create a VCL snippet page appears.
- In the Name field, type an appropriate name (for example,
Retrieve-Origin-Information
). - From the Type controls, select within subroutine. The Select subroutine menu appears.
- From the Select subroutine menu, select fetch (vcl_fetch).
-
In the VCL field, add the following VCL logic:
1 2 3
# save the variables for access in deliver set beresp.http.Your-Backend-Name = beresp.backend.name; set beresp.http.Your-Backend-IP-Port = beresp.backend.ip ":" beresp.backend.port;
- Click Create to create the snippet.
Change the response header to a request header
Create another regular VCL Snippet that changes the response header to a request header before logging the information:
- Click the VCL snippets link. The VCL snippets page appears.
- Click Create snippet. The Create a VCL snippet page appears.
- In the Name field, type an appropriate name (for example,
Remove-Origin-Information
). - From the Type controls, select within subroutine. The Select subroutine menu appears.
- From the Select subroutine menu, select deliver (vcl_deliver).
-
In the VCL field, add the following VCL logic:
1 2 3 4 5 6 7 8 9 10 11
if (fastly_info.state ~ "(MISS|PASS)") { # only on a miss or pass # save the responses back to req.request because # request headers are not sent back to the client set req.http.Your-Backend-Name = resp.http.Your-Backend-Name; set req.http.Your-Backend-IP-Port = resp.http.Your-Backend-IP-Port; } # remove the identifying information from the response unset resp.http.Your-Backend-Name; unset resp.http.Your-Backend-IP-Port;
- Click Create to create the snippet.
EXPLANATION: If a response header like beresp.http.Your-Backend-Name
exists in vcl_fetch
when VCL is processed, then a corresponding response header resp.http.Your-Backend-Name
will exist in vcl_deliver
and will be logged by default. This means that response headers will be included in response output, exposing origin information (e.g., IP address, port number, or origin name) in the process. Fortunately, request headers are not passed back to the client, but their information remains accessible from vcl_log
.
Set up remote log streaming
Once your snippets are created, you can set up remote log streaming for tracking purposes. Do this by adding req.http.Header-Name
to the Log format field in the logging endpoint you configured in the remote log streaming guide. Using the example above, you would add req.http.Your-Backend-Name
and req.http.Your-Backend-IP-Port
. Once you've reviewed the changes you've made, click the Activate button to deploy these configuration changes for your service.
NOTE: When you configure remote log streaming to capture this information, remember the VCL is executed twice, once on the shield node and again on the edge node. This means that the first log (from the shield node) will have the origin's information and the second log (from the edge node) will have the shield's information.