[Kimchi-devel] [PATCH] [Kimchi 1/1] Fix issue #1170: ERROR:KCHVM0090E: Unable to create a password-less libvirt connection to the remote libvirt daemon

Aline Manera alinefm at linux.vnet.ibm.com
Tue Nov 14 13:04:24 UTC 2017


I am not being able to apply it due the following error:

[alinefm at alinefm-TP440 kimchi]$ git am -3 
/home/alinefm/mail-patches/\[Kimchi-devel\]\ \[PATCH\]\ \[Kimchi\ 1_1\]\ 
Fix\ issue\ #1170\:\ ERROR\:KCHVM0090E\:\ Unable\ to\ create\ a\ 
password-less\ libvirt\ connection\ to\ the\ remote\ libvirt\ daemon.eml
Applying: Fix issue #1170: ERROR:KCHVM0090E: Unable to create a 
password-less libvirt connection to the remote libvirt daemon
error: corrupt patch at line 11
error: could not build fake ancestor
Patch failed at 0001 Fix issue #1170: ERROR:KCHVM0090E: Unable to create 
a password-less libvirt connection to the remote libvirt daemon
The copy of the patch that failed is found in: 
/home/alinefm/wok/.git/modules/src/wok/plugins/kimchi/rebase-apply/patch
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.
To restore the original branch and stop patching, run "git am --abort".
[alinefm at alinefm-TP440 kimchi]$ vi 
/home/alinefm/mail-patches/\[Kimchi-devel\]\ \[PATCH\]\ \[Kimchi\ 1_1\]\ 
Fix\ issue\ #1170\:\ ERROR\:KCHVM0090E\:\ Unable\ to\ create\ a\ 
password-less\ libvirt\ connection\ to\ the\ remote\ libvirt\ daemon.eml
[alinefm at alinefm-TP440 kimchi]$ git am --abort


Seems the patch gets corrupted in some way.

Also, looking again to the patch, maybe it is simpler to just remove the 
sdtout parameter from Popen.

On 10/31/2017 04:06 AM, Fu.Lin wrote:
>
> Long ago, I submitted an issue: 
> https://github.com/kimchi-project/kimchi/issues/1170 ...
>
> Test by following script:
>
>     import subprocess
>     import os
>     import time
>
>     def _check_remote_libvirt_conn(remote_host,
>             user='root', transport='ssh'):
>
>         FNULL = open(os.devnull, 'w')
>         dest_uri = 'qemu+%s://%s@%s/system' % (transport, user,
>     remote_host)
>         cmd = ['virsh', '-c', dest_uri, 'list']
>         proc = subprocess.Popen(cmd, stdout=FNULL, stderr=FNULL,
>                 shell=True, preexec_fn=os.setsid)
>         timeout = 0
>         while proc.poll() is None:
>             time.sleep(1)
>             timeout += 1
>             if timeout == 5:
>                 print("killed")
>                 os.killpg(os.getpgid(proc.pid), signal.SIGTERM)
>
>     if __name__ == '__main__':
>         _check_remote_libvirt_conn('localhost')
>
> `proc` is timeout every time although the connection is normal tested 
> by `virsh -c <uri>` list
>
>
> The following is the reason of changing(From Python Document 2.7.14):
>
>  Do not use stdout=PIPE or stderr=PIPE with this function as that can 
> deadlock based on the child process output volume. Use Popen with the 
> communicate() method when you need pipes.
>
>  Using shell=True can be a security hazard. See the warning under 
> Frequently Used Arguments for details.
>
> At the same time, if not change, it does not work.
>
>
> diff --git a/model/vms.py b/model/vms.py
> index 6da4f3b..92e771f 100644
> --- a/model/vms.py
> +++ b/model/vms.py
> @@ -1885,10 +1885,11 @@ class VMModel(object):
>      def _check_remote_libvirt_conn(self, remote_host,
>                                     user='root', transport='ssh'):
>
> +        FNULL = open(os.devnull, 'w')
>          dest_uri = 'qemu+%s://%s@%s/system' % (transport, user, 
> remote_host)
>          cmd = ['virsh', '-c', dest_uri, 'list']
> -        proc = subprocess.Popen(cmd, stdout=subprocess.PIPE,
> -                                shell=True, preexec_fn=os.setsid)
> +        proc = subprocess.Popen(cmd, stdout=FNULL, stderr=FNULL,
> +                                shell=False, preexec_fn=os.setsid)
>          timeout = 0
>          while proc.poll() is None:
>              time.sleep(1)
>
> -- 
> Regards
> River (Fu.Lin)
>
>
> _______________________________________________
> Kimchi-devel mailing list
> Kimchi-devel at ovirt.org
> http://lists.ovirt.org/mailman/listinfo/kimchi-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/kimchi-devel/attachments/20171114/996dcc62/attachment.html>


More information about the Kimchi-devel mailing list