JupyterLab: "Assertion failed: Connection reset by peer [10054]": what could I do about it?

[maajdl]

Hello,

I am unable to use JupyterLab for more that a few hours. The same for JupyterNotebook.
Below is a copy of the console screen when the crash occurs as well as the Windows logs.
I have this problem since a long time, and I have re-installed Anaconda3 at least 4 times without resolving the issue.
I have seen that I am not alone wih this problem, but I know many people who do not have this problem.

Would you have a suggestion about how to look for the cause and find a practical solution?
What should I check for?
What could I try?
How could I get mor information about what happens?

Thanks for your help,

Michel

==============================================================
[I 23:26:08.445 LabApp] JupyterLab extension loaded from C:\Users\myasus\Anaconda3\lib\site-packages\jupyterlab
[I 23:26:08.445 LabApp] JupyterLab application directory is C:\Users\myasus\Anaconda3\share\jupyter\lab
[I 23:26:08.445 LabApp] Serving notebooks from local directory: C:\Users\myasus\Documents_Work\Jupyter notebooks
[I 23:26:08.445 LabApp] The Jupyter Notebook is running at:
[I 23:26:08.445 LabApp] http://localhost:8888/?token=080bc3d02831103d726de11e1d1ada48197df38b4e55e25c
[I 23:26:08.445 LabApp] or http://127.0.0.1:8888/?token=080bc3d02831103d726de11e1d1ada48197df38b4e55e25c
[I 23:26:08.445 LabApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 23:26:08.507 LabApp]
To access the notebook, open this file in a browser:
file:///C:/Users/myasus/AppData/Roaming/jupyter/runtime/nbserver-6468-open.html
Or copy and paste one of these URLs:
http://localhost:8888/?token=080bc3d02831103d726de11e1d1ada48197df38b4e55e25c
or http://127.0.0.1:8888/?token=080bc3d02831103d726de11e1d1ada48197df38b4e55e25c
[I 23:26:11.255 LabApp] Build is up to date
[I 23:26:11.588 LabApp] New terminal with automatic name: 1
TermSocket.open: 1
TermSocket.open: Opened 1
[I 23:26:14.235 LabApp] Kernel started: 5e0d118d-6f3d-40c1-8b3e-40d58b4071d0
[I 23:26:17.504 LabApp] Adapting from protocol version 5.1 (kernel 5e0d118d-6f3d-40c1-8b3e-40d58b4071d0) to 5.3 (client).
[I 23:26:17.504 LabApp] Adapting from protocol version 5.1 (kernel 5e0d118d-6f3d-40c1-8b3e-40d58b4071d0) to 5.3 (client).
[W 23:26:20.123 LabApp] Got events for closed stream None
2019-09-01 23:26:40.726003: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow
binary was not compiled to use: AVX2
2019-09-01 23:26:40.735857: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library nvcu
da.dll
2019-09-01 23:26:40.967865: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1640] Found device 0 with properties:
name: GeForce GTX 980M major: 5 minor: 2 memoryClockRate(GHz): 1.1265
pciBusID: 0000:01:00.0
2019-09-01 23:26:40.974296: I tensorflow/stream_executor/platform/default/dlopen_checker_stub.cc:25] GPU libraries are statically li
nked, skip dlopen check.
2019-09-01 23:26:40.981406: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1763] Adding visible gpu devices: 0
2019-09-01 23:26:44.220150: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1181] Device interconnect StreamExecutor with strengt
h 1 edge matrix:
2019-09-01 23:26:44.228235: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1187] 0
2019-09-01 23:26:44.232446: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1200] 0: N
2019-09-01 23:26:44.245595: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1326] Created TensorFlow device (/job:localhost/repli
ca:0/task:0/device:GPU:0 with 3236 MB memory) -> physical GPU (device: 0, name: GeForce GTX 980M, pci bus id: 0000:01:00.0, compute
capability: 5.2)
Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:371)
Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:371)
Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:371)
Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:371)
Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:192)
Assertion failed: Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:Connection re
set by peer [371)
10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:371)
Exception in callback BaseAsyncIOLoop._handle_events(1656, 1)
handle: <Handle BaseAsyncIOLoop._handle_events(1656, 1)>
Traceback (most recent call last):
File "C:\Users\myasus\Anaconda3\lib\asyncio\events.py", line 88, in _run
self._context.run(self._callback, *self._args)
File "C:\Users\myasus\Anaconda3\lib\site-packages\tornado\platform\asyncio.py", line 138, in _handle_events
handler_func(fileobj, events)
File "C:\Users\myasus\Anaconda3\lib\site-packages\terminado\management.py", line 196, in pty_read
s = ptywclients.ptyproc.read(65536)
File "C:\Users\myasus\Anaconda3\lib\site-packages\winpty\ptyprocess.py", line 189, in read
data = self.fileobj.recv(size)
ConnectionResetError: [WinError 10054] An existing connection was forcibly closed by the remote host
Assertion failed: Connection reset by peer [10054] (C:\ci\zeromq_1549025594292\work\src\signaler.cpp:371)

==============================================================
Windows log, Logged: 2019-09-02 02:11:23

Faulting application name: python.exe, version: 3.7.3150.1013, time stamp: 0x5cc0c771
Faulting module name: KERNELBASE.dll, version: 6.3.9600.19425, time stamp: 0x5d26b6e9
Exception code: 0x40000015
Fault offset: 0x000000000000908c
Faulting process ID: 0x1944
Faulting application start time: 0x01d5610bdda68fe4
Faulting application path: C:\Users\myasus\Anaconda3\python.exe
Faulting module path: C:\WINDOWS\system32\KERNELBASE.dll
Report ID: 31cb60c2-cd16-11e9-828b-5cc5d446f6c1
Faulting package full name:
Faulting package-relative application ID:

==============================================================
Windows log, Logged: 2019-09-02 02:11:23

Faulting application name: python.exe, version: 3.7.3150.1013, time stamp: 0x5cc0c771
Faulting module name: KERNELBASE.dll, version: 6.3.9600.19425, time stamp: 0x5d26b6e9
Exception code: 0x40000015
Fault offset: 0x000000000000908c
Faulting process ID: 0xca0
Faulting application start time: 0x01d5610be11811c9
Faulting application path: C:\Users\myasus\Anaconda3\python.exe
Faulting module path: C:\WINDOWS\system32\KERNELBASE.dll
Report ID: 326286f1-cd16-11e9-828b-5cc5d446f6c1
Faulting package full name:
Faulting package-relative application ID:

[sigiesec]

I think this is a duplicate of #1808, but unfortunately there is no known solution :(

[bluca]

It is - and usually there's some networking monitoring application that's killing TCP connections on the system, see that issue for some tips

[maajdl]

Thanks bluca,
Could you suggest me how to find out this TCP killer ?
Thanks
Michel

[valmelnikov]

The hint on some other piece of software killing the connection helped me: I removed all the default apps which were coming with my ASUS laptop (some stupid gaming and cloud disk app and may be some other) and finally, after3 years of restarting the Jupyter 3-5 times a day, I have it running on one session forever. Firewall is not the reason, I have it on now and it does not kill the Jupyter.

[maajdl]

Spot on!
Thanks a lot valmelnikov!
As I have an ASUS computer I disabled two ASUS startup programs without success.
But then I disabled the service 'AsusGameFirstService' and this was the guilty!
Now I can enjoy Jupyter for more than a few hours.

[srg9000]

Is there a way I can find such services? Is there a way to not get the server affected by such services?