[WIP] Named pipe cleanup #85

blink1073 · 2018-01-19T11:14:35Z

Potential fix for #59.

blink1073 · 2018-01-19T11:40:39Z

@SimonBiggs @dalito, do you mind trying out the snippet below with these changes?

You can make the changes by overwriting the winpty_wrapper.py file locally with the raw contents of the file in this PR, found in python -c "import winpty; print(winpty.__path__[0])".

from winpty import PtyProcess

proc = PtyProcess.spawn('python')
proc.write('print("hello, world!")\r\n')
proc.write('exit()\r\n')
while proc.isalive():
	print(proc.readline())

SimonBiggs · 2018-01-19T11:47:30Z

I won't get a chance till I'm back on my work computer on Monday. (It's Friday evening here now).

SimonBiggs · 2018-01-19T11:47:55Z

If it's not resolved by then though I shall give it a go :)

SimonBiggs · 2018-01-19T11:48:16Z

(All Linux at home)

dalito · 2018-01-19T16:41:02Z

I still get the same two errors (the order is random). Here is the result:

Python 3.6.4 (v3.6.4:d48eceb, Dec 19 2017, 06:54:40) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> from winpty import PtyProcess
>>> proc = PtyProcess.spawn('python')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\ptyprocess.py", line 113, in spawn
    inst = cls(proc)
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\ptyprocess.py", line 47, in __init__
    self._server.bind(address)
OSError: [WinError 10048] Normalerweise darf jede Socketadresse (Protokoll, Netzwerkadresse oder Anschluss) nur jeweils einmal verwendet werden
>>> proc = PtyProcess.spawn('python')
>>> Exception in thread Thread-1:
Traceback (most recent call last):
  File "C:\dev_amd64\Python36\lib\threading.py", line 916, in _bootstrap_inner
    self.run()
  File "C:\dev_amd64\Python36\lib\threading.py", line 864, in run
    self._target(*self._args, **self._kwargs)
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\ptyprocess.py", line 337, in _read_in_thread
    data = pty.read(4096)
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\winpty_wrapper.py", line 86, in read
    ReadFile(self.conout_pipe, data, length, None, None)
OSError: exception: access violation writing 0x0000000000000000

Do you have an idea why there are two different errors seen randomly? What can cause a timing problem?

Thanks for looking at this. I can continue to test possible fixes over the weekend.

blink1073 · 2018-01-20T11:25:02Z

The port error is most likely due to an antivirus software or system policy limiting access to ports. The default port is 20128. Can you please try changing that line locally to default to 8888 (the one typically used by the Jupyter Notebook), to see if that clears it up? We could allow for an environment variable to set the default port if it is an issue. I'm still looking into the access violation...

blink1073 · 2018-01-20T11:31:41Z

@dalito, I've updated the file in the PR with more debug info, and an attempt to read a second time if the first fails, can you please try that as well?

dalito · 2018-01-20T12:53:36Z

Here is the result

Python 3.6.4 (v3.6.4:d48eceb, Dec 19 2017, 06:54:40) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> from winpty import PtyProcess
>>> proc = PtyProcess.spawn('python')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\ptyprocess.py", line 113, in spawn
    inst = cls(proc)
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\ptyprocess.py", line 47, in __init__
    self._server.bind(address)
OSError: [WinError 10048] Normalerweise darf jede Socketadresse (Protokoll, Netzwerkadresse oder Anschluss) nur jeweils einmal verwendet werden
>>> proc = PtyProcess.spawn('python')
size 0
length 0
>>> conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 0
length 0
conout_pipe 392
size 8
length 8
conout_pipe 392
exception: access violation writing 0x0000000000000000
Exception in thread Thread-1:
Traceback (most recent call last):
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\winpty_wrapper.py", line 90, in read
    ReadFile(self.conout_pipe, data, length, None, None)
OSError: exception: access violation writing 0x0000000000000000

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\dev_amd64\Python36\lib\threading.py", line 916, in _bootstrap_inner
    self.run()
  File "C:\dev_amd64\Python36\lib\threading.py", line 864, in run
    self._target(*self._args, **self._kwargs)
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\ptyprocess.py", line 337, in _read_in_thread
    data = pty.read(4096)
  File "C:\dev_src\_pyve\stable_py36\Lib\site-packages\winpty\winpty_wrapper.py", line 93, in read
    ReadFile(self.conout_pipe, data, length, None, None)
OSError: exception: access violation writing 0x0000000000000000

blink1073 · 2018-01-24T21:40:53Z

Unfortunately I am no closer to understanding what is going on here.

SimonBiggs · 2018-01-25T03:45:01Z

@blink1073 let me know if there is anything I can do. It might be possible to do a remote session if you wanted to do more hands on troubleshooting.

SimonBiggs · 2018-01-25T03:50:26Z

Just for reference here is me running the results of this pull request on my machine.

Microsoft Windows [Version 6.1.7601]
Copyright (c) 2009 Microsoft Corporation.  All rights reserved.

C:\Users\sbiggs>python
Python 3.5.4 | packaged by conda-forge | (default, Aug 10 2017, 01:38:47) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> from winpty import PtyProcess
>>>
>>> proc = PtyProcess.spawn('python')
size 0
length 0
>>> conout_pipe 348
proc.write('print("hello, world!")\r\n')
size 0
length 0
conout_pipe 348
24
>>> size 0
proc.write('exit()\r\n')
length 0
conout_pipe 348
8
size 0
>>> length 0
while proc.isalive():
conout_pipe 348
size 0
... length 0
        print(proc.readline())conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 0
length 0
conout_pipe 348
size 8
length 8
conout_pipe 348
exception: access violation writing 0x0000000000000000
Exception in thread Thread-1:
Traceback (most recent call last):
  File "C:\Users\sbiggs\AppData\Local\Continuum\Miniconda3\lib\site-packages\winpty\winpty_wrapper.py", line 88, in read
    ReadFile(self.conout_pipe, data, length, None, None)
OSError: exception: access violation writing 0x0000000000000000

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\sbiggs\AppData\Local\Continuum\Miniconda3\lib\threading.py", line 914, in _bootstrap_inner
    self.run()
  File "C:\Users\sbiggs\AppData\Local\Continuum\Miniconda3\lib\threading.py", line 862, in run
    self._target(*self._args, **self._kwargs)
  File "C:\Users\sbiggs\AppData\Local\Continuum\Miniconda3\lib\site-packages\winpty\ptyprocess.py", line 331, in _read_in_thread
    data = pty.read(4096)
  File "C:\Users\sbiggs\AppData\Local\Continuum\Miniconda3\lib\site-packages\winpty\winpty_wrapper.py", line 91, in read
    ReadFile(self.conout_pipe, data, length, None, None)
OSError: exception: access violation writing 0x0000000000000000

SimonBiggs · 2018-01-25T03:59:12Z

As an extra step our IT administrator temporarily disabled my antivirus. This didn't stop the OSError from occuring.

dalito · 2018-01-25T08:43:32Z

To add to @SimonBiggs tests: Also when run as admin the error stays.

ccordoba12 · 2018-01-25T14:16:13Z

@dalito, so this is not fixed by your PR?

dalito · 2018-01-25T17:20:09Z

I got two different errors randomly: This one and the "OSError: [WinError 10048]" which is fixed by #87. After applying #87 there will only be this error for me.

andfoy · 2018-01-25T22:22:10Z

AFAIK, this is a problem related to the Python distribution itself. The strange fact is that it should be resolved

SimonBiggs · 2018-01-26T21:21:42Z

My process of installing python was to install a recent Miniconda and then run something like 'conda install python=3.5' to downgrade from 3.6 back to 3.5. (Some of my dependencies were struggling with 3.6 on conda for some reason) If you like I can uninstall Python and reinstall it with clear reproducible steps to see if it is then reproducible on other machines.

…

On Fri., 26 Jan. 2018, 9:22 am Edgar Andrés Margffoy Tuay, < ***@***.***> wrote: AFAIK, this is a problem related to the Python distribution itself. The strange fact is that it should be resolved — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGQVe8Bnsnx2mRmU9_0m5VBHsFB5hMvUks5tOP6TgaJpZM4RkVDG> .

ccordoba12 · 2018-01-27T16:52:43Z

@SimonBiggs, could you see what happens in Python 3.6?

SimonBiggs · 2018-01-28T01:42:22Z

I did make a python 3.6 environment `conda create --name py36 python=3.6` it still didn't work under that.

…

On Sun., 28 Jan. 2018, 3:52 am Carlos Cordoba, ***@***.***> wrote: @SimonBiggs <https://github.com/simonbiggs>, could you see what happens in Python 3.6? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AGQVe7BgNRdzTv7q91zvhils85JBPO94ks5tO1RcgaJpZM4RkVDG> .

ccordoba12 · 2018-02-06T20:19:09Z

@blink1073, what to do about this PR? Do you want to close or merge it?

blink1073 · 2018-02-06T20:21:53Z

Closing, not sure what else to try here.

blink1073 added 3 commits January 19, 2018 05:14

named pipe cleanup

bd0a13f

use wide string

1d9c742

variable name

7fc2b98

blink1073 closed this Jan 19, 2018

blink1073 reopened this Jan 19, 2018

formatting

ae27e2e

debug info

b313bbb

dalito mentioned this pull request Jan 20, 2018

OSError: [WinError 10048] on spawning of PtyProcess #86

Closed

blink1073 closed this Feb 6, 2018

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP] Named pipe cleanup #85

[WIP] Named pipe cleanup #85

blink1073 commented Jan 19, 2018 •

edited

Loading

blink1073 commented Jan 19, 2018

SimonBiggs commented Jan 19, 2018

SimonBiggs commented Jan 19, 2018

SimonBiggs commented Jan 19, 2018

dalito commented Jan 19, 2018

blink1073 commented Jan 20, 2018

blink1073 commented Jan 20, 2018

dalito commented Jan 20, 2018

blink1073 commented Jan 24, 2018

SimonBiggs commented Jan 25, 2018

SimonBiggs commented Jan 25, 2018

SimonBiggs commented Jan 25, 2018

dalito commented Jan 25, 2018

ccordoba12 commented Jan 25, 2018

dalito commented Jan 25, 2018

andfoy commented Jan 25, 2018

SimonBiggs commented Jan 26, 2018 via email

ccordoba12 commented Jan 27, 2018

SimonBiggs commented Jan 28, 2018 via email

ccordoba12 commented Feb 6, 2018

blink1073 commented Feb 6, 2018

[WIP] Named pipe cleanup #85

[WIP] Named pipe cleanup #85

Conversation

blink1073 commented Jan 19, 2018 • edited Loading

blink1073 commented Jan 19, 2018

SimonBiggs commented Jan 19, 2018

SimonBiggs commented Jan 19, 2018

SimonBiggs commented Jan 19, 2018

dalito commented Jan 19, 2018

blink1073 commented Jan 20, 2018

blink1073 commented Jan 20, 2018

dalito commented Jan 20, 2018

blink1073 commented Jan 24, 2018

SimonBiggs commented Jan 25, 2018

SimonBiggs commented Jan 25, 2018

SimonBiggs commented Jan 25, 2018

dalito commented Jan 25, 2018

ccordoba12 commented Jan 25, 2018

dalito commented Jan 25, 2018

andfoy commented Jan 25, 2018

SimonBiggs commented Jan 26, 2018 via email

ccordoba12 commented Jan 27, 2018

SimonBiggs commented Jan 28, 2018 via email

ccordoba12 commented Feb 6, 2018

blink1073 commented Feb 6, 2018

blink1073 commented Jan 19, 2018 •

edited

Loading