I just ran into this problem so I think it is still valid. The problem definitely has to do with the PKI token size when using the Apache2 WSGI front end for Keystone. I played around with the token size by deleting the number of endpoints and found that the exact same REST API call which previously failed executed without the error. I contacted Graham Dumpleton on the mod_wsgi project and asked him if there were a way to increase the maximum size of the response header and received the following response:
"This is a hardwired limitation within the Apache function used by mod_wsgi to parse the response headers returned from a WSGI application running in daemon mode."
He does however have a fix up his sleeve and is thinking about making it.
I just ran into this problem so I think it is still valid. The problem definitely has to do with the PKI token size when using the Apache2 WSGI front end for Keystone. I played around with the token size by deleting the number of endpoints and found that the exact same REST API call which previously failed executed without the error. I contacted Graham Dumpleton on the mod_wsgi project and asked him if there were a way to increase the maximum size of the response header and received the following response:
"This is a hardwired limitation within the Apache function used by mod_wsgi to parse the response headers returned from a WSGI application running in daemon mode."
He does however have a fix up his sleeve and is thinking about making it.