I have posted previously about similar issues with our usage of the REST API but I believe this may be different. We built a project in our development environment and it worked really well, but now that we've gone to our production OpenText environment, we are seeing completely random 500 errors on the auth call. Usually if we retry the call, it clears up. Each call we have done is using the same credentials and method of authenticating, yet roughly 10 - 20% (out of 1000s of calls) are failing intermittently.
One difference between OpenText DV and PD in our environment, is that PD has multiple front end servers, but we do NOT load balance. I have tried pointing to multiple front end servers and the behavior is the same, so I don't believe it is a problem isolated to one particular node.
The documentation indicates that the 500 error is triggered when credentials are not passed, but we are definitely passing them. We are hitting a virtual directory we set up called OTCSAPI and it only uses anonymous authentication. The top-level server setting only accepts anonymous as well.
Is there some type of throttling in Content Server that might be causing calls to fail when it is busy?
The anonymous user is the built-in IUSR account. Any ideas? The error we are getting is here - some of it is proprietary but it may help you:
Message: Method and/or Endpoint execution failed (Inner: 500 - 500 Internal Server Error); ServiceName: REST_otauth; ServiceGuid: 0cf4cde1-9414-44d6-8482-03a12fd1af97; InnerExceptionMessage: ;
We are passing in a connection string of username=usernamehere;password=passwordhere