qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <philmd@redhat.com>
To: Cleber Rosa <crosa@redhat.com>,
	David Gibson <david@gibson.dropbear.id.au>
Cc: ldoktor@redhat.com, ehabkost@redhat.com, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] Cryptic errors from PIP install if missing openssl-devel
Date: Thu, 29 Aug 2019 11:24:30 +0200	[thread overview]
Message-ID: <24d0d5be-d206-33a0-cd8c-29825e2f8516@redhat.com> (raw)
In-Reply-To: <20190829032746.GA488@localhost.localdomain>

On 8/29/19 5:27 AM, Cleber Rosa wrote:
> On Thu, Aug 29, 2019 at 11:51:17AM +1000, David Gibson wrote:
>> On Thu, Aug 29, 2019 at 11:31:25AM +1000, David Gibson wrote:
>>> If I attempt to run "make check-acceptance" on my POWER9, RHEL8.1
>>> machine when the openssl-devel package isn't installed, I get the
>>> following very cryptic error:
>>>
>>>   VENV    /home/dwg/qemu/build/rhel8/tests/venv
>>>   PIP     /home/dwg/qemu/tests/requirements.txt
>>> Command "/home/dwg/qemu/build/rhel8/tests/venv/bin/python -u -c "import setuptools, tokenize;__file__='/tmp/pip-build-la4el5r5/cryptography/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" install --record /tmp/pip-1efs22iz-record/install-record.txt --single-version-externally-managed --compile --install-headers /home/dwg/qemu/build/rhel8/tests/venv/include/site/python3.6/cryptography" failed with error code 1 in /tmp/pip-build-la4el5r5/cryptography/
>>>
>>> Using V=1 doesn't give any more useful information, and it's not
>>> (easily) possible to manually re-run the given command since it relies
>>> on things in /tmp that are removed once the attempt finishes.
>>>
>>> I only figured out it was openssl-devel being missing that was the
>>> problem by (mostly) guesswork.  It would be really great if we could
>>> generate a more helpful error here.
>>>
>>> In addition, if I rerun "make check-acceptance" it no longer even
>>> attempts the PIP install, since tests/venv already exists in my build
>>> environment.  It then sort of works, but I think it might be hitting
>>> other errors because of the missing python packages.  Sorry that's a
>>> bit vague - I also seem to be getting unrelated errors that I'm still
>>> trying to figure out.
>>
>> Fwiw, I also get an equally cryptic error that I haven't figured out
>> the cause for on my 32-bit Fedora container environment:
>>
>>   VENV    /home/dwg/src/qemu/build/i386/tests/venv
>>   PIP     /home/dwg/src/qemu/tests/requirements.txt
>>   Failed building wheel for bcrypt
>> Could not build wheels for bcrypt which use PEP 517 and cannot be installed directly
>> You are using pip version 19.0.3, however version 19.2.3 is available.
>> You should consider upgrading via the 'pip install --upgrade pip' command.
>>
> 
> This is certainly caused by pip not being able to install paramiko on
> those systems.  I have dealt with paramiko (and its many dependencies)
> before on the avocado remote runner plugin (which is not being used
> here) and it was not fun.
> 
> My personal goal was to rely on the ssh binary as an ssh client, which
> should be more ubiquitous, and with that I added a simple wrapper to
> Avocado:
> 
>   https://avocado-framework.readthedocs.io/en/71.0/api/utils/avocado.utils.html#module-avocado.utils.ssh

I did not know this module.

> I guess we should consider changing the (few) tests that require
> paramiko to use that module instead.

Clean way to resolve this issue.

>> In this case the check definitely doesn't work - it doesn't appear to
>> have installed avocado in the venv.
>>
>>   AVOCADO tests/acceptance
>> /home/dwg/src/qemu/build/i386/tests/venv/bin/python: No module named avocado
>>
> 
> My hope is that with pure Python modules in requirements.txt, failures
> will be much harder to come by.  Either way, it'd be nice to improve

Is there a way to check for this before submitting/merging patches?

Some checkpatch.py lines that report:

  The 'paramiko' module is not pure Python and can
  not be added in requirements.txt.

> the venv creation, at the very least add a cleanup if it fails to
> complete successfully.
> 
> As a workaround I'd suggest two things:
> 
>  1) remove paramiko from requirements.txt
>  2) set the env var CONTINUOUS_INTEGRATION=1 before running `make
>     check-acceptace` (the tests in linux_ssh_mips_malta.py check for
>     that and don't run if it's set).
> 
> But, we'll need to address those failures definitely ASAP.  Anyway,
> thanks for reporting it.
> 
> - Cleber.
> 
>> -- 
>> David Gibson			| I'll have my music baroque, and my code
>> david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
>> 				| _way_ _around_!
>> http://www.ozlabs.org/~dgibson
> 
> 
> 


  reply	other threads:[~2019-08-29  9:25 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-29  1:31 [Qemu-devel] Cryptic errors from PIP install if missing openssl-devel David Gibson
2019-08-29  1:51 ` David Gibson
2019-08-29  3:27   ` Cleber Rosa
2019-08-29  9:24     ` Philippe Mathieu-Daudé [this message]
2019-08-29  9:46       ` Philippe Mathieu-Daudé
2019-09-03 15:08         ` Cleber Rosa
2019-09-03 15:22           ` Philippe Mathieu-Daudé
2019-09-03 15:47             ` Cleber Rosa
2019-09-03 15:56               ` Lukáš Doktor
2019-09-03 15:57                 ` Lukáš Doktor
2019-09-03 17:27           ` Eduardo Habkost
2019-09-03 22:33             ` Cleber Rosa
2019-09-03 15:18       ` Cleber Rosa
2019-08-30 17:56 ` Eduardo Habkost
2019-08-31  1:48   ` David Gibson
2019-09-03 15:31     ` Cleber Rosa
2019-09-04 19:57     ` Cleber Rosa
2019-09-05  2:09       ` David Gibson
2019-09-05 13:38         ` Cleber Rosa
2019-09-05 14:06           ` Eduardo Habkost
2019-09-05 15:24             ` Cleber Rosa
2019-09-06  0:34           ` David Gibson
2019-09-06 14:43             ` Eduardo Habkost
2019-09-06 14:52               ` Cleber Rosa
2019-09-06 15:14                 ` Eduardo Habkost
2019-09-06 15:29                   ` Cleber Rosa

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=24d0d5be-d206-33a0-cd8c-29825e2f8516@redhat.com \
    --to=philmd@redhat.com \
    --cc=crosa@redhat.com \
    --cc=david@gibson.dropbear.id.au \
    --cc=ehabkost@redhat.com \
    --cc=ldoktor@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).