- From: <sird@rckc.at>
- Date: Mon, 7 Dec 2009 00:19:59 +0800
- To: Thomas Roessler <tlr@w3.org>
- Cc: public-web-security@w3.org
- Message-ID: <8ba534860912060819y3626f194la677a8a642b4a1bd@mail.gmail.com>
Comments.. 1.- The section 4.6.2 on step 5 I think should include other headers like User_Agent (with _ instead of -), Content_Length, etc.. Same for Proxy_ and Sec_ since Apache sortof sucks.. and Range/Request Range, and etc.. as shown by kuza55 on some ppt some time ago. 2.- 4.6.3 is not clear. It is obvious the UA should check first for which type of authentication, but then if I read correctly you allow the script to set their own Authentication header via setRequestHeader.. but if the header is missing then you fall down to the 4th and 5th arguments of open. This makes the UA to make 2 requests [one to know the auth method and the other to do the real request]? Both requests have the data sent by send() (before and after 401)? What about redirects that require different Authentication methods? If the user is now under (for example) a digest auth session, but the page/redirected page responds with Authentication: Basic, does the UA should prompt the user for user/password again? This is a dangerous downgrade attack (think active network attackers). If the session already has a username/password HTTP auth session and open() has user/pass? it should be replaced by the new one? Are you sure? Are you really sure? There are several attack scenarios there.. and unless I missed something in my opinion the specification is not specific enough =/ 3.- Do you really want to return to the user ALL http headers with getAllResponseHeaders? think on Set-Cookie + httpOnly Anyway.. just a few thoughts.. Greetings!! -- Eduardo http://www.sirdarckcat.net/ Sent from Hangzhou, Zhejiang, China On Sun, Dec 6, 2009 at 8:35 PM, Thomas Roessler <tlr@w3.org> wrote: > The XMLHttpRequest spec is in Last Call till 16 December: > > > XMLHttpRequest > > W3C Working Draft 19 November 2009 > > This Version: > > http://www.w3.org/TR/2009/WD-XMLHttpRequest-20091119/ > > A review from a security perspective would be a Good Thing. > > Particularly interesting pieces: > > - this is the place where the same origin policy for XMLHttpRequest is > defined > - behavior upon redirects > - needs security considerations on, e.g., DNS rebinding > > Any takers? > > Thanks, > -- > Thomas Roessler, W3C <tlr@w3.org> > > > > > > > > >
Received on Sunday, 6 December 2009 16:20:59 UTC