In the installation guide, we are instructed to modify the cd_dynamic_conf.xml of the website which is to be Experience-Manager-Enabled. The example configuration and explanation are as follows:

      <Publication Id="23">
        <Host Domain="example.com" Port="91" Protocol="http" Path="/local-fr" />
  <StorageMapping IdentifyPublicationByProperty="publicationUrl" />

This example makes it possible to retrieve metadata for a Web page located in the http://example.com:91/local-fr/ by checking the corresponding Page in the Publication with ID 23 in the Content Manager.

So far I have not been able to find documentation that explains what these settings do, or more to the point, what contribution they make to getting Experience manager working. I have managed to get this working by specifying the domain name of my website in the Domain attribute, and leaving the Path empty.

  1. What part does the dynamic conf play in the workings of Experience Manager?
  2. How can I verify this part of my setup, isolated from whether Experience Manager as a whole works
  3. What symptoms would Experience Manager display if this were not correctly configured?
  4. Where might I find conceptual explanations of what is going on here?

1 Answer 1


There are 2 things that cd_dynamic_conf is used for in Experience Manager:

  1. Determine the current site's publication context
  2. Execute TCDL stuff when "publishing" via Session Preview

The first part you pretty much cover already in your question, though you should be aware that you can also let Tridion find it out by itself (like CWA, as long as you have a unique path to that publication) by leaving the "StaticMappings" element empty.

The TCDL execution is done by the "Dynamic" engine (or REL for the friends), since you don't have a deployer anymore between CM and CD. REL will execute dynamic links and such other constructs (DCPs, etc) without transforming them first into Java or .NET code, which the deployer would have.

  • Aha! - so the dynamic conf controls the execution of REL. Why does it need the domain? And why can you specify a path? This looks suspiciously like cd_link_conf Aug 22, 2013 at 14:42
  • As I said, you don't need to specify any of that if you use the dynamic mappings instead of static mappings. It needs to be able to find the publication by url, that's why it needs the domain & path.
    – Nuno Linhares
    Aug 22, 2013 at 18:01
  • And "knowing" the publication allows XM to pick up pages/components from the right place in the blueprint, I suppose? Aug 23, 2013 at 8:16
  • Also allows it to know which Publication Target to use - and indirectly which Web Service to use for Session Preview.
    – Nuno Linhares
    Aug 23, 2013 at 11:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.