How to fix HTTP Error 503 – The service is unavailable

• Author
• Recent Posts

Surender Kumar

Surender Kumar has more than twelve years of experience in server and network administration. His fields of interest are Windows servers, Active directory, PowerShell, web servers, networking, Linux, virtualization, and Kubernetes. He loves writing for his blog.

Latest posts by Surender Kumar (see all)

• kube-scheduler: The Kubernetes scheduler - Fri, Sep 29 2023
• Kubernetes logs: Pod logs, container logs, Docker logs, kubelet logs, and master node logs - Mon, Sep 25 2023
• Kubernetes DaemonSets - Wed, Sep 6 2023

Cause of HTTP Error 503

HTTP Error 503 is another very common error you might face with an IIS-hosted website in addition to the 500 – Internal server error.

HTTP Error 503. The service is unavailable

The primary reason for this error is a stopped or disabled application pool. The application pool is the IIS component that initiates a worker process, which then allows the web server to process incoming requests for a website. When something is wrong with the application pool, the web server throws an HTTP Error 503.

Rapid-Fail Protection

There can be various reasons for an application pool failure, such as an unhandled exception in the code, an expired password for the application pool identity, or something else. By default, IIS is configured to disable a malfunctioning application pool if its worker process faces five consecutive crashes within a period of five minutes. This setting is known as Rapid-Fail Protection, which is available under the Advanced Settings of an application pool. This is shown in the following screenshot:

Viewing or modifying the rapid fail protection setting for an application pool

• Enabled—The value True indicates that the rapid-fail protection feature is active, which is the default. Set the value to False to disable it.
• Failure Interval (minutes)—The time interval to specify the number of minutes (by default, 5) to count the failures for a process.
• Maximum Failures—The maximum number of failures that are allowed within the time interval mentioned above (by default, 5).

While troubleshooting an application pool-related problem, you can modify these parameters, if needed.

Debugging HTTP Error 503

To determine the exact reason why the application pool is disabled or stopped, open the Event Viewer, navigate to the System log, and look for error or warning events whose source is Windows Process Activation Service (WAS). WAS is responsible for starting/stopping the application pool and keeping track of worker processes in the IIS.

The following screenshot shows that the application pool is disabled because its identity is invalid:

Checking the system log to identify the cause of application pool failure

If you couldn't find anything in the System log, take a look in the Application log since the ASP.Net-related exceptions are logged there. As discussed above, when a worker process experiences five fatal errors or exceptions within five minutes, rapid-fail protection kicks in and automatically disables the faulting application pool.

Based on event logs, if you conclude that the reason for application pool failure is a bug in the application code, but you still want to keep the application pool running until your developer fixes it, you could temporarily adjust the behavior of rapid-fail protection, either by increasing the Maximum Failures or the Failure Interval to a higher value. You could even disable this feature by using the following PowerShell command:

(Get-IISAppPool awesomewebsite.com).failure.rapidFailProtection
Set-WebConfigurationProperty '//*[@name="awesomewebsite.com"]//failure' -Name rapidFailProtection -Value $False

Viewing and disabling the rapid fail protection for an application pool with PowerShell

Remember that the rapid-fail protection feature is there for the safety of the web server, so disabling it shouldn't be considered a solution to the problem. Instead, fixing the underlying cause (or bug causing the behavior) in the application code is really important. Anyway, disabling it sometimes comes in handy if developers are taking longer to fix the bug; meanwhile, you want to keep the web application up and running. Once rapid-fail protection is disabled, the web server won't disable the faulting application pool, and visitors will no longer see HTTP Error 503.

Fixing HTTP Error 503

A simple resolution for HTTP Error 503 is to start the application pool corresponding to your website. If you get an application pool identity-related error or warning, make sure the username and password for the identity are correct. If a custom username is used, make sure its password is not expired. If possible, set the password to never expire for that user.

If the application pool stops repeatedly but there is no application pool identity-related error in the system log, the issue is likely due to some fatal error or unhandled exception in the code. In that case, the application log can give you some useful insights. The following screenshot shows a typical error logged in the application log that occurred due to an unhandled exception:

Checking the application log to identify the cause of application pool failure

If you see such errors or warnings, notify your application developers, and ask them to fix the exception. If needed, share the detailed error message recorded in the event log.

Common 503 substatus codes

The following table covers the common HTTP 503 substatus codes, along with their possible causes and troubleshooting advice:

%SystemRoot%\System32\inetsrv\appcmd.exe set config /section:serverRuntime /appConcurrentRequestLimit:100000

%SystemRoot%\System32\inetsrv\appcmd.exe set config /section:system.webserver/serverRuntime /appConcurrentRequestLimit:100000

Conclusion

You can see that, like most IIS errors, HTTP Error 503 can be fixed with the help of detailed errors and failed request tracing. However, in the event of application pool-related errors, the HTTP.sys driver throws the 503.0 error because the application pool isn't available to handle the request. Thus, the detailed error or failed request tracing rule is not even triggered. In such situations, you can rely on Windows event logs to troubleshoot.