Some machines don't log into Kinetic 2022.2

GInclan · January 24, 2023, 3:34pm

Hi, I thought that trying here might get a faster response.

We just updated from 10.2.500 to Kinetic 2022.2, everything went fine and we have most users with their client software upgraded and working.

However, some machines refuse to log in the user into Kinetic. We have a self signed certificate for our server, since this is intranet access only and just installing the certificate in the client machines worked for all the other computers.

But some(10) still refuse to connect, I am assuming that is something related to SSL, but I am no expert and can’t find a way to debug this. I have a support case open, but I am struggling to get help from them. At least a fast response, since I have 10 users without access now.
Is there any way to debug and find the cause? The failing computers have windows 8.1 Prof, but some other computers with that OS conected without issue. One windows 10 computer also fails.

Thank you, any help will be appreciated

hmwillett · January 24, 2023, 3:36pm

Browser or client?
What’s the error?

Joseph_Martin · January 24, 2023, 5:02pm

You can probably find more information in the event viewer on the computers that are failing, under the Epicor Software Corporation Client Log.
Also, try logging in as the failing user on another computer, if it works there then it is machine related and not user related at least.

My guess would be corrupted personalizations or a .NET version requirement that is not installed.

GInclan · January 24, 2023, 7:21pm

It is the thick client(windows), not the web client. It is certainly machine related, since the same user works in another computer without issue.
I’ll check those logs.

GInclan · January 24, 2023, 7:33pm

Forgot to add the error, it is:
Application Error

Exception caught in: mscorlib

Error Detail

Message: An error occurred while sending the request.
Inner Exception Message: The underlying connection was closed: An unexpected error occurred on a send.
Program: CommonLanguageRuntimeLibrary
Method: ThrowForNonSuccess

Client Stack Trace

at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
at Microsoft.Extensions.Http.Logging.LoggingHttpMessageHandler.d__5.MoveNext()
— End of stack trace from previous location where exception was thrown —
at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
at Microsoft.Extensions.Http.Logging.LoggingScopeHttpMessageHandler.d__5.MoveNext()
— End of stack trace from previous location where exception was thrown —
at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
at Ice.Cloud.ProxyBase1.<ExecuteAsync>d__60.MoveNext() --- End of stack trace from previous location where exception was thrown --- at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task) at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task) at Epicor.Utilities.AsyncHelper.RunSync[TResult](Func1 method)
at Ice.Cloud.ProxyBase1.Execute(String methodName, RestValueSerializerBase serializer, ProxyValuesIn valuesIn, ProxyValuesOut valuesOut) at Ice.Cloud.ProxyBase1.<>c__DisplayClass58_0.b__0(Context _)
at Polly.Policy1.<>c__DisplayClass32_0.<Execute>b__0(Context ctx, CancellationToken ct) at Polly.Retry.RetryEngine.Implementation[TResult](Func3 action, Context context, CancellationToken cancellationToken, IEnumerable1 shouldRetryExceptionPredicates, IEnumerable1 shouldRetryResultPredicates, Func1 policyStateFactory) at Polly.RetryTResultSyntax.<>c__DisplayClass12_01.b__0(Func3 action, Context context, CancellationToken cancellationToken) at Polly.Policy1.ExecuteInternal(Func3 action, Context context, CancellationToken cancellationToken) at Polly.Policy1.Execute(Func3 action, Context context, CancellationToken cancellationToken) at Polly.Policy1.Execute(Func2 action, Context context) at Ice.Cloud.ProxyBase1.CallWithCommunicationFailureRetry(String methodName, ProxyValuesIn valuesIn, ProxyValuesOut valuesOut, RestRpcValueSerializer serializer)
at Ice.Cloud.ProxyBase1.CallWithMultistepBpmHandling(String methodName, ProxyValuesIn valuesIn, ProxyValuesOut valuesOut, Boolean useSparseCopy) at Ice.Cloud.ProxyBase1.Call(String methodName, ProxyValuesIn valuesIn, ProxyValuesOut valuesOut, Boolean useSparseCopy)
at Ice.Proxy.Lib.SessionModImpl.Login()
at Ice.Core.Session.GetSessionId(String asUrl, String companyId, String plantId)
at Ice.Core.Session.InitSessionMod(String asUrl, Boolean fwVerCheck, String companyID, String plantID, String sessionId)
at Ice.Core.Session.InitSession(Action setCredentials, String asUrl, Guid licenseTypeId, String pathToConfigurationFile, Boolean fwVerCheck, String companyID, String plantID, Boolean useChannelCacheForServices, String sessionID)
at Ice.Core.Session…ctor(String userID, String password, String asUrl, Guid licenseTypeId, String pathToConfigurationFile, Boolean fwVerCheck, String companyID, String plantID, Boolean useChannelCacheForServices, String sessionID)
at Ice.Core.Session…ctor(String userID, String password, Guid licenseType)
at Ice.Lib.LogOn.CreateSession(String userID, String password, String appServerUri, Guid licenseType, SessionTokenType sessionTokenType, Object azureADOwnerWindow)
at IceShell.Apps.LogonDialog.logOn(String userID, String password, Boolean promptUpdatePassword)
at IceShell.Apps.LogonDialog.DoWorkLogon()

Inner Exception

The underlying connection was closed: An unexpected error occurred on a send.

at System.Net.HttpWebRequest.EndGetRequestStream(IAsyncResult asyncResult, TransportContext& context)
at System.Net.Http.HttpClientHandler.GetRequestStreamCallback(IAsyncResult ar)

Inner Exception

Unable to read data from the transport connection: Se ha forzado la interrupción de una conexión existente por el host remoto.

at System.Net.TlsStream.EndWrite(IAsyncResult asyncResult)
at System.Net.PooledStream.EndWrite(IAsyncResult asyncResult)
at System.Net.ConnectStream.WriteHeadersCallback(IAsyncResult ar)

Inner Exception

Se ha forzado la interrupción de una conexión existente por el host remoto

at System.Net.Sockets.Socket.EndReceive(IAsyncResult asyncResult)
at System.Net.Sockets.NetworkStream.EndRead(IAsyncResult asyncResult)

Hally · January 24, 2023, 10:11pm

I would attempt to reinstall the client. Perhaps the autoupdate on the clients did now work correctly.

I’m going to assume the effected client machines has all the required .net perquisites

I hope that helps.

GInclan · January 24, 2023, 10:56pm

I’ve tried that(re-installing), but no change. Strangely enough the Web Client does work, it is something related to the Thick Client and it would seem that it is no entirely SSL at fault.
Also, checked the logs and I only see the record for the error, no other messages are in there.

Hally · January 24, 2023, 11:01pm

A clean install or just an upgrade, perhaps some files are not being overwritten for some reason.

I would also be renaming\deleting the c:\ProgramData/Epicor folder in case you have any artifacts in the cache…

Which begs me to ask two things:

Do you use the Alternate Cache folder in your sysconfig? If the cache data won’t be in c:\programdata\epicor folder
There isn’t a second user logged into this computer with Epicor running, preventing the update working correctly?

GInclan · January 24, 2023, 11:20pm

Simon,

Yes, we’ve tried the clean re-install. We deleted the whole c:\Epicor folder, along with the c:\ProgramData\Epicor one. We are trying desperate measures now and deleting antivirus software to check it out.

No, we don’t use alternate cache folder and the machine is just for one user. No other remote (or local, for that matter) user is logged.

hmwillett · January 24, 2023, 11:26pm

Have you tried clearing the AppData ones?

C:\Users\username\AppData\Local\Epicor

I usually just rename the folders in there as _OLD to see if it helps before deleting them entirely.

GInclan · January 24, 2023, 11:33pm

Just tried those, Hannah, but still no. The antivirus is taking too long to uninstall, so still waiting to try after that. I am testing in the computer that I have closer, since I can’t go to the others fast enough. Luckily the user is out at the moment.

bmanners · January 25, 2023, 12:04am

Check the date Time setting on the bad machines. It has to be in the correct timezone and time or you can get errors.

Brett

GInclan · January 25, 2023, 1:05am

Thank for the suggestion Brett, but in this case all the machines have the same timezone and time… Contacted support again today, just to find out the person who was seeing my case is out until Jan 25, so thank you very much Epicor. The fact that support changes Urgent cases as Medium,even when we are talking about users can’t log in is surprising.

hmwillett · January 25, 2023, 1:33am

Escalate the case and get your CAM involved. They’ll typically go to bat for you if you give them the case number.

CTapia · January 25, 2023, 10:46pm

Hi @GInclan

I have the same error, today I upgraded to Epicor and some computers have connected and work fine, but others have the same error and cannot connect. Have you found any solution for that problem?

Carlos

ckrusen · January 25, 2023, 11:40pm

Grasping a at straws…

Run a TRACERT to the server. Do this on the workstations that work and those that don’t. Look for differences.

Try flushing the workstations DNS cache.

Rich · January 26, 2023, 12:39am

Other things to review:

Verify that your nonworking systems are configured to support TLS 1.2 - Google that for more detail.

Verify the Windows Shortcut is starting the correct Client and that the Sysconfig file defined is also correct and has the proper AppServer URL and Auth values - compare one of the working Sysconfig files to the nonworking Sysconfig to ensure they are the same.

Try using the 64 bit version of the client - Epicor64.exe or Kinetic64.exe.

GInclan · January 26, 2023, 12:39am

Support has finally reported, but with no good news. They claim that Kinetic 2022.2 doesn’t support windows 8.1. Even though I do have some computers with Windows 8.1 working! So, I am thinking this the end of line.

Strangely enough, those prerequisites are not in the installation or update guide, so not sure where did that come from.

Thank you all for your suggestions!

GInclan · January 26, 2023, 12:43am

Hi Rich,
I wasn’t aware there is actually a 64 bit version of the software, will try it. Although support indicates that no 32 bit OS is supported, strangely enough Kinetic is still 32 bit also.
In any case, thanks again, I’ll give it a rest and update the computers, since this is taking a toll on me already.

Hally · January 26, 2023, 1:05am

Not wanting to hijack this thread, but what is the guidance around using the 64bit version? I have never seen anything about it in installation documentation, although It’s been there for a long time.