[Pkg-libvirt-maintainers] Bug#564034: Bug#564034: Bug#564034: Bug#564034: netcat doesn't exit and virt-manager hangs

Guido Günther agx at sigxcpu.org
Thu Jan 7 11:57:01 UTC 2010


On Thu, Jan 07, 2010 at 12:42:55PM +0100, Václav Ovsík wrote:
> On Thu, Jan 07, 2010 at 11:19:19AM +0100, Guido Günther wrote:
> > We need a similar patch for libvirt - however it needs to be disto
> > specific since different netcat versions have different needs -
> > unforuntely.
> 
> I'm not certain. Consider client running Fedora (without the patch)
> connecting to server running Debian. I think it should be upstream if it
> not disrupts other systems.
> 
> Citation from https://bugzilla.redhat.com/show_bug.cgi?id=517778
> 
> = Comment #7 From  Daniel Berrange  2009-08-19 11:06:04 EDT  -------
> = 
> = virt-manager runs netcat on the remote end of the SSH connection. In theory
> = closing the FD should cause netcat to see the EOF, and exit, causing SSH to
> = exit. I reckon netcat is not behaving nicely though and thus holding open the
> = connection. The Debian netcat has certainly got such bugs, and Fedora one has
> = patched many bugs like that. We should explicitly kill() the SSH pid.  
> 
> This is probably about how should netcat behave and which implementations
> (patches) are correct. I don't know.
> I think adding `-q 0' for nc execution upstream should not hurt in either case.
> Cheers
We had this in virt-manager an Cole dropped it since Fedora's nc didn't
support -q0.
 -- Guido





More information about the Pkg-libvirt-maintainers mailing list