Net Data Design, LLC Blog

Windows 7 and Windows Server 2008 R2 ship with IIS 7.5. It’s called Application Pool Identities. Application Pool Identities allows you to run Application Pools under an unique account without having to create and manage domain or local accounts. The name of the Application Pool account corresponds to the name of the Application Pool. The image below shows an IIS worker process (w3wp.exe) running as the DefaultAppPool identity.

Task Manager - Worker Process

Application Pool Identity Accounts

Worker processes in IIS 6 and 7 run as NETWORKSERVICE by default. NETWORKSERVICE is a built-in Windows identity. It doesn’t require a password and it has only user privileges, i.e. it is relatively low-privileged. Running as a low-privileged account is a good security practice because then a software bug can’t be used by a malicious user to take over the whole system.

The problem is however that over time more and more Windows system services started to run as NETWORKSERVICE and services running as NETWORKSERVICE can tamper with other services running under the same identity. Because IIS worker processes run third-party code by default (Classic ASP, ASP.NET, PHP code) it was time to isolate IIS worker processes from other Windows system services and run IIS worker processes under unique identities. The Windows operating system provides a feature called “Virtual Accounts” that allows IIS to create unique identities for each of its Application Pools. Click here for more information about Virtual Accounts.

Configuring IIS Application Pool Identities

If you are running IIS 7.5 on Windows Server 2008 R2 you don’t have to do anything. For every Application Pool you create the IIS Admin Process (WAS) will create a virtual account with the name of the new Application Pool and run the Application Pool’s worker processes under this account.

If you are running Windows Server 2008 you have to change the IdentityType property of the Application Pool you created to “AppPoolIdentity”. Here is how:

• Open the IIS Management Console (INETMGR.MSC).
• Open the Application Pools node underneath the machine node. Select the Application Pool you want to change to run under an automatically generated Application Pool Identity.
• Right click the Application Pool and select “Advanced Settings…”

Configuring AppPool Identity

• Select the “Identity” list item and click the button with the three dots.
• The following dialog appears.

Selecting AppPool Identity

• Select the Identity Type “ApplicationPoolIdentity” from the combo box

To do the same step via command-line you can simply call the appcmd command-line tool the following way:

Securing Resources

Whenever a new Application Pool is created the IIS management process creates a security identifier (SID) representing the name of the Application Pool itself, i.e. if you create an Application Pool with the name “MyNewAppPool” a security identifier with the name “MyNewAppPool” is created in the Windows Security system. From this point on resources can be secured using this identity. The identity is not a real user account however, i.e. it will not show up as a user in the Windows User Management Console.

You can try this by selecting a file in Windows Explorer and adding the “DefaultAppPool” identity to its Access Control List (ACL).

1. Open Windows Explorer
2. Select a file or directory.
3. Right click the file and select “Properties”
4. Select the “Security” tab
5. Click the “Edit” and then “Add” button
6. Click the “Locations” button and make sure you select your machine.
7. Enter “IIS AppPool\DefaultAppPool” in the “Enter the object names to select:” text box.
8. Click the “Check Names” button and click “OK”.

By doing this the file or directory you selected will now also allow the “DefaultAppPool” identity access.

Securing Resources for your Process Identity

You can do this via the command-line using the ICACLS tool. The following example gives full access to the DefaultAppPool identity.

On Windows 7 and Windows Server 2008 R2 the default is to run Application Pools as this security identifier, i.e. as the Application Pool Identity. To make this happen a new identity type with the name “AppPoolIdentity” was introduced. If the “AppPoolIdentity” identity type is selected (default on Windows 7 and Windows Server 2008 R2) IIS will run worker processes as the Application Pool identity. With every other identity type the security identifier will only be injected into the access token of the process. If the identifier is injected content can still be ACLed for the AppPool identity but the owner of the token is probably not unique. Here  is an article that explains this concept.

Accessing the Network

Using the NETWORKSERVICE account in a domain environment has a great benefit. Worker process running as NETWORKSERVICE access the network as the machine account. Machine accounts are generated when a machine is joined to a domain. They look like this:

<domainname>\<machinename>$,

for example:

mydomain\machine1$

The nice thing about this is that network resources like file shares or SQL Server databases can be ACLed to allow access for this machine account.

What about AppPool identities?

The good news is that Application Pool identities also use the machine account to access network resources. No changes are required.

Note:

This article http://learn.iis.net/page.aspx/246/using-fastcgi-to-host-php-applications-on-iis-70/ walks you through the process. This article was written for IIS 7, but will apply to IIS 7.5 as well.

In addition, if your server is 64 bit, you should set ‘enable 32 bit applications‘ to true in your application pools settings.

We, as an SMB (small and medium business), have been using Amazon’s S3 for sometime now. We are a small software development company that specializes in web applications. We also host web applications and sites. The benefit of S3 is it gives us the ability to keep complete off-site back-ups as well as share certain files with outside contractors and clients.

This probably isn’t anything earth shattering to most, but it certainly beats our old system of taking back-up tapes and DVD off-site every night. That method had too many short-comings to say the least.

Recently, for one of our secure document storage web applications, we started incorporating a data provider that utilizes the S3 API. Setting a search provider for S3 proved to be a bit of a challenge, but so far it has worked out fairly well.

So lets get back to the topic at hand. For the SMB looking for a cost effective way to keep file storage organized and highly available, I’m hard pressed to come up with another viable solution. If you’re the owner of a SMB or work for one, what options have you explored? What method of [off-site] back-up are you using? I’d really like to hear from you…

A serious SQL injection attack has injected a malicious iframe on more than 100,000 susceptible websites. ScanSafe reports that the injected iframe loads malicious content from 318x.com, which eventually leads to the installation of a rootkit-enabled variant of the Buzus backdoor trojan. A Google search on the iframe resulted in over 132,000 hits as of December 10, 2009.

Keep in mind that 99.9% of these attacks stem from poor coding. It very important to implore beginners (and some more experienced programmers) to code against them. In brief:

• Constrain Data, Check for known good data by validating for type, length, format, and range.
• Use type-safe SQL parameters for data access. Use parameters with stored procedures or dynamically constructed SQL command strings.
• Use a low-permission database account for data access.
• Hide data errors, don’t give clues at to what maybe acceptable to the database.

Now, there is a reason the items above look simple. They Are!

All we need are programmers that will listen and not take short cuts.

2.8.6 fixes two security problems that can be exploited by registered, logged in users who have posting privileges.  If you have untrusted authors on your blog, upgrading to 2.8.6 is recommended.

The first problem is an XSS vulnerability in Press This discovered by Benjamin Flesch.  The second problem, discovered by Dawid Golunski, is an issue with sanitizing uploaded file names that can be exploited in certain Apache configurations. Thanks to Benjamin and Dawid for finding and reporting these.