IpSuperTcp channels: SuperChannel TimeOut or Actively refused

Does this happen server- or client-side ?

Is it possible to attach debugger once this happens to see where exactly execution is paused / which exceptions are raised?

A post was split to a new topic: .NET RO server hanging if it is running over night

It happens client side. I just received an email it occurs daily now and sometimes multiple times a day.

It’s not possible to attach a debugger. Could it work if I ask them to create a dump file and debug it in my Visual Studio to see what happened?

I assume simple restart of the client app clears this out?

Yuo mean you would ask them to create a memory dump using Task Manager and then you will open this file in Visual Studio? Yes, this would help.

Yes. The have to kill the client app with the Task Manager and reopen it. They are already making my ears bleed today because they are very annoyed with it :tired_face:

I just asked their IT department to create e memory dump if it happens so you will here from me soon.

Hello,

We at last got a dump file for this issue. As I’ve mentioned before, the application just hangs and with a restart of the client app (kill it using the Task Manager and restarting it) our client can work again. This happens between 2 to 5 times a day. Visual Studio and WinDbg gave me the following information:

If you want, we can send you the dump file, but it’s quite large. So we can use WeTransfer, but to which email address can we send it?

Thank you.

Please send it or a link to it to support@
However it is way too low-level.
Could you also provide a netstat -a result when the app hangs?

Hello,

Here we are again. A new year, same problems… So, the problem got a lot better since our last post, but now it just exploded. “Connection refused” problems every day when some users start the application (if the application runs, it’s mostly ok), sometimes multiple times… The last thing we tried was to disable DeepSec serverside, but it didn’t help. We’ve attached the netstat files (baseline when everything is ok and a file when the issue occurs). Our application runs on port 8090 TCP-IP.

Thank you.

NetstatABN Without DeepSec With issue 26 02 2021.txt (46.5 KB) Baseline NetstatABN Without Deep Sec 24 02 2021.txt (22.9 KB)

Hello

There are several interesting yet worrying entries in the files you provided.
At first let’s take a look at connections to the port 8090. Almost all of them are in ESTABLISHED state as they should be.

Now let’s take a look at something that listens at port 82 (AFAIK it is something Tor-related)
There’s a dozen of leaked sockets for connections on this port (TIME_WAIT status).
Also there is a lot of leaked IPv6 connections (lines like

TCP [::1]:58823 [::1]:83 TIME_WAIT

)
Also there are leaked sockets described by lines like

TCP 10.198.51.6:58773 10.197.18.196:443 TIME_WAIT

Looks like something tries to connect via https to some host and fails without closing connection properly.

Please note that none of these ports are related to your server app.
Could you contact your network administrator and check if there is any unwanted software on your server host?

Hello Anton,

Sorry for the delay. I’ve sent your concerns to my client. To be continued.
Thank you very much for taking the time for this problem.

Sincere greetings.