All of lore.kernel.org
 help / color / mirror / Atom feed
From: Markus Armbruster <armbru@redhat.com>
To: Amador Pahim <apahim@redhat.com>
Cc: "Kevin Wolf" <kwolf@redhat.com>,
	"Lukáš Doktor" <ldoktor@redhat.com>,
	"Fam Zheng" <famz@redhat.com>,
	"Eduardo Habkost" <ehabkost@redhat.com>,
	"Stefan Hajnoczi" <stefanha@gmail.com>,
	qemu-devel@nongnu.org, "Max Reitz" <mreitz@redhat.com>,
	"Cleber Rosa" <crosa@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v6 7/7] qemu.py: include debug information on launch error
Date: Fri, 18 Aug 2017 16:01:38 +0200	[thread overview]
Message-ID: <87k2210ybh.fsf@dusky.pond.sub.org> (raw)
In-Reply-To: <CALAZnb1naw=TtrsDaBw3ehESCp7ET42DUuc=cSTQTXpsKsvoyw@mail.gmail.com> (Amador Pahim's message of "Fri, 18 Aug 2017 14:42:39 +0200")

Amador Pahim <apahim@redhat.com> writes:

> On Tue, Aug 15, 2017 at 10:57 AM, Markus Armbruster <armbru@redhat.com> wrote:
>> Amador Pahim <apahim@redhat.com> writes:
>>
>>> When launching a VM, if an exception happens and the VM is not
>>> initiated, it is useful to see the qemu command line that was executed
>>> and the output of that command.
>>>
>>> Before this patch:
>>>
>>>     >>> import qemu
>>>     >>> vm = qemu.QEMUMachine('qemu-system-aarch64', debug=True)
>>>     >>> vm.launch()
>>>     Traceback (most recent call last):
>>>       File "<stdin>", line 1, in <module>
>>>       File "qemu.py", line 175, in launch
>>>         self._launch()
>>>       File "qemu.py", line 189, in _launch
>>>         self._post_launch()
>>>       File "qemu.py", line 154, in _post_launch
>>>         self._qmp.accept()
>>>       File "qmp/qmp.py", line 145, in accept
>>>         self.__sock, _ = self.__sock.accept()
>>>       File "/usr/lib64/python2.7/socket.py", line 206, in accept
>>>         sock, addr = self._sock.accept()
>>>     socket.timeout: timed out
>>>     >>>
>>>
>>> After this patch:
>>>
>>>     >>> import qemu
>>>     >>> vm = qemu.QEMUMachine('qemu-system-aarch64', debug=True)
>>>     >>> vm.launch()
>>>     DEBUG:qemu:Error launching VM. Command: 'qemu-system-aarch64
>>>     -chardev socket,id=mon,path=/var/tmp/qemu-5298-monitor.sock
>>>     -mon chardev=mon,mode=control -display none -vga none'.
>>>     Output: 'qemu-system-aarch64: No machine specified, and there
>>>     is no default\nUse -machine help to list supported machines\n'.
>>>     Traceback (most recent call last):
>>>       File "<stdin>", line 1, in <module>
>>>       File "qemu.py", line 175, in launch
>>>         self._launch()
>>>       File "qemu.py", line 189, in _launch
>>>         self._post_launch()
>>>       File "qemu.py", line 154, in _post_launch
>>>         self._qmp.accept()
>>>       File "qmp/qmp.py", line 145, in accept
>>>         self.__sock, _ = self.__sock.accept()
>>>       File "/usr/lib64/python2.7/socket.py", line 206, in accept
>>>         sock, addr = self._sock.accept()
>>>     socket.timeout: timed out
>>>     >>>
>>>
>>> Signed-off-by: Amador Pahim <apahim@redhat.com>
>>> ---
>>>  scripts/qemu.py | 12 +++++++++++-
>>>  1 file changed, 11 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/scripts/qemu.py b/scripts/qemu.py
>>> index e9a3a96d13..43fd0b072c 100644
>>> --- a/scripts/qemu.py
>>> +++ b/scripts/qemu.py
>>> @@ -158,7 +158,10 @@ class QEMUMachine(object):
>>>              self._remove_if_exists(self._created_files.pop())
>>>
>>>      def launch(self):
>>> -        '''Launch the VM and establish a QMP connection'''
>>> +        '''
>>> +        Try to launch the VM and make sure we cleanup and expose the
>>> +        command line/output in case of exception.
>>> +        '''
>>>
>>>          if self.is_running():
>>>              raise QEMULaunchError('VM already running.')
>>> @@ -169,6 +172,7 @@ class QEMUMachine(object):
>>>                                    'before launching again.')
>>>
>>>          try:
>>> +            self._iolog = None
>>>              self._qemu_full_args = None
>>>              self._qemu_full_args = (self._wrapper + [self._binary] +
>>>                                      self._base_args() + self._args)
>>> @@ -176,9 +180,15 @@ class QEMUMachine(object):
>>>              self._pending_shutdown = True
>>>          except:
>>>              self.shutdown()
>>> +            LOG.debug('Error launching VM.%s%s',
>>> +                      ' Command: %r.' % ' '.join(self._qemu_full_args)
>>> +                      if self._qemu_full_args else '',
>>> +                      ' Output: %r.' % self._iolog
>>> +                      if self._iolog else '')
>>
>> This nested interpolation is too ugly to live :)
>
> I find it beautiful :)
> Killing it anyway.
>
>>
>> Nesting is easy to avoid:
>>
>>                LOG.debug('Error launching VM\n. ' Command: %r. Output: %r.'
>>                          % (' '.join(self._qemu_full_args)
>>                             if self._qemu_full_args else '',
>>                             self._iolog
>>                             if self._iolog else ''))
>
> In this case we will have the strings "Command: " and "Output:" even
> with no content to expose.

No command should not happen.

No output would be bad news.  Making that explicit in the logs feels
right to me.

>> Why %r and not %s?
>
> It's a matter of preference. I tend to avoid allowing messages to
> break the log line. Anyway, I'm improving this.

That's a really good point.  I'd go as far and call %s wrong now :)

Output could be arbitrarily long, though.

>> Are you sure '\n' is appropriate in the argument of LOG.debug()?
>>
>> Why cram everything into a single LOG.debug()?
>>
>>                LOG.debug('Error launching VM')
>>                LOG.debug('Command: %s' % (' '.join(self._qemu_full_args)
>>                                           if self._qemu_full_args else ''))
>>                LOG.debug('Output: %s' % self._iolog if self._iolog else ''))
>
> Sure, I'm ok with both ways.
>
>>
>>>              raise
>>>
>>>      def _launch(self):
>>> +        '''Launch the VM and establish a QMP connection.'''
>>>          self._pre_launch()
>>>          devnull = open(os.path.devnull, 'rb')
>>>          self._popen = subprocess.Popen(self._qemu_full_args,

  reply	other threads:[~2017-08-18 14:01 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-31  8:51 [Qemu-devel] [PATCH v6 0/7] scripts/qemu.py fixes and cleanups Amador Pahim
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 1/7] qemu.py: use poll() instead of 'returncode' Amador Pahim
2017-07-31 14:07   ` Eduardo Habkost
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 2/7] qemu.py: fix is_running() return before first launch() Amador Pahim
2017-08-01 10:09   ` Stefan Hajnoczi
2017-08-01 10:50     ` Eduardo Habkost
2017-08-01 12:56       ` Amador Pahim
2017-08-02 13:35         ` Stefan Hajnoczi
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 3/7] qemu.py: use python logging system Amador Pahim
2017-07-31 14:50   ` Eduardo Habkost
2017-07-31 15:05     ` Amador Pahim
2017-08-15  8:10   ` Markus Armbruster
2017-08-15 19:58     ` Eduardo Habkost
2017-08-16  6:17       ` Markus Armbruster
2017-08-18 12:24         ` Amador Pahim
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 4/7] qemu.py: improve message on negative exit code Amador Pahim
2017-08-15  8:26   ` Markus Armbruster
2017-08-18 12:24     ` Amador Pahim
2017-08-18 13:57       ` Markus Armbruster
2017-08-18 14:33         ` Amador Pahim
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 5/7] qemu.py: use os.path.null instead of /dev/null Amador Pahim
2017-08-01 10:11   ` Stefan Hajnoczi
2017-08-22  2:24   ` Philippe Mathieu-Daudé
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 6/7] qemu.py: cleanup and improve launch()/shutdown() Amador Pahim
2017-08-15  8:45   ` Markus Armbruster
2017-07-31  8:51 ` [Qemu-devel] [PATCH v6 7/7] qemu.py: include debug information on launch error Amador Pahim
2017-08-15  8:57   ` Markus Armbruster
2017-08-18 12:42     ` Amador Pahim
2017-08-18 14:01       ` Markus Armbruster [this message]
2017-08-18 14:49         ` Amador Pahim
2017-08-21  9:14           ` Markus Armbruster

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=87k2210ybh.fsf@dusky.pond.sub.org \
    --to=armbru@redhat.com \
    --cc=apahim@redhat.com \
    --cc=crosa@redhat.com \
    --cc=ehabkost@redhat.com \
    --cc=famz@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=ldoktor@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@gmail.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.