One of my deployment pipelines on Azure DevOps fails when executing the build module images action of Azure IoT Edge task. I am trying to deploy a custom module developed using the Azure IoT SDK for C# (.NET 6).
Error message:
##[error]/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (1.26.12) or chardet (3.0.4) doesn't match a supported version!
I tried to include this solution as CmdLine task before the build task. It worked a few runs and then failed again.
The pipeline already had a "Temporary fix" installing iotedgedev separately as a workaround for this bug
The deployment logs don't tell much about where to focus. I wonder what could be the cause of this issue? and if there is a quick fix or something to avoid while developing the application (i.e. warning messages while building or something like that)
I've had the same problem. After checking previous success Pipelines it seems that then new iotedgedev version 3.3.5 has a problem in DevOps Build Task (I haven't investigate more)
To fix that, at least temporally, I have added a Command Line Script before Build Modules Images Task that installs version 3.3.4.
pip install -U iotedgedev==3.3.4
It has worked for me, but as I said is a temporally patch.
As this appears to happen again and again and again and again (in different flavours), I felt it made sense to explain why this happens. It's right there in the error message actually, in my case:
.local/lib/python3.11/site-packages/requests/__init__.py:102:
RequestsDependencyWarning: urllib3 (1.26.9) or
chardet (5.1.0)/charset_normalizer (2.0.12) doesn't match a supported version!
So, apparently the requests package needs its dependencies at a certain version level to work correctly. But pip doesn't show which versions:
$ pip show requests | grep Requires
Requires: certifi, charset-normalizer, idna, urllib3
Let's look at init.py then, this is where the error occured:
if chardet_version:
major, minor, patch = chardet_version.split(".")[:3]
major, minor, patch = int(major), int(minor), int(patch)
# chardet_version >= 3.0.2, < 5.0.0
assert (3, 0, 2) <= (major, minor, patch) < (5, 0, 0)
....and on it goes, and we can now fiddle around with the expected version strings until the warning goes away or update the requests package to a version that accepts its required packages at the correct versions.
TL;DR: One has to upgrade or downgrade the affected packages (request or its dependencies) to get it right, or fiddle around with requests/__init__.py until the warning goes away :-)
Short answer -> force pyOpenSSL to version 22.0.0
(Edited 09.01.20223)
pip install -U iotedgedev pyOpenSSL==22.0.0 urllib3 requests
The cause seems to be a reported bug related to a dependency issue connected to a new pyOpenSSL release: github.com/Azure/iotedgedev/issues/589
pyOpenSSL>=20.0.1 on iotedgedev requirements resolves pyOpenSSL-22.1.0 creating some conflict with urlib or chardet
In this case, a temporary fix needed a "quick" temporary fix :D
This is quite strange as pathlib seems already built in python > 3.4, anyone got an idea what raises the error?
ps. Package is already in pypi, the recipe was created directly from grayskull.
I'm using coc-tsserver on Neovim to program with Next.js, but the Auto Import feature is not working. When I try use a Next module, for example the <Link> tag, it recognizes the "import Link from 'next/link'" with no errors, but it's not able to find the module when the import line is missing. I already tried to config tsserver to use the external Eslint package, but it's still not working (even the yarn Eslint package is not able to find next modules when it's not specified).
How can I auto import Next.js modules on Eslint/coc-tsserver?
From the tsserver documentation
https://github.com/neoclide/coc-tsserver
you may need to add code below to your coc-config
"javascript.suggest.autoImports": true,
"typescript.suggest.autoImports": true
with :CocConfig
I am trying to get an application started on Openshift, using the Python cartridge 3.3. I've added my dependencies into the requirements.txt, but the application fail to load on pandas dependency. I've searched through the web and there's only one mention of issues related to pandas and numpy but cannot seem to help. I've attempted to install pandas via ssh and end up with the following message after a while
Installing collected packages: pandas
Running setup.py install for pandas
Connection to [app].rhcloud.com closed by remote host.
Connection to [app].rhcloud.com closed.
Has anyone already used pandas on Openshift? Is there potential issues due to C dependencies?
Thank you
I had the same problem. It seems that with OpenShift there's a problem compiling algos.cwhich is part of pandas.
By installing manually I was able to avoid OpenShift's closing of the connection:
See my answer to this similar question.
How can I install Gunicorn using Cygwin?
I cant seem to find the Gunicorn Package on Cygwin...Unless I'm understanding the issue on github wrong, I think it does support Gunicorn.
Also I found this in the changelog:
0.17.2 / 2013-01-07
optimize readline
make imports errors more visiblle when loading an app or a logging class
fix tornado worker: don’t pass ssl options if there are none
fix PEP3333: accept only bytetrings in the response body
fix support on CYGWIN platforms