Skip to content
Anyone else seeing ...
 
Notifications
Clear all

Anyone else seeing high memory usage with the new CrowdStrike sensor update?

1 Posts
1 Users
0 Reactions
4 Views
(@crm_trailblazer_7)
Estimable Member
Joined: 3 months ago
Posts: 129
Topic starter   [#12274]

Just rolled out the latest sensor version (7.15.17806.0) across our Windows Server 2022 estate and memory usage is consistently 30-40% higher than the previous stable build. This isn't just a spike—it's sustained.

Our baseline monitoring shows:
* Previous build (7.13.x): Avg. working set ~180-220 MB per sensor.
* Current build (7.15.x): Avg. working set ~280-320 MB per sensor.
* Pattern holds across identical VM templates (8 vCPU, 32GB RAM).

We're not running any exotic configurations. Standard Falcon Prevent/Insight/Discover. The increase is significant enough to trigger our resource alerts on several dense hypervisors.

Before I open a ticket and get the usual "within expected parameters" response, I need data.
* Is this isolated to our environment or a wider pattern?
* Has anyone done a proper before/after comparison on Server 2019 or Win10/11?
* Any specific module (`csagent`, `csa`, `r30`) showing the delta?

If you're seeing this, please share your build version and memory metrics. Vendor demos are one thing, but production telemetry doesn't lie.


Show me the query.


   
Quote