Closed pampua84 closed 1 year ago
I've recently come across this while we were load testing one of our other apps that relies on APIs we have implemented into our Identity Server application.
We noticed that the default cap for pool size is set by SQL Server to be 100 if you don't specify it in the connection string.
So we set Max Pool Size=32767;
on the end of our connection string and despite heavy load testing, we have not encountered these issues again and the active connections to our SQL Server have managed themselves back down after the increased load.
Hello, first of all thanks for the reply, I wanted to ask you if you had ever tried to use such a solution:
instead of manually specifying the MaxPoolSize in the connection string. This actually solves the problem, perhaps this could also have been solved by implementing a better horizontal scaling plan on the servers. What do you think about it? Thank you
I would also consider adding caching to the configuration stores, if you're not already doing that. In the stack trace originally posted, the timeout is occurring during a query for config data. Normally that configuration changes very infrequently, so caching it makes a lot of sense. Docs on configuration caching can be found here: https://docs.duendesoftware.com/identityserver/v6/data/configuration/#caching-configuration-data.
Hello everybody, very often i have this kind of error on indetity server:
and this is the stacktrace:
How can I understand what causes a load so high that the connection pool is saturated? Have you ever met such problem? Thank you.