1000s of policies checked during a Task Sequence

Amnon Feiner 25 Reputation points
2026-09-19T04:24:42.8133333+00:00

I have a weird situation where during a task sequence launch via software center, thousands of policies that look like this are evaluated, cuasing the TS to start after about an hour.
PolicyID      : ScopeId_E9C75880-3496-4E56-BB67-DDEFFCA95E59/AuthList_E2E71567-B6BF-4E2A-B046-9EE717894382/VI

PolicyVersion : 76.00

PolicyHash    :
After running reset and cleanup, the count is much lower and the TS starts withing 5 minutes, however it happens the next day as well. I check base lines, deployments and everything I can think of, no luck.
How do I find where they come from or the cause?

Microsoft Security | Intune | Configuration Manager | Deployment

2 answers

Sort by: Most helpful
  1. Rahul Jindal 11,731 Reputation points
    2026-09-19T06:16:50.1933333+00:00

    It is probably the referenced content. What is the TS for? How is the assignment created handling downloading of the content - before or during the execution of the TS?

    Was this answer helpful?


  2. AI answer

    2026-09-19T04:24:59.07+00:00

    Those entries are task sequence and dependent policy data being pulled and expanded at task sequence start.

    Key points supported by the logs and documentation:

    1. At the beginning of a task sequence, the client downloads the task sequence policy body from the database and stores it in the task sequence environment (TSEnv.dat). MP_GetPolicy logs this activity.
    2. The task sequence policy includes dependent policies, not just the task sequence itself.
    3. During steps such as Install Application, the client retrieves application policy from the task sequence environment, decompresses it, and stores policy in WMI under root\ccm\policy\actualconfig.
    4. Large task sequence policy is a known cause of slow or failed task sequence processing. The expanded client-side policy can exceed 32 MB even when the task sequence object in the database is smaller.

    To find the source of the large number of evaluated policies:

    1. Check MP_GetPolicy on the management point.
      • Search by the Deployment ID or Task Sequence ID.
      • This shows the request where the task sequence policy body is retrieved.
    2. Check SMSTS.log during task sequence startup.
      • Look for policy retrieval from the task sequence environment and policy decompression.
      • If the delay happens around Install Application, that step is reading application policies from TSEnv.dat and expanding them.
    3. Review what is deployed to the same collection as the task sequence.
      • A documented way to reduce task sequence policy size is to reduce the number of software updates deployed to the same collection.
    4. Review task sequence size in the Configuration Manager console.
      • Add the Size (KB) column to the task sequence list to identify large task sequences.
      • Keep each task sequence under 2 MB in the database.
    5. Reduce policy carried by the task sequence.
      • Move functional sections into child task sequences using Run Task Sequence.
      • Reference PowerShell scripts from a package instead of embedding script text in Run PowerShell Script steps.
      • Review custom task sequence variables, because the task sequence environment has an 8-KB limit and variables also contribute to policy size.

    Most likely cause in this scenario:

    • The task sequence deployment is carrying a large amount of dependent policy, commonly from application installs, software updates targeted to the same collection, or an oversized task sequence structure. Reset and cleanup temporarily lowers the local policy footprint, but the next policy refresh brings the same content back.

    What to focus on first:

    1. MP_GetPolicy for the task sequence deployment request.
    2. SMSTS.log to see which phase is spending time reading/decompressing policy.
    3. The task sequence Size (KB) value.
    4. Applications and software updates associated with the task sequence deployment collection.

    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.