IIS application-pool recycling replaces a worker process without permanently stopping its application pool. A planned recycle can be useful for a supported application, but it is not a substitute for investigating memory leaks, failed dependencies or slow requests.
Understand the application impact
Identify the exact pool and every application assigned to it. Recycling a shared pool affects all of those applications. Ask the application owner how sessions, in-process caches, long-running requests and background jobs behave when a worker process is replaced. A successful recycle is not proof that users avoided disruption.
Overlapped recycling normally allows a replacement worker to start while the old worker drains requests, subject to the application’s configuration and shutdown limits. Data kept only in process memory can still disappear. Stateful applications need a tested session and job design before frequent recycling is introduced.
Record the existing configuration
Use IIS Manager to inspect the pool’s Recycling settings and Advanced Settings. Capture the existing interval, scheduled times, memory thresholds and recycle-event logging. Save an IIS configuration backup before changing a production server.
& $env:windir\system32\inetsrv\appcmd.exe add backup BeforePoolRecycling
Import-Module WebAdministration
Get-ItemProperty 'IIS:\AppPools\BusinessApp' -Name recycling.periodicRestartChoose one deliberate trigger
In IIS Manager, select Application Pools, choose the pool and open Recycling. For a daily maintenance time, configure Specific times and select an agreed quiet period. The wizard also exposes fixed intervals and other conditions. Review all enabled triggers together: adding a scheduled time while leaving a separate interval enabled can produce more recycles than intended.
Avoid copying an arbitrary memory threshold from another server. The suitable value depends on process architecture, available memory and the application’s normal behaviour. A threshold that is too low can create repeated cold starts during the busiest part of the day.
Enable logging and test the user journey
Record the relevant recycling reasons so an unexpected process replacement can be correlated with its cause. During a pilot maintenance window, recycle the selected pool once, observe the replacement worker process and test the actual authenticated application, not merely the IIS default page.
Restart-WebAppPool -Name 'BusinessApp'
Get-WebAppPoolState -Name 'BusinessApp'
& $env:windir\system32\inetsrv\appcmd.exe list wpDistinguish a recycle from a failure
If users report recurring interruptions, compare worker-process changes with the recycling schedule, IIS/WAS events, application logs and resource trends. A pool stopped by rapid-fail protection is a different incident from a scheduled recycle. Increasing or disabling recovery thresholds without diagnosing the crash can hide a serious problem.
Check application warm-up, certificate and database connections, session persistence and long-running work after the test. Preserve logs across the event so you can tell whether the problem precedes the recycle or results from it.
Rollback and reference
Restore the pool’s previously recorded trigger values and logging settings if the new policy causes disruption. Restore the whole IIS backup only when you intend to revert all configuration changes covered by that backup; doing so can overwrite unrelated changes made later.
Microsoft documents the interval, memory and schedule controls in periodic restart settings for application pool recycling. The default interval and available settings should be checked against the server configuration rather than assumed from a generic checklist.