OperationCanceledException #2977
Unanswered
mansohrstrom
asked this question in
General
Replies: 1 comment
-
|
From the stack traces this doesn't look like pool exhaustion to me, the failures are occuring after you already have established connections.
What do the stack traces look like when you use HTTP/1.1? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
We are struggling with a problem on version 2.3 of Yarp in a .NET 9 api hosted on IIS 10 (Windows server 2022).
Client => GW => Inner API
We have default route settings. When we do load tests and issue multiple requests in sequence without delays we suddenly get exceptions.
We have found that setting
request.Headers.ConnectionClose = true;
will delay the exception but they will just occur later. We have tried to fiddle with HttpClient settings in yarp.json (MaxConnectionsPerServer etc) without any luck. We've disabled HTTP2 in client/IIS. Does anyone recognize this and can point us in this right direction? It feels like a connection starvation somewhere. We use AppInsights for logging at GW and sometimes the exceptions/logs aren't even sent to Azure from GW (because the http connection pool is starved?).
We suspect something in IIS because we can't replicate it locally (dev laptop).
Below are complete stack trace from client + GW + inner API.
Grateful for any help!
Mans
Our .NET based test client gets:
And the inner API (route target) reports:
Beta Was this translation helpful? Give feedback.
All reactions