I'm unable to reproduce this bug even though it has been reported by someone else also. Could you give some more info? What environment are you running (DB, OS, webserver, mongrel or fastcgi, ruby version, rails version, etc) and what do you do to produce the error (just loading the general settings page or setting a particular filter)?

Probably this problem is only in development mode.
it similar http://dev.rubyonrails.org/ticket/6039

# ./script/console development
Loading development environment.
allow_concurrency=false
SitealizerController: missing default helper path sitealizer_helper
>> Object.const_defined? :TextFilterPlugin
=> true
>> Dependencies.clear
=> []
>> Object.const_defined? :TextFilterPlugin
=> false
>> TextFilter.find(1) rescue $!.message
=> "superclass mismatch for class MacroPost" 

#1092 duplicate?

I'm getting this same error. I'm running on eApps VPS hosting (CentOS), using rails 1.2.3, ruby 1.8.4, MySQL 5.0.27, with apache and fastcgi. The error seems to occur after visiting the main admin page and submitting the form, or immediately upon login, and is reset when I restart apache... If I type in admin urls by hand (like /admin/contents) I'm able to edit/delete posts. It seems to still affect the system in production mode, but I'm not an expert with rails.. I ran rake environment RAILS_ENV=production. The blog itself seems to break sometimes when the rest of the system breaks, but I haven't found a pattern for this. I'm still able to edit blog entries with xmlrpc.

Oops, the same error occurs when trying to view a blog post after submitting it from TextMate via xmlrpc. This is on the view post page, and the same error appears on the blog home page after this is triggered. On resetting apache, the public side of the blog is fixed and the article was saved.

Looks as tho its a rails problem:

[http://www.ruby-forum.com/topic/53563#30458]

This is fixed in SVN by loading the textfilter plugins first in environment.rb.