Service Fabric HTTP client image store upload timeout when running in .NET 6 application

Question

We have a custom deployment engine that uses the Service Fabric HTTP client (Microsoft.ServiceFabric.Client.Http) to deploy Service Fabric applications. It runs fine on .NET Framework 4.8, but if we use .NET 6, our second application upload to the Service Fabric image store times out, along with all subsequent uploads from within the same deployment process. The first app upload takes about 15 seconds, and the second times out after 100 seconds (the HTTP default timeout) even though the app content is the same size. It happens every time we run, but if I switch back to .NET Framework 4.8, it works fine (all apps upload within about 15 seconds).

I have tried disabling all of the new socket and IPv6 stuff in my process, but no luck.

Answer 1

@Thomas Hammond

Welcome to Microsoft Q&A Platform, thanks for posting your query here.

Currently, .NET 6 is still not fully supported in SF.

Checked with internal team on this, The GA support for .NET 6 is targeted for SF 9.0 release, the release date is not yet decided.

Support for IHost and IWebHost in AspNetCoreCommunicationListener is shipped in 5.2.1363, but the docs and samples are work in progress. Team is working on more framework support.

Similar ask here: https://learn.microsoft.com/en-us/answers/questions/636450/service-fabric-net-6-support.html

Hope that helps.
Please 'Accept as answer' if it helped, so that it can help others in the community looking for help on similar topics.

Answer 2

Ironically, our .NET 6 apps running on Service Fabric (a VM in a SF cluster) seem to work fine. I was referring to an external app running on a regular non-SF machine (Windows 10) using .NET 6 and the SF HTTP client to deploy applications into a cluster.