Connection Time out when connection Oracle On-Prem DB

Pramod C 1 Reputation point
2020-11-25T17:37:19.583+00:00

We have an .NET Core 3.1 Api Application hosted in Azure Web App which connects to an on-prem Oracle DB through express route.

The connection works fine and we were able to run the app without any issues.

The issue happens once the load increase. We were load testing the APIs and we are getting connection time out error frequently

We have set min connection pool to 50 and max connection pool to 100. It will run fine initially and when load increase we are seeing this error.

We are calling the API through APIM and there is a cache set in APIM for 180 seconds.

We have cleared all sessions in DB and ran the tests and after running the tests we still have 89 active sessions which decreases gradually over a period of time.

ODP.NET core is used for calling the oracle db. We are opening and closing the connections and each objects are in 'using' block.

Not sure why this happens. Any inputs are appreciated.

Azure ExpressRoute
Azure ExpressRoute
An Azure service that provides private connections between Azure datacenters and infrastructure, either on premises or in a colocation environment.
342 questions
0 comments

3 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. SaiKishor-MSFT 17,216 Reputation points
    2020-11-25T22:59:57.62+00:00

    @Pramod C

    Since you mentioned that it works fine initially with low load, I suspect this to be a bandwidth related issue. To troubleshoot further-

    • Please check to see if your express route bandwidth threshold is being reached. You can use Express Route monitoring tools available on Azure to check the same. If the bandwidth threshold is being reached, please see if you can upgrade the bandwidth speed and see if that helps with the timeout errors.
    • You could further do an Iperf test to determine the bandwidth being achieved over this Express route connection to further understand if you are getting the speed as expected.
    • You can also try connecting via internet or S2S VPN if there is a way to do this and see if that makes any difference to see if the network is the bottleneck or the endpoints.

    Please let me know further if you need any more assistance and we will be glad to help. Thank you!

    0 comments
  2. Pramod C 1 Reputation point
    2020-11-26T13:49:51.15+00:00

    @SaiKishor-MSFT

    Not sure it's an issue with express route as when we have connection min pool set to 10 we are getting the same issue for 600 users after some time. When we increased the pool size to 50 and maxpool 100 the error is not happening for 600 users but started happening for 1000 users.

    Thanks for the input. We will check the bandwidth of express route and see any bottleneck is over there.

    0 comments
  3. Pramod C 1 Reputation point
    2020-11-26T15:49:22.093+00:00

    @SaiKishor-MSFT

    We have checked the Express route and it has an allowance of 200mb/s and when we ran the load test for 1000 users it's not taking more than 50mb/s.