web
You’re offline. This is a read only version of the page.
close
Skip to main content

Announcements

News and Announcements icon
Community site session details

Community site session details

Session Id :
Customer experience | Sales, Customer Insights,...
Suggested Answer

General performance degradation

(1) ShareShare
ReportReport
Posted on by 4

Hello,

For more than two months our production CRM is having performance degradation where:

  1. Forms take excessive time to load
  2. CRUD actions take ~40 seconds (individual CRUD action, like save, delete, update)
  3. Plug-ins timing out
  4. Cloud flows running longer (the Dataverse connector)

We contacted MS support day one we noticed the issues; however, we got meaningful response, or rather one generic email a week…

We did a lot of investigations, but performance still is unstable. What we found was that plug-ins fail, not because logic is too complicated, but because "service.Update" takes more than a half a minute per action. Same happens with Dataverse connector. In addition, we do not see this degradation in UAT, even though the environments are identical. Plug-ins did not change at all for more than two years. Below I attach API calls and Plug-in performance:

From images we can see that API calls are going over the peak limit by MS account in one day. In plugins we see instability throughout whole month, where in past it would be almost straight line on ~300ms.

What could be the issue? Do those MS accounts go over peak limit impact the general CRM performance? Even when Plug-in statistics show the speed dropped, why it still takes over 30s per “service” action? Where can I look to see more potential issues?

I have the same question (0)
  • Suggested answer
    11manish Profile Picture
    1,215 Super User 2026 Season 2 on at

    Given that this has persisted for more than two months and affects forms, CRUD operations, plug-ins, and Power Automate, I would continue to treat this as a Microsoft support escalation rather than an application issue. When engaging Microsoft, provide evidence such as:

    Plug-in Profiler or Plug-in Trace Log timings showing that service.Update() is the bottleneck.
    API response time metrics before and after the degradation.
    Comparative performance between UAT and Production.
    Environment Analytics screenshots (API usage, capacity, long-running operations).
    Correlation IDs (x-ms-request-id or ActivityId) from slow API calls.
    Exact timestamps (UTC) of slow requests to aid backend log analysis.

    If you haven't already, request that the case be escalated to the Dataverse Product Group/Engineering team. With identical solutions performing differently across environments and the latency occurring within the Dataverse service itself, this is the type of issue that typically requires investigation of the underlying platform infrastructure rather than further application-level troubleshooting.

  • rajeshjamrodh Profile Picture
    31 on at

    Hi,

    A degradation in production performance within just two months is generally unlikely to be caused by a single factor. Multiple aspects should be analyses before identifying the root cause.

    As per Microsoft best practices, please verify the following: 

    Review the Power Platform Admin Centre to check the overall health of the production environment, including storage capacity, database usage, file storage, and log consumption.

    Monitor database growth and ensure sufficient free capacity is available.

    If records are frequently shared and unshared, review the growth of the PrincipalObjectAccess (POA) table, as excessive POA records can significantly impact Dynamics 365 performance.

    Analyse plugins, synchronous workflows, Power Automate flows, JavaScript, integrations, and long-running queries that may be affecting response times.

    Review audit logs, bulk data operations, duplicate detection jobs, and system jobs for abnormal growth.

    Monitor SQL performance, API throttling, and integration latency if external systems are involved.

    Establish monthly health checks and monitor key performance metrics proactively so that corrective actions can be taken before they impact production.

    Following Microsoft's performance and governance best practices, along with regular monitoring of environment health and database growth, will help identify the root cause and maintain optimal system performance.

     

  • ParthPatel249 Profile Picture
    4 on at

    The escalation advice is solid. Collecting Correlation IDs, Plugin Trace Logs, and UTC timestamps before engaging Microsoft engineering is the right move. However, I wouldn't fully rule out the application layer yet.

    Looking at the API analytics, there's a massive spike hitting close to 1 million calls around 04:00 AM. That pattern strongly suggests a scheduled batch job, nightly sync, or a Cloud Flow with retry storms running at that time. When an org breaches Dataverse service protection limits, throttling is applied at the org level, meaning every user, plugin, and flow gets queued, which explains why service.Update takes 30+ seconds even though the plugin logic itself hasn't changed. The plugin isn't slow; it's waiting for a throttled platform to respond.

    UAT being unaffected doesn't automatically confirm a platform infrastructure issue. UAT simply isn't receiving the same API volume, so it never hits the throttle threshold.

    My suggestion: identify what's generating that 04:00 spike first (check scheduled Cloud Flows, Azure integrations, and any nightly jobs using the Dataverse connector). Then escalate to Microsoft with evidence that you've already ruled out the application layer. That's what gets a ticket taken seriously at the engineering level.

  • PP-21090025-0 Profile Picture
    4 on at

    Is there a way to see if environment is being throttled?

Under review

Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.

Helpful resources

Quick Links

Season of Sharing Community Challenge Winners!

Congratulations to our community stars!

Women in Power Builds Momentum

Expanding mentorship, skilling, and AI innovation

Congratulations to the July Top 10 Community Leaders

These are the community rock stars!

Leaderboard > Customer experience | Sales, Customer Insights, CRM

#1
11manish Profile Picture

11manish 80 Super User 2026 Season 2

#2
Muhammad Shahzad Shafique Profile Picture

Muhammad Shahzad Sh... 35 Most Valuable Professional

#3
Daniyal Khaleel Profile Picture

Daniyal Khaleel 32 Most Valuable Professional

Last 30 days Overall leaderboard

Product updates

Dynamics 365 release plans