I've had WordPress-mu 1.2.1 installed for at least a month running as a beta service where I work. On Tuesday one of our admins upgraded the webserver housing wordpress-mu which involved a change to Apache 2.2.4 and PHP 5.2.2. Within the next 24 hours we noticed that the Magpie feed client wasn't picking up new posts even though they were appearing in the feed. Within the next hour the Magpie cache expired and we realized that the reader was timing out while trying to read the feed. Upon further investigation I noticed that there is extraneous data after the header and before the body and again after the body of the HTTP response. It looks as if there is a two byte hexadecimal string between the header and body (e.g. 29e6) and a 0 combined with linefeeds and newlines after the body of the data. The hexadecimal string seems to be unique depending on the URL.
I've looked though the code and diff'd it with a clean copy to ensure nothing was touched in Wordpress-mu. As well I've checked the Apache, Wordpress-MU, and PHP config files to make sure they were not changed. We are using the sub-directory method of hosting blogs.
I'm writing to see if anyone has had similar experiences or any ideas as to what may be causing this. Also, the time outs are only evident in certain clients (notably wget and Magpie). The garbage data does not seem to be present in wpmu 404 pages or in static pages. I appreciate any insight anyone can offer on this.