Outgoing service/database spans disappear after daily IIS app pool recycle; only a VM restart restores them

[zaidanrizq]

Hi everyone,

We're seeing a recurring problem with distributed traces on one of our IIS-hosted .NET applications, and we'd appreciate any ideas.

Problem

Every day at 00:30, an automated job recycles the application pool through IIS Manager. After this recycle:

• The service still appears in Dynatrace, and incoming requests are still captured. (confirm/adjust)
• The spans for outgoing calls to other services and to the database are no longer captured. The PurePaths/traces end at the entry point of our service.
• The problem continues until we restart the whole VM. After the reboot, tracing works normally until the next scheduled recycle.

Environment

• OneAgent version: 1.347.49.20260928-064841
• OneAgent monitoring mode: Full-Stack
• Host OS: Windows Server 2022 Datacenter 21H2 2009, ver. 10.0.20348
• Technologies: 
  .NET (.NET Framework 3.5.1.0), ASP.NET (3.5.1.0), CLR (FullCLR 2.0.50727.9179), IIS App Pool (10.0.20348.1), MSSQL Client (2.0.50727.9176), WCF (3.5.1.0)

[AntonPineiro]

Hi,

I would open a support case.

Best regards