I'm having a weird problem. I have two code bases on one server. One is cutting edge stuff in my home directory. The other is a more stable release in the /var/www/... directory. I have noticed that when I pull the code from git to 开发者_开发知识库my home directory, the other code base picks up some functionality of the fresh code. The about page stays the same, but new features are visible on the stable code base.
The two codebases each have their own subdomain, sessions are stored in the databases, which are separate, and the cookies have unique names. It's not a browser cache issue - the code is server side and the sites have distinct colors and logos.
There is no memcache, apc, or eaccellerator installed.
Here's where it gets weirder. If I rename the cutting edge code directory in my home directory, and then rename it back, the stable code base goes back to behaving normally. (Unless I've made a database change. Then, I have to clear the model cache in the stable code's file system)
The code in my home directory is a git clone. The code in the stable directory is a copy of the code in the home directory at some stable point in time (cp -av /home/foo /var/www/foo)
Edit: The actual files in the /var/www/foo directory do not change. If I open one up in vi, the code is the proper code.
Most likely candidate is you have some mod_rewrite rules, either in your Apache config files, or in .htaccess files.
精彩评论