AWS looks best for sites that get a lot of traffic. A memcached system sounds fast. Still I’d rather spend my time on dev vs systems engineering – until I need to…

Ah yeah. That’s definitely something that lingered from the MVC 2.0 project. New projects in MVC 3.0 don’t have Default.aspx files anymore. I don’t believe any of my current projects [running on Mono] do either. Something I’ve done while converting older projects is create a brand new sandbox project with the new version and compare configs, file layout, etc. I’ve also just created new projects and then imported all the important stuff from the old project; just to make sure it’s correct for the new version. Sometimes those lingering settings for the old version will really hang up stuff.

The AWS thing is pretty fun to play with. I treat AWS as the ad-hoc, full control, detail oriented type solution for folks who want that kind of control. Azure, while more expensive is a great solution if you want to spin up a distributed solution fast and don’t want the headache of configuring it all yourself. I haven’t tried any others yet. I’ve heard good things about AppHarbor, which I think runs on AWS right now.

My thought for a solution I’m playing with is to setup on AWS and wire up a simple MemCached handler for session/cache, MongoDB for data persistence, but it’s all going to require some work to get rolling. I envision some kind of auto-scaling config that will allow me to spin up many instances without interacting with the systems… But this is me just talking about something I haven’t death with yet.

Now I’m contemplating following your EC2 post. Looks pretty involved though. I think Heroku is easier and still free up to 750 hrs/mo.

I don’t believe I even set a DocumentIndex file. Make sure you’re using a config section. This takes the “routes” from the requested URL and hands them off to the Mono handler. From there, the default MVC route handlers will take over. Here’s a section of my config (I used the Mono-project mod_mono config tool to do mine, which may require a few minor tweaks if you’re using a custom mono path).



Allow from all
Order allow,deny
MonoSetServerAlias mvc3
SetHandler mono
SetOutputFilter DEFLATE
SetEnvIfNoCase Request_URI "\.(?:gif|jpe?g|png)$" no-gzip dont-vary


My default routes work with this. Make sure you have a default route specified in your global.asax.cs file.

I finally got back to this, after converting everything to razor and some other stuff. And its sort of working: virtual dirs have path issues, but it is working as the root url except for the actual root url. That is, all urls with paths work, but the root url doesn’t. I tried setting the DirectoryIndex but it doesn’t help. It works on IIS so I shouldn’t have to mess with my routes. Should be an apache config setting but I don’t know what….

Any ideas?

Sorry for the delay on this. This script will do a few things. First: delete /opt/mono-2.10 (default prefix), which is where all the mono files will be installed to. Second, you can “reverse” the apt-get install by doing: “apt-get remove “. (That’s kind of a guess, but if you’re starting with a fresh machine and you’re wanting to roll back all the dependency installed added by the script). You may have to follow that up with “apt-get autoremove” to clean up.

That should be it. I think the major piece is the mono install itself which is super easy to clean up since the script prefixes that to a custom directory.

Hope that helps! ]]> http://www.integratedwebsystems.com/2011/08/install-mono-2-10-3-on-ubuntu-using-bash-script/comment-page-1/#comment-10137 Oscar Sat, 14 Jan 2012 08:24:12 +0000 http://www.integratedwebsystems.com/?p=1207#comment-10137 Hello, How do you 'uninstall' the changes made by the script ? TIA, Oscar Hello,

How do you ‘uninstall’ the changes made by the script ?

TIA, Oscar