You can generate robots.txt as part of your package build, and/or have a
scheduled task t oregenerate it.  then you can stick it on a fast non-ASPNET
application, or lighttpd, or CDN


On Wed, May 20, 2009 at 11:36 PM, Jimmy Shimizu <[email protected]>wrote:

> I would agree on first point, seems like unecessary overhead. I just want
> the possibility to add certain explicit filetypes without having to fool
> around with the config (like site-validation etc). I'm gonna write a
> controller-action for robots.txt anyway because I want that automated.
>
> so basically its *.html for google, aswell as *.ico for favicon (for the
> retarded browsers that doesn't support icon-filename in markup)
>
>
> On 20 maj 2009, at 20:37, Ken Egozi wrote:
>
> 1. isn't it better to have static files not reach MRHHF at all?
> 2. I really don't like the "if there's a real file, do not process MR
> logic".
> on many occasions, when server is not secured correctly, a loophole allows
> an attacker to upload files (not change or delete, but add) to the site.
>
> now if you have a url that triggers an MR action under
> "someurl.anextension", and you use the "if there's a file bypass MR",  all
> an attacker needs to do is to upload a malicious markup file named
> someurl.anextension, and you're in really big trouble.
>
> This kind of attack have happened to me and to people I know on several
> shared hosting scenarios. Usually the attacker adds the common default
> documents (index.htm) for that, and I solved that problem by removing all
> default documents from the IIS.
> however the said scenario will open up a whole new bunch of attack
> surfaces.
>
>
>
> On Wed, May 20, 2009 at 9:06 PM, Henrik Feldt <[email protected]> wrote:
>
>>
>> Yes, it's a part of it, but if monorail also has a hit on the url which
>> has
>> a static file, you can override the static file with the routing. I moved
>> a
>> lot of knowledge from the http handler to the routing module.
>>
>> Currently, routingmoduleex:
>> if (File.Exists(request.PhysicalPath))
>>                        {
>>                                return; // Possibly requesting a static
>> file, so we skip routing altogether
>>                        }
>>
>> But path="*" on mrhttphfac means it will try to find a controller for it
>> anyway. Instead of doing this in routingmoduleex I'm returning a route
>> with
>> a property isstaticfile which then the mrhandler delegates to a static
>> file
>> handler. Because the handler fac originally was written under the
>> assumption
>> it won't get static requests, it tries to find a controller/action for
>> this,
>> which obviously fails causing this annoying exception you mailed about
>> before.
>>
>> I'm still in the testing stages for this, because what I'm after is the
>> dynamic content compression in iis7 and I'm not sure how returning a
>> custom
>> handler for static files affects this.
>>
>> I extended the classes with a base class and different types of routing
>> matches.
>>
>> Regards,
>> Henrik
>>
>> -----Original Message-----
>> From: [email protected]
>> [mailto:[email protected]] On Behalf Of Jimmy Shimizu
>> Sent: den 20 maj 2009 14:54
>> To: [email protected]
>> Subject: Re: robots.txt and sitemaps with routing
>>
>>
>> What do you do when matching a static file? Do you explicitly check the
>> filesystem for existance?
>>
>> I think I'd still keep my static-folder for content that always will be
>> static (to avoid unnessecary processing of files through MR), but for
>> certain root-files this approach isn't applicable.
>>
>> Henrik Feldt wrote:
>> > Hello Jimmy,
>> >
>> > What I did, was to modify the routing in the routing module and also
>> > add a property "IsStaticFile" on the RouteMatch, so that the routing,
>> > which is responsible for rewriting the url, checks for static files,
>> > rather than letting MRHttpHandlerFactory do it. What MRHHF does is to
>> > check if the route is for static.
>> >
>> > I also made the routing aware of the controller tree, because in my
>> > opinion there's no use in routing to mr handlers if there's no
>> > controller/action for the route anyway; then we might fail with
>> > ASP.Net's error handling (+ being able to set 404 pages through web
>> > admin API in IIS7, or error handlers in web.config in iis6) like
>> > default and save ourselves the trouble of mapping every single file
>> extension manually in Web.config.
>> >
>> > Regards,
>> > Henrik
>> >
>> > -----Original Message-----
>> > From: [email protected]
>> > [mailto:[email protected]] On Behalf Of Jimmy
>> > Shimizu
>> > Sent: den 20 maj 2009 12:22
>> > To: [email protected]
>> > Subject: Re: robots.txt and sitemaps with routing
>> >
>> >
>> > I'm using RoutingModuleEx, basic MonoRail routing that is mapped like
>> this:
>> >
>> > <add name="MonoRail" path="*" verb="*"
>> > type="Castle.MonoRail.Framework.MonoRailHttpHandlerFactory,
>> > Castle.MonoRail.Framework" preCondition="integratedMode"/>
>> >
>> > I was under the impression that with this approach, static files are
>> > never handled correctly. If that is not the case, I'd love to hear
>> > about the solution :)
>> >
>> > I get Url smaller than 2 tokens if I try to request a static file.
>> >
>> > Jokin Cuadrado wrote:
>> >
>> >> What routing are you using?
>> >>
>> >> the one i use check if a file exist before handling the dynamic url,
>> >> so just put a static robot.txt in the root and it will work.
>> >>
>> >>
>> >> On Mon, May 18, 2009 at 12:01 PM, Jimmy Shimizu
>> >> <[email protected]>
>> >>
>> > wrote:
>> >
>> >>
>> >>
>> >>> How do you guys solve the issue with files that are supposed to be
>> >>> found directly under the siteroot when using advanced routing
>> >>> (meaning, catching * with Monorail)?
>> >>>
>> >>> I was planning on using a specific controller that servers
>> >>> robots.txt and sitemaps dynamically, but when someone needs to
>> >>> verify for example domain-ownership with analytics or webmastertools
>> >>> or such, how do you handle that?
>> >>>
>> >>>
>> >>>
>> >>
>> >>
>> >>
>> >
>> >
>> >
>> > >
>> >
>>
>>
>>
>>
>>
>
>
> --
> Ken Egozi.
> http://www.kenegozi.com/blog
> http://www.delver.com
> http://www.musicglue.com
> http://www.castleproject.org
> http://www.gotfriends.co.il
>
>
>
>
>
> >
>


-- 
Ken Egozi.
http://www.kenegozi.com/blog
http://www.delver.com
http://www.musicglue.com
http://www.castleproject.org
http://www.gotfriends.co.il

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Castle Project Users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/castle-project-users?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to