Refactor communication infra

Adam Litke agl at us.ibm.com
Tue Jan 15 14:15:41 UTC 2013


On Tue, Jan 15, 2013 at 03:40:17PM +0200, Dan Kenigsberg wrote:
> On Mon, Jan 07, 2013 at 03:10:38PM -0500, danken at redhat.com wrote:
> > Dan Kenigsberg has posted comments on this change.
> > 
> > Change subject: Refactor communication infra
> > ......................................................................
> > 
> > 
> > Patch Set 15: (2 inline comments)
> > 
> > ....................................................
> > File vdsm_api/BindingJsonRpc.py
> > Line 37:         self._reactors = reactors
> > Line 38: 
> > Line 39:     def _createTcpReactor(self, cfg):
> > Line 40:         address = cfg.get("ip", "0.0.0.0")
> > Line 41:         port = cfg.get("port", 4044)
> > sorry, I do not follow. which error handling code would have been required? to handle the case of not having cfg['ip'] ? having a default here only hide such an error instead of crying out loud about it.
> 
> Saggi, Adam, could you help me here?
> 
> I do not understand the need for the default port here.
> Could you elaborate on the justification?

I'll give Saggi a chance to defend it.  On the face of it I'd agree with you Dan
that the initialization code in clientIF should be structured similarly to the
BindingXMLRPC code.  Configuration for all ports (regardless of how the
underlying code is structured) should probably be in a flat namespace and
together in the vdsm.conf file.

-- 
Adam Litke <agl at us.ibm.com>
IBM Linux Technology Center



More information about the vdsm-patches mailing list