Orchestrator: Exception from HRESULT: 0x80070006

Hi All,

I am facing an issue when running two Jobs at the same time (not always). The 2nd Job, instead of being set to Pending Status and wait until the 1st Job finishes, throws the following error:
Executor start process failed, reason System.Runtime.InteropServices.COMException (0x80070006): Controlador no válido. (Excepción de HRESULT: 0x80070006 (E_HANDLE))
at UiPath.Core.Session.OpenInteractiveWindowsSession(String username, String password, OpenSessionFlags flags, String application, String arguments, Int32 delay, IntPtr userToken, TS_CONNECT_OPTIONS options)
at UiPath.Service.Impl.Executor.StartProcessInSession(Guid executorInstanceId, String username, String password, SettingsDictionary execSettings, IntPtr userToken) in D:\a\1\s\Robot\UiPath.Service\Impl\Executor.cs:line 449
at UiPath.Service.Impl.Executor.<>c__DisplayClass17_1.<b__0>d.MoveNext() in D:\a\1\s\Robot\UiPath.Service\Impl\Executor.cs:line 161

The Robot runs on a Windows 10 machine.

Thanks

Hi @cnavarrova,
Please have a look here:

Maybe this will somehow help.

Thanks for your reply, but it did not solve the problem.

Can you please write more details. Which version of Orchestrator? EE or CE edition? How robots are configured?

  • EE Orchestrator 2018.3.1
  • 1 Robot - 1 Machine
  • The problem does not occur always. Only with those 2 specific processes.

I’m not sure but based on your error it looks like something is wrong with credentials for those two processes. Like one process is trying to log into same account where other process is already working. Please check if for example same thing will happen for separate accounts (of course if you have unattended robot possibility).

They have to be under the same account because we only have one user so far.
Anyway, in theory, when the 1st process ends, it should log out, so I don’t know why the 2nd process could still find the 1st process working…

I’m just guessing but maybe logoff process is taking so long that second job is trying to start when logoff is still in place and it stuck at Pending state.

Looking at the exception it makes sense what you say. However, in such case it would be an Orchestrator issue, because even if it takes too long to logoff the next process should wait in Pending until the user is “free”.

The solution I found was a bit odd. The previous process finishes sending an HTML mail, and although the process ends, outlook keeps active, even if the mail has already been delivered.
I think this was the issue, since adding an Outlook Kill activity allowed the next process to start properly…

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.