Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updating from ASP.NET Core 2.0 to 2.1 (net471) causes unstable Azure App Service #3389

Closed
kspearrin opened this issue Aug 6, 2018 · 2 comments
Labels
✔️ Resolution: Duplicate Resolved as a duplicate of another issue

Comments

@kspearrin
Copy link

kspearrin commented Aug 6, 2018

I wasn't quite sure where to post this issue, so please feel free to move it to the correct area if this is not appropriate.

I have a production API application deployed to Azure App Services that has been stable and running without issue on ASP.NET Core 2.0.x with .NET Framework 4.7.1.

After updating the app to ASP.NET Core 2.1.2 (still with .NET Framework 4.7.1) and deploying to Azure things have become unstable. The application is running and is usable, however, CPU is spiking up and down and application response times are all over the charts. Reverting the update and re-deploying 2.0 causes things to go back to normal and stable.

Here are some charts from a period of this deployment:

Application Response Times

enter image description here

CPU Percentage

enter image description here

The only change in these deployments was updating the project libraries from 2.0 to 2.1 following the migration from ASP.NET docs and then reverting the upgrade back to 2.0. It should also be noted that these graphs show ASP.NET Core 2.1 in a period that is normally in low demand (middle of the night) and charts are usually at their lowest levels.

App Insights does not seem to point to a single endpoint as causing these issues. Distribution of slow/spiking response times seems uniform across the application.

I took a 5 minute performance profile from Kudu during the unstable 2.1 deployment. I am not really an expert at reading these things, but from one of the performance spikes I was able to see some significant sampling coming from IISMiddleware and HostingApplication.

image

I am not sure if this .diagsession file contains sensitive information or not, so I have not attached it publicly here. If someone from Microsoft would like me to send it over, please let me know and I will be glad to.

After trying to debug this more, I decided to deploy using netcoreapp2.1 rather than net471 to see of that would change anything. After deploying ASP.NET Core 2.1 with netcoreapp2.1 things are now perfectly stable as was before with ASP.NET Core 2.0 and net471. So it would seem that the issue is only present when using net471.

@Tratcher
Copy link
Member

Tratcher commented Aug 6, 2018

@halter73 @sebastienros

@halter73
Copy link
Member

halter73 commented Aug 6, 2018

Thanks for the report. This appears to be a duplicate of aspnet/KestrelHttpServer#2694 which we're currently looking into.

@halter73 halter73 closed this as completed Aug 6, 2018
@halter73 halter73 added the ✔️ Resolution: Duplicate Resolved as a duplicate of another issue label Aug 6, 2018
@ghost ghost locked as resolved and limited conversation to collaborators Dec 3, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
✔️ Resolution: Duplicate Resolved as a duplicate of another issue
Projects
None yet
Development

No branches or pull requests

3 participants