Mai-lan's Blog

TechNet/MSDN IIS7 front-end servers running 100% virtualized using Hyper-V

mailant — Thu, 29 May 2008 06:46:00 GMT

We get a lot of questions about best practices for Hyper-V and IIS. The answer is pretty simple so far...it just seems to work, aside from the normal overhead for virtualization on a workload. Customers are already starting to or planning to deploy IIS on Hyper-V. MS IT is the first in line and has just released a report describing their experience with front-end IIS servers on a couple of big MS web properties.

A couple of months ago, Microsoft IT migrated all of their IIS7 front-end servers for TechNet and MSDN web sites to run virtualized on Hyper-V RC0 with 4 virtual CPUs and 10GB RAM. The hardware used for the virtualized IIS servers are 2 Intel quad-core CPUs and 32GB RAM (2GB are reserved for the Windows Server 2008 parent partition). Backend databases are still on physical servers at this point.

These are big sites -- TechNet recieves 1 million hits a day and MSDN gets 3 million per day. After a couple of months bake time, the results of the migration were officially released yesterday here in some detail. These VMs are using Windows Server 2008 Enterprise RTM with the standard install option. The general consensus from the experience by our IT folks? It just worked...

DiscountASP.NET IIS Manager UI extensions

mailant — Fri, 02 May 2008 05:42:00 GMT

One of the hosters who we work with, DiscountASP.NET (an ASP.NET focused web hoster based in sunny Southern California), has just launched a beta of two new UI extensions for the IIS Manager for their customers.

One of them is a web.config backup/restore module, which will backup your web.config file and restore it. I think that this is a very useful function...I do it all the time using appcmd when I am messing around with config. I can see how DiscountASP.NET customers would like to see it integrated in the UI. The second module is a feedback module where customers of DiscountASP.NET can send feedback and suggestions.

Pretty cool! I love seeing partners take advantage of the IIS extensibilty model, which, among other things, makes any extensions like these delegate-able to remote users as well.

Managing CPU utilization for IIS worker processes that use FastCGI module

mailant — Tue, 29 Apr 2008 04:42:00 GMT

Since there have been several questions in the IIS FastCGI forums regarding using IIS CPU Limits feature together with FastCGI, I wanted to talk a little more about how what you should do if you are trying to control CPU while running PHP on IIS7.

First off, It is not possible to use FastCGI when CPU limits feature is turned on. This is because CPU limits feature uses Job Object to manage CPU utilization by worker process, and FastCGI also uses Job Object to ensure that child FastCGI processes are killed when parent worker process is terminated. Since a process can belong to only one Job Object, FastCGI module returns the 500 error if it determines that Job Object is already being used. This behavior is by design.

Even though you can't use the IIS CPU limits feature, you can still control CPU utilization by IIS worker processes. The recommended approach is to use Windows Server Resource Manager (WSRM). WSRM is part of the Windows Server 2008 and it can be used to manage server processor and memory usage with standard or custom resource policies. It is pretty nifty. Refer to this article to get instructions on how to install and configure WSRM for IIS. Note that WSRM documentation article states that WSRM does now work with processes that use Job Objects. This statement is not exactly correct - WSRM cannot manage the processes within Job Object, only if Job Object sets limits that conflict with WSRM (such as memory utilization, process priority, etc). Since Job Object in FastCGI handler does not set any of those limits, WSRM is able to manage FastCGI processes successfully. The WSRM team is going to revise the documentation text to point that out. We wanted to give you a heads up on that detail now so that you could take advantage of WSRM today if you needed it for hosting ASP.NET, PHP or other Fast-CGI compliant applications on IIS7.

IIS7 Diagnostics for Developers talk at MIX08

mailant — Fri, 07 Mar 2008 08:26:00 GMT

I just got back from MIX08 in Vegas where I gave a talk called “Everything You Need To Know About Debugging and Diagnostics with IIS7.” It was a fun talk. There were a fewIT administrators and a whole lot of developers. When we have talked about this topic with customers in the past, we have focused on the IT administrator perspective. This time, we took it from a different angle, which is to think through when and how you would leverage the diagnostic platform as a developer.

At the end of the day, both developers and IT professionals use most of the same tools (with a few differences to be called out in this post). The primary difference is the environment and the dev lifecycle in which the tool is used.

Advice From the Front Lines (Or, What I Learned From Brad, our IIS Escalation Engineer)

Before I went to MIX, I talked to Bret, the IIS Escalation Engineer who works closely with our product team, asking for his input on how he diagnoses problems with IIS apps. I figured that he has a great perspective on this, since he has seen all sorts of problems in all sorts of configurations. Bret provided a list of diagnostic patterns that I included in my presentation at MIX, namely:

Don’t assume something is happening. Do get facts. (Using features like performance counters, event counters, FREB logs, RSCA)
Don’t go down the “switch this setting and see”. Do find the root cause, and then resolve the problem there. (Again, FREB, detailed errors, RSCA)
Do isolate the problem and get a good problem statement. (ditto on tools)

When you think about it, these three concepts are both critical to effective troubleshooting and amazingly hard sometimes when you are in the thick of figuring out why your application just won’t work the way you want it. It all comes down to figuring out the root cause of your problem. Bret talked about one case where someone he worked with once spent 2 hours trying to troubleshoot a server unavailable error by getting netmon traces and performing a whole host of other diagnostics. The root cause was that the w3svc was not started.

Once you have determined the root cause of the problem, you can isolate it even further to figure out what you need to fix. Bret gave another example where if a customer is having problems with Windows Auth, try using NTLM only. Does that work? If so, then you know there is a problem with Negotiate and Kerberos config.

That’s what the IIS7 platform diagnostic tools are all about – narrowing down to root cause and isolating the problem. Our diagnostic are built into IIS and integrate with ASP.NET Trace events and System.Diagnostics.

2 Very Useful Diagnostic Tools in the Dev LifeCycle
Detailed errors and Failed Request Tracing are two features that apply all the time. Detailed Errors you just “get” for free with the Web server. Failed Request Tracing requires a tiny amount of work to use – namely, setting up rules and looking at logs.

When you are developing a web app, you are constantly checking to see if it works in the browser. Me, I am prone to configuration errors, especially if I get myself in trouble messing around the applicationhost.config or web.config file directly – I stay in safer waters when I use the IIS Manager or appcmd to modify configuration. (AppCmd backup has become a reflexive twitch for me before I start to fiddle with an app.) I tend to see a lot of 500.19 errors with wrong configuration syntax. Detailed Errors are smart enough to tell me the exact plan in the configuration file that is not working.

Failed Request Tracing is incredibly useful to pinpoint what is failing where in the request pipeline. You can set up rules that trace a range of errors (like 400-500). Combined, the two tools are pretty handy for the quick “gut check” when you are developing your code. Since ASP.NET Trace events and System.Diagnostics data are also integrated in the Failed Request Tracing logs, you can use the single log file to view the data that you get from HTTP errors, ASP.NET Traces, and System.Diagnostic calls.

When you move to functional testing, you can add on a couple more layers of diagnostics, namely WCAT, orphaning processes, and some interesting performance counters

Functional Testing with Web Capacity Analysis (WCAT)

Web Capacity Analysis Tool (WCAT) for stress testing. It’s amazing how few people know about WCAT. This is the tool that we use internally for our own performance testing. It lets you throw all kinds of load against your dev box (or production server) to simulate high load. And it’s free. You can get it from the IIS6 Resource Kit or download it from www.iis.net Download Center.

Orphaning Failed Processes For Debugging

You might want to orphan failed processes, which basically means making sure the server doesn’t kill off a worker process that hangs. The Web server will automatically recycle a failed process by ending it and starting a replacement. If you enable the process orphaning, the Web server leaves the failed worker process running and starts a new process up in its place.

You can also configure the orphaned process so that the server can run a command, like launching a debugging tool, on the orphaned process. (The Executable attribute is how you could attach a command to the orphaned process in the UI and orphanActionExe is the config setting.)

Monitoring Performance Counters

Performance counters are handy for functional testing. Mukhtar, who does our performance testing for IIS, looks at different data for managed and unmanaged requests.

For unmanaged requests, Mukhtar looks at "Processor(_Total)\\% Processor Time". If this counter falls well short of 100%, Mukhtar starts digging. He also looks at MBits/sec in the wcat log summary section to see if the network was maxed. Another counter he checks out is "System\\Context Switches/sec" which is basically Context Switches/Request in the wcat log summary section, to learn if the problem is scaling. And of course, the "Memory\\Available Mbytes” is important because it tells you if you ran out of RAM.

When working with managed requests, Mukhtar pays close attention to all counters GC-related to see if there are any unusual spikes.

In addition to these counters, there are also IIS-specific counters that are useful to monitor. I would check out the IIS6 perf counters to watch (http://www.microsoft.com/technet/prodtechnol/WindowsServer2003/Library/IIS/71490aae-f444-443c-8b2a-520c2961408e.mspx?mfr=true), which are still valid, as well as some of the newer ones in IIS7 that I talked about back in January: http://blogs.iis.net/mailant/archive/2008/01/10/new-worker-process-performance-counters-in-iis7.aspx.

Other Tools in Production

Once you are running your code in production, all the same diagnostic tools apply plus additional tools like Runtime State and Controls API (RSCA). RSCA lets you see requests that are in-flight and taking more than 0 seconds to execute.

You can view RSCA data through the IIS7 Manager or using appcmd (“appcmd list requests”). Unless you are using WCAT for stress testing (or demoing, in my case), you typically don’t use RSCA when you are developing and in functional test mode because you aren’t generating enough load to see where a request is getting bogged down in the pipeline.

Another tool that you can use in production is Event Tracing for Windows (ETW). You can use our LogParser (also in the IIS6 resource pack) to parse the verbose logs that you get with ETW. Bret our IIS escalation engineer recommends using ETW for complicated issues with a specific request. He also recommends DebugDiag, a tool in the IIS6 resource pack that lets you find memory leaks, process terminations, and slow responses.

Best Practices

For the developing phase (which includes functional testing), I called out:

Use failed request log files to find error patterns and offending URLS, along with detailed errors
Consider using process orphaning
Use ASP.NET health monitoring to receive configurable alerts about errors
Use WCAT to stress application before product

When you are tuning, you should consider:

Use design patterns in the Performance Tuning whitepaper (here’s the link that I forgot to add to my deck and promised the audience that I would put in this blog: http://download.microsoft.com/download/9/c/5/9c5b2167-8017-4bae-9fde-d599bac8184a/Perf-tun-srv.docx). This isn’t just about tuning IIS, though, it includes all of Windows Server 2008.
Enable Output Caching for semi-dynamic pages. In my demo, we saw an enormous difference in performance by moving all content with a .php extension to kernel mode caching.
Set 32 bit IIS worker processes in Wow64 mode in per-AppPool settings (there was a typo in the deck that didn’t include the very important 32 bit part, sorry)
If you * script-mapped all requests to ASP.NET in IIS6, Integrated Pipeline is much faster than an IIS6 * scriptmap solution. Try together with IIS7 URL Authorization.
Move your highly trafficked default document to top of the Default Documents list so that the Web server doesn’t always go down the list to find it.

While doing production-level troubleshooting, it is good to:

Use Failed Request Tracing to capture hard-to-repro errors. Keep the rules on in production to make sure that you’re able to react quickly.
Set fine-grained Failed Request Tracing rules to keep your log history valid. By default, the limit to Failed Request Tracing logs is 50. After that, they start to recycle. You can expand that limit to a higher number if you would like. Best practice, though, is to build the right rule set.
Check key performance counters for clues on application health, which I mentioned earlier.

Extensibility, PHP and Diagnostics

One of the things that I talked about today is the new modular platform for IIS7. You can replace one of your modules with your own, or extend the platform by building in new functionality. That’s what we are doing with our IIS extensions like secure FTP Server for Windows Server 2008. When you build on the new IIS7 platform, you get the benefit of other features in the platform like remote delegated administration (!) and the diagnostics functionality.

If you write your own module that enables virtual directory hosting, you can use Failed Request Tracing and all of the other diagnostics that I talk about in this blog with it. For FastCGI-compliant application frameworks like PHP and Ruby, there are a couple of diagnostics idiosyncrasies to note:

In Detailed Errors you would see potential errors from the FastCGI module and not a particular application.

Your PHP application will not generate a Failed Request Tracing event unless the event occurs before entering the PHP engine and after leaving the PHP engine. Same thing with ETW. Since the PHP engine doesn’t emit Failed Request Tracing events or ETW events, the app will be pretty much silent while being processed by the PHP engine itself.

Questions from the Audience

The Q&A time after the session was a lot of fun. I know that I am not capturing all of questions here, sorry about that -- in particular, I can’t remember all the questions that came up during the presentation itself. But here is a list of some of the questions that I caught at the end of the talk:

How do I migrate my old IIS6 applications to IIS7? Try using the CTP for Web Deploy, downloadable here. Make sure you grab the walkthroughs (under the Documentation link on the download page) – the walkthroughs are basically discrete labs on common tasks with Web Deploy.
Are you doing anything with PowerShell? Yes, we are working on Technical Preview for a PowerShell provider for our configuration system.
Can you use kernel model caching on content that requires authorization? No, if the content requires authorization, you need to use user mode caching, which can be configured on the same page as kernel mode caching.
I have a lot of old ADSI scripts that create web sites in my hosted environment. When the scripts create the new sites, are they IIS7 sites or IIS6 sites since it’s using a legacy technology? If the script is creating a new site, it should be an IIS7 site. To get the scripts to run, though, you will need to install the IIS legacy components which are an optional install in setup.
Why does IIS always bring along the legacy components? The IIS web server doesn’t require the legacy components for its core functionality. Other functionality will bring it along, though. For example, if you are using SMTP with your IIS server, you will need to install the legacy components for IIS so that SMTP with work (it needs the metabase).
Is there any migration between the metabase on the new XML configuration system? No, we do not have a conversion tool that transforms the metabase into the right config files. We will focus our efforts in a great Web application and server migration story that takes the config settings from one store to the other, rather than trying to convert the storage itself.
What is the relationship between the error dialogs that I pop in my app and Detailed Errors? There is actually no correlation between the app error dialogs and Detailed Errors. Detailed Errors are all about HTTP error codes and mapping the error code to probably cause. Your error dialogs are likely to application logic-specific.
Can I view Detailed Errors remotely? No, Detailed errors can only be viewed on the local server.
Can II6 applications work on IIS7? Yes. ISAPI apps will run in Classic Mode without modification.
If you are a developer and you get a feature delegated remotely to you, do you have a limited view scoped down to just what you are allowed to do?

Thanks For Listening!

If there are any of the MIX08 IIS diagnostics talk attendees out there, thanks for coming and listening! You were a great crowd. I will try to get the video posted once that is out.

Introducing the new IIS.NET web site

mailant — Wed, 27 Feb 2008 16:26:00 GMT

So...notice anything different today? :-) Thanks to the efforts of a whole team of people here in Redmond, Telligent, Goldman Design and ORCS Web, we have significantly overhauled our www.IIS.net web site. As you can tell from the depth of the content and the participation of the product group in forums and blogs, fostering a strong and vibrant technical community is a huge deal for us. We care about you in particular, and how you learn, use, and extend IIS in all the ways that you do. To help with that process, we've rolled out some infrastructure and tools across www.iis.net that helps us get information out to you and for you to get information back to us.

Here's a quick recap of some of the changes:

Updated look and feel that maximizes content, which is what the web site is all about.
Dynamic home page content. It will change every day, driven by the latest forum posts, blogs, and news.
Personalized blog photos, so you can see who you are talking to through the site. That's kind of fun.
http://learn.iis.net. This is the center of the technical content universe on IIS.NET. This is where you go for overviews and deep dives on content ranging from IIS7 features to how to host IIS in shared hosting environment.
Hosting guidance that covers key technologies across the Microsoft stack for shared hosting providers. These guidelines are all about best practices and what you need to do differently in a multi-tenent environment...critical information about Windows Server, IIS, ASP.NET and other products that hosters use to provide services to customers.
Under the hood, we've replaced our content publishing system with a wiki. We will be able to open up the wiki to contributors later on, once we put some more scaffolding in place around content reviews and approvals.
Video support. We've got 10 videos in learn.iis.net now, and we are going to be adding a lot more. We have a deep committment to media integration as a first class content type and will be lining up a series of new features including podcasts of audio and video content around IIS.
Ad design. Ad revenue helps us fund great content. We have made some changes to increase our revenue potential in a way that does not distract from our goal of building a vibrant technical community.

It being a web site, we'll continue to tweak it along the way, naturally. For example, we're going to be adding a Quick Reference, which will be a new and innovative way to learn about a single topic through text description, video overviews, code samples, and schema descriptions. Enjoy the changes and let us know what you think at iisnet@microsoft.com.

Thanks,

Mai-lan

IIS Migration and Synchronization tool, Web Deploy (Technical Preview 1)

mailant — Mon, 28 Jan 2008 07:12:00 GMT

We recently released IIS7's Web Deploy, which gives you the ability to migrate and synchronize IIS6 and IIS7 applications(download the x86 bits here and the x64 bits here).

Take a minute and think about the tasks that you do right now to move a web application from one server to another. There's all sorts of backing up, restoring, copying, praying, perhaps some cursing, and then validating on the machine. Consider, for example, the need to back up a web application before a migration, a change to the site, or by schedule. You can use the archive functionality in WebDeploy to accomplish this with a few simple lines of script.

Taking an archive or snapshot of a web site to restore or synchronize it later requires the following command on or against an IIS6.0 site:

msdeploy.exe -verb:sync -source:metakey=lm/w3svc/1 -dest:archivedir=c:\archive > wdeployarchive.log

(If you want to do the same action on an IIS 7.0 site, just replace metakey=lm/w3svc/1 with appHostConfig=”Default Web Site”.)

The > wdeployarchive.log argument makes sure that the results of the archive will be listed in the log file. This archive now works as a backup for the site, to sync to an IIS 6.0 server or to migrate to an IIS 7.0 server.

You can create an archive with most providers, excluding directory and file path providers. You can also re-sync the source provider, like the metakey, with the same archive or create new archives to achieve versioning. One note for the Technical Preview 1: you cannot add to an archive once it has been created.

Now let's say that you've (or someone has) trashed the web site on your server. You can restore the web site from your archive with this command for the IIS 6.0 site:

msdeploy.exe -verb:sync -source:archivedir=c:\archive -dest: metakey=lm/w3svc/1 > wdeployarchive.log

(Likewise,if you want to do the same action on an IIS 7.0 site, just replace metakey=lm/w3svc/1 with appHostConfig=”Default Web Site”.)

That means two lines of script for the archive and restoral of the Web site, with basic logging. If you want to learn more about how to use Web Deploy, check out these walk-throughs.

We are really excited about Web Deploy here in Redmond -- we're looking for your feedback, so try it out and let us know your thoughts in our forums on the Web Deployment team blog.

TechNet Radio Broadcasts for "Managing PHP on Windows" and "Deploying and Managing a Customized Web Server"

mailant — Mon, 14 Jan 2008 04:51:00 GMT

Right before the holidays, we did some radio broadcasts on our team targeting the IT Professional and focusing on specific topics of interest in IIS7.

Isaac Roybal, product manager, interviewed me on what it means to deploy and manage a customized web server. I talked about how IIS's componentized architecture gives you fine-grained control over your server footprint and functionality, including how to install and manage on Server Core. Here's a link to that interview:

http://www.microsoft.com/technet/community/tnradio/archive/jan022008.mspx

Then I sat down with Ruslan Yakushev, Senior Program Manager for FastCGI, to talk about managing PHP on Windows. I thought Ruslan did a great job walking through the highlights and best practices for managing PHP using FastCGI on IIS6 and IIS7. Here's a link to that interview:

http://www.microsoft.com/technet/community/tnradio/archive/jan082008.mspx

We'll have more of these radio broadcasts coming out over the next couple of months. You can put a voice to the name of some of the people that bring you the product and IIS.NET

Integrated Pipeline and Hostable Web Core: Two good things to know if you develop on top of IIS...

mailant — Sat, 12 Jan 2008 23:36:00 GMT

IIS7 is a big release for developers. For example, we've changed our extensibility model in this version to make it easier for customers and vendors to build web applications and extend the Web server. I think two of the most interesting innovations in this release are the integrated pipeline and hostable Web core. Mike Volodorsky wrote a great article on one of the reasons why the integrated pipeline is so cool in this month's MSDN magazine. Check it out if you can at: http://msdn.microsoft.com/msdnmag/issues/08/01/PHPandIIS7/default.aspx. Hostable Web core (HWC) is a feature that we haven't talked enough about but will start adding more information to iis.net (stay tuned). HWC opens up a lot of interesting scenarios if you are looking to build solutions or tools that hosts IIS, like a control panel. Kanwal did an excellent job of introducing Hostable Web core to developers in this blog: http://blogs.iis.net/ksingla/archive/2007/12/20/ins-amp-outs-of-hostable-web-core.aspx.

New worker process performance counters in IIS7

mailant — Thu, 10 Jan 2008 06:38:00 GMT

In IIS7 RC0, one of the neat new features we added were two performance counters that give you a window into what’s happening on a per worker processes and the application level at runtime. This is useful because it lets you have a finer-grained monitoring view on servers that are running a large load. You can check out the overall health of specific processes and apps on a well-loaded server with these two new counters, which let you monitor and track over time per worker process data, like when/how requests are queuing or if the output cache truly is working. All of our worker process counter instances are named <pid>_<apppool_name>. If you wanted to search in perfmon for a counter, you could use the *_AppPoolName to get all WP instances running that app pool.

WAS_W3WP

This counter set exposes Web Admin Service related counters for the worker process. Use this counter if you want to check in on a specific worker process. This performance counter tracks the following info:

Total Health pings: Total number of health pings received by the process.
Total Runtime Status Queries: Total number of Runtime Status queries received by the process.
Health Ping Reply Latency: Time, in 100-nanosecond intervals, taken by worker process to reply to last health ping.
Active listener channels: Number of currently active listener channels in the worker process.
Active protocol handlers: Number of currently active protocol handlers in the worker process
Total WAS Messages Received: Total number of messages received by the worker process from Web Admin Service.
Messages Sent to WAS: Total count of messages sent to WAS
Total Requests Served: Total number of requests served by the worker process. This counter is only meaningful when request based recycling is enabled for the application pool.
Total Messages Sent to WAS: Total number of messages sent to Web Admin Service by the worker process.

W3SVC_W3WP

This counter set exposes HTTP request processing related counters for the worker process. You will want to use this to check in on how the web server WAS object is handling requests to do things like make sure that the output caching is behaving as expected.

Total HTTP Requests Served: Total number of HTTP requests served by the worker process.
Requests / Sec: HTTP requests/sec being processed by the worker process.
Active Requests: Current number of requests being processed by the worker process.
Total Threads: Total number of threads available to process requests in the worker process.
Active Threads Count: Number of threads actively processing requests in the worker process.
Maximum Threads Count: Maximum number of threads to which the thread pool can grow as needed.
Current File Cache Memory Usage: Current number of bytes used by user-mode file cache.
Maximum File Cache Memory Usage: Maximum number of bytes used by user-mode file cache.
Output Cache Current Memory Usage: Current number of bytes used by output cache.
Current Files Cached: Current number of files whose contents are present in user-mode cache.
Total Files Cached: Total number of files whose contents were ever added to the user-mode cache (since service startup).
File Cache Hits: Total number of successful lookups in the user-mode file cache (since service startup).
File Cache Misses: Total number of unsuccessful lookups in the user-mode file cache (since service startup).
File Cache Flushes: Total number of files removed from the user-mode cache (since service startup).
Active Flushed Entries: Number of file handles cached in user-mode that will be closed when all current transfers complete.
Total Flushed Files: Total number of file handles that have been removed from the user-mode cache (since service startup).
Current URIs Cached: URI information blocks currently in the user-mode cache.
Total URIs Cached: Total number of URI information blocks added to the user-mode cache (since service startup).
URI Cache Hits: Total number of successful lookups in the user-mode URI cache (since service startup).
URI Cache Misses: Total number of unsuccessful lookups in the user-mode URI cache (since service startup).
URI Cache Flushes: Total number of URI cache flushes (since service startup).
Total Flushed URIs: The number of URI information blocks that have been removed from the user-mode cache (since service startup).
Current Metadata Cached: Number of metadata information blocks currently present in user-mode cache.
Total Metadata Cached: Total number of metadata information blocks added to the user-mode cache (since service startup).
Metadata Cache Hits: Total number of successful lookups in the user-mode metadata cache (since service startup).
Metadata Cache Misses: Total number of unsuccessful lookups in the user-mode metadata cache (since service startup).
Metadata Cache Flushes: Total number of user-mode metadata cache flushes (since service startup).
Total Flushed Metadata: Total number of metadata information blocks removed from the user-mode cache (since service startup).
Output Cache Current Items: Number of items current present in output cache.
Output Cache Total Hits: Total number of successful lookups in output cache (since service startup).
Output Cache Total Misses: Total number of unsuccessful lookups in output cache (since service startup).
Output Cache Total Flushes: Total number of flushes of output cache (since service startup).
Output Cache Total Flushed Items: Total number of items flushed from output cache (since service startup).
File Cache Hits / sec: Rate of successful lookups in file cache during last sample interval.
Metadata Cache Hits / sec: Rate of successful lookups in metadata cache during last sample interval.
Uri Cache Hits / sec: Rate of successful lookups in URI cache during last sample interval.
File Cache Misses / sec: Rate of unsuccessful lookups in file cache during last sample interval.
Metadata Cache Misses / sec: Rate of unsuccessful lookups in metadata cache during last sample interval.
Output Cache Misses / sec: Rate of unsuccessful lookups in output cache during last sample interval.
Uri Cache Misses / sec: Rate of unsuccessful lookups in URI cache during last sample interval.

Hope these help!

Automating IIS from the Command-Line/Scripting Webcast

mailant — Sat, 15 Dec 2007 06:27:00 GMT

I gave a presentation recently on automating IIS from the command-line (you can check it out at: https://www.livemeeting.com/cc/mseventsbmo/view?id=1032352182&role=attend&pw=D80FDE8D). I focused on how to manage IIS using PowerShell, WMI, and AppCmd. I talked a lot about Server Core as well, and the management options for IIS with Server Core.

Here are the questions from that presentation with answers in case you are interested.

Mika: Hi, I only just joined and might have missed something :( Is the demo script available somewhere?

Mai-lan: I’ve attached a couple of the appcmd batch files that I showed during our presentation.

Salam: Is IIS 7 going to support sFTP, I saw some SSL icons on FTP

Mai-lan: IIS is shipping a new secure FTP server as a downloadable module from IIS.NET at the end of February. If you download the module to your Windows Server 2008 installation, it will integrate into your existing installation, including the management of the server through IIS7 Manager.

salam: 1st Q, how come you were able to create a 2nd web site on the same port and IP where the default one is running without getting an error?

Mai-lan: You can do this by having two different host headers. Host headers provide for an alias so that they can be differentiated and processed by the right pool. You have one IP address per server and thousands of host-header based web-sites. You can copy and paste the following lines into a batch file or just paste them in an elevated command-line:

REM CREATE SITE 1 DIRECTORY
Md %systemdrive%\inetpub\wwwroot\site1

REM CREATE DEFAULT.HTM IN SITE DIRECTORY
echo 'this is site 1' >> %systemdrive%\inetpub\wwwroot\site1\default.htm

REM ADD SITE1 VIA APPCMD
%windir%\system32\inetsrv\appcmd add site -name:site1 -physicalPath:%systemdrive%\inetpub\wwwroot\site1 -bindings:http/*:80:site1

REM ADD HOST HEADER TO HOSTS FILE SO THAT THE NAME 'SITE1" CAN BE RESOLVED BY DNS
echo 127.0.0.1 site1 >> %windir%\system32\drivers\etc\hosts

REM DO THE SAME FOR SITE 2
Md %systemdrive%\inetpub\wwwroot\site2
echo 'this is site 2' >> %systemdrive%\inetpub\wwwroot\site2\default.htm
%windir%\system32\inetsrv\appcmd add site -name:site2 -physicalPath:%systemdrive%\inetpub\wwwroot\site2 -bindings:http/*:80:site2
echo 127.0.0.1 site2 >> %windir%\system32\drivers\etc\hosts

REM REQUEST SITE1 AND SITE 2 VIA IE
start http://site1
start http://site2

Chris: would you please show again how to load the assembly, and get the iis object?

Mai-lan: Sure. Here it is:

# Load assembly explicitly

$mwaPath = join-path $env:systemroot system32\inetsrv\microsoft.web.administration.dll

$mwa = [System.Reflection.Assembly]::LoadFrom($mwaPath)

# create configuration object

$cfg = new-object Microsoft.Web.Administration.ServerManager

salam: Did you create a virtual directory or a new web site? if it is a VD, then I have no problem, but if it was a new web site, then please explain
Mai-lan: This is a good point. If you create a new site using MWA, you have to create a new virtual directory since that is a required parameter. For example, here’s the syntax: Sites.Add("NewSite", "http", "*:8080:", "d:\\MySite"). Here is a good article about using MWA with examples: http://www.iis.net/articles/view.aspx/IIS7/Extending-IIS7/Using-Microsoft-Web-Administration/How-to-Use-Microsoft-Web-Administration?Page=2. I was using AppCmd in the demo. I was creating a site, but I was NOT actually creating a site that would really work because in the demo I didn’t add the physical path and the bindings. When you create a site using AppCmd, you have to use the syntax of appcmd add site /name:string /id:uint /physicalPath:string /bindings:string to really add a site that works. Note that AppCmd will not throw you an error if you create a site without physical path and bindings. You will create a site (like I did in the demo) but it doesn’t have the required information to be fully functional. I’ll demo it with all parameters next time, I don’t want it to be misleading.
tom: Can you describe what are SCRegEdit.wsf, Netsh/Netdom?

Mai-lan: They are Server Core tools for management (not specific to IIS). SCRegEdit.wsf is included in Server Core to more easily enable automatic updates, enable Terminal Server, Remote Admin Mode, enable remote IPSec Monitor management and configure DNS SRV record weight and priority. Netdom is there to let you join a domain. You can learn more at: http://technet2.microsoft.com/windowsserver/longhorn/en/library/bab0f1a1-54aa-4cef-9164-139e8bcc44751033.mspx?mfr=true. There is also a newsgroup on Server Core: http://forums.microsoft.com/TechNet/ShowForum.aspx?ForumID=582&SiteID=17.

Mika: Thanks for a concise summary on IIS admin automation! I'm looking forward on http://blogs.iis.net/mailant

Arthur: not sure if this is the right forum, but related to automated install, from what I have seen IIS will only install on the C drive with IIS 7 and we cannot point the install to another drive, do you know if there is a reason for that?

Mai-lan: INETPUB goes to C drive and there is no way to override that. Sorry...

chris: Excelent Job! Thank You

Mai-lan: No problem, thanks for coming.
chris: When I try to run appcmd on my server the program is not found. Could it that I need a later version of 2008? (Server Longhorn)Mai-lan: AppCmd is located in the %systemroot%\system32\inetsrv\ directory. Because it is not path of the PATH automatically, you need to use the full path to the executable. Alternatively, you canto simplify all the dir switching, you can add the inetsrv directory to the path on your machine so that you can access AppCmd.exe directly from any location.

Health model and KB round-ups

mailant — Sat, 15 Dec 2007 05:25:00 GMT

I have troubleshooting on the mind this week since we just released the IIS7 health model (you can check it out at http://technet2.microsoft.com/windowsserver2008/en/library/b19873a2-9f72-40c8-b73d-89f39cda62781033.mspx?mfr=true). When we talk about diagnostics, we talk a lot about the new diagnostic features like FREB in IIS7 -- Mike has blogged about these in depth at http://blogs.iis.net/mvolo/archive/2007/07/26/troubleshoot-iis7-errors-like-a-pro.aspx. What's neat about the health model is it provides information about every error message that you will encounter in IIS. Take, for example, error code 2269: http://technet2.microsoft.com/windowsserver2008/en/library/b19873a2-9f72-40c8-b73d-89f39cda62781033.mspx?mfr=true. It gives you specific steps to fix and verify the fix for error message "The worker process for app pool '%1', PID='%2', failed to initialize the http.sys communication when asked to start processing http requests and therefore will be considered ill by W3SVC and terminated. The data field contains the error number." Really useful information for free, phone-less troubleshooting. The health model might not have the final solution to every problem, but it is a great first step for troubleshooting any error.

We built the health model in close partnership with PSS and our documentation team. The process of building the health model is an interesting one. Saad Ladki, the Program Manager who shipped the health model, was explaining this to some folks today. Events are manifested to the OS as part of the development process. We pick up events from the manifest and categorize those events into entities. Then we add health state, what we call "resolvers" (how to solve the problem) and "verifiers" (how to verify the that the problem is solved). This information is reviewed by the product team (extensively) and finally posted online as the health model. The most important thing that we focus on in this process is to make sure that our content is as accurate and actionable as possible.

Along the same lines of troubleshooting, here's another good resource for you. Bernard, one of our IIS MVPs, keeps an interesting blog at http://msmvps.com/blogs/bernard/. On his blog, Bernard lists a monthly round-up of IIS Knowledge Base (KB) articles. A useful resource!

Windows Server 2008 RC1 available for download!

mailant — Fri, 07 Dec 2007 07:38:00 GMT

Windows Server 2008 RC1 is available for download! We are feature and quality complete with this release. In this build, we address feedback that we have gotten from customers in beta and our RC0 release. There is nothing quite like deploying IIS7 in a diverse set of production environments worldwide, ranging from huge enterprises and hosters to smaller business environments, to shake out those last issues that need fixing. You can download Windows Server 2008 RC1 at: http://www.microsoft.com/windowsserver2008/audsel.mspx

Here are some of the improvements that we made in this release:

FastCGI fixes, most of which are already described here: http://blogs.iis.net/mailant/archive/2007/11/12/rtm-of-fastcgi-for-iis6.aspx
Support a new default file extension and MIME type for a Silverlight application. This allows Silverlight content to work without manual configuration on IIS. We also have added default MIME type support the .VSTO extention as a content type.
Fix to ensure that editing inherited elements from collection defaults persists the new properties.
Fix to allow serving of classic ASP custom error pages without 500 status errors.
Fix that lets Web Enrollment Certificate Service work after upgrading from Windows Server 2003.
Security improvements, including a fix that prevents running code in one application that accesses the data in another application pool, even if those applications are isolated by IIS.
Stress and performance improvements, including fixes for several memory leaks.
Fixes for several deadlock conditions which hang IIS and prevents pages from being served when an IIS configuration file is used from a UNC share.
ETW support for the GENERAL_RESPONSE_ENTITY_BUFFER event. In previous builds, this event was not handled correctly by ETW which rendered logs containing that event unparsable.
Fix for stale cached data getting written into the config system when IISAdmin shuts down.
Fix for the HTTP request failures that you saw after you upload an invalid web.config file or one of its dependencies to a web site over HTTP. Our development manager Taylor refers to this bug as the equivalent of “locking your keys in the car” when making changes over HTTP – once you uploaded that invalid config file over HTTP, you couldn't edit that site to repair the damage over HTTP.

Windows Media Services 2008 RC1 is also available for download only on Windows Server 2008 RC1; you can learn more about it in Vishal's blog: http://blogs.iis.net/vsood/archive/2007/12/06/windows-media-services-2008-rc1-for-windows-server-2008-rc1-is-now-live.aspx.

On to Windows Server 2008 RTM! The IIS team is really excited to be so close to getting the final product into your hands for use. IIS7 is filled with so many great new features. Stay tuned on upcoming changes that we are making to http://www.iis.net to add more content and web site features that will help you learn more about IIS7 and all the ways that it’s breaking new ground for world-class web hosting.

If you posted a comment on some of my blog posts recently...

mailant — Fri, 07 Dec 2007 06:58:00 GMT

I just realized that I had a bunch of comments stacked up in an unapproved queue. Sorry! I had some spam issues with the blog and switched to moderated comments. I've gone through and pushed the comments through now, and will be answering them. I really do want to hear from you so I will go huddle with Pete, who is our IIS.NET guru and figure out how I can have a spamless blog that lets real comments come through. Sorry for the delay and keep sending the comments and questions my way.

FastCGI Extension for IIS6 – GoLive to RTM release upgrade

mailant — Fri, 07 Dec 2007 06:40:00 GMT

As it has been already mentioned in download instructions for FastCGI Extension RTM for IIS 6, we do not support in-place upgrade of FastCGI Extension from pre-RTM builds to RTM build. If you have been running pre-RTM build (such as GoLive release or Intermediate release), you will need to uninstall it first and then install FastCGI RTM. This post describes some situations to watch out for during upgrade process.

Un-installing FastCGI Extension may cause other handlers to serve requests that were originally mapped to FastCGI.

When you un-install either GoLive or RTM version of FastCGI Extension, it will remove the FastCGI script maps from IIS configuration (you can recreate previous script maps when you re-install by using fcgiconfig.js -syncini command). This may cause situations when some other handler in IIS may start serving requests that were originally served by FastCGI Extension. For example, if you had a “.php” file extension mapped to FastCGI and also had “*” MIME mapping or “.php” MIME mapping then after you uninstall FastCGI Extension the php specific requests may get processed by static file handler. This may cause the source code of PHP script to be sent as a response to a client.To avoid this situation you need to make sure that you do not have MIME type mappings defined for the same extension that is mapped to FastCGI extension. Also make sure that you do not use “*” MIME type mapping.

Running FastCGI RTM installer without prior removal of previous FastCGI version will not upgrade the fcgiext.dll.

If you try to run FastCGI RTM installer on a machine that already has GoLive or Intermediate releases of FastCGI Extension installed, then the RTM installer will complete successfully, but the actual FastCGI extension binary - fcgiext.dll - will not be upgraded. This is an incorrect behavior that is caused by a bug in RTM installer package.·

To avoid this situation – remove the previous version of FastCGI Extension before running FastCGI RTM installation package.
To check if your machine is not in this situation – search the machine for all instances of fcgiext.dll and make sure that all of them have the version number 6.1.36.1. If you found fcgiext.dll in %WINDIR%\system32\inetsrv or %WINDIR%\SysWOW64\inetsrv that has any other version number then read on.
To fix the installation if your machine has the wrong version of FastCGI binary – go to “Add Remove Programs” UI and remove all instances of FastCGI. After that run the installer for FastCGI RTM version.

IIS on Vista Patches As Of 11/15/2007

mailant — Fri, 16 Nov 2007 02:31:00 GMT

So far, we have only had two patches for IIS7 on Vista.

Behavior: We have seen some CGI applications (especially PHP applications that use the built-in CGI component) fail on Vista because we are unable to support multiple "cookies" for authentication or personalization.
We fixed that problem with this GDR: http://support.microsoft.com/kb/932385
Behavior: We learned that if you modified config for any delegated feature, IIS ended up removing your custom values in config.
We fixed that problem with this QFE: http://support.microsoft.com/kb/930451

Once you get these two patches on your Vista installation of IIS, you will be up to date as of today for any known fixes for IIS on Vista. These fixes are also built into Windows Server 2008.