Re: [PATCH] #3: Fix xm_internal / remote interaction

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jul 04, 2007 at 05:12:53AM +0100, Daniel P. Berrange wrote:
> On Tue, Jul 03, 2007 at 04:14:30PM +0100, Richard W.M. Jones wrote:
> > I'm not quite sure what the problem is (although the problem is in
> > xm_internal), but when you use xm_internal over remote, it sometimes
> > doesn't initialize its internal cache correctly, so it thinks that
> > there are no inactive domains.
> > 
> > The fix is a one-liner which I hit upon by accident -- I don't really
> > understand why it works:
> 
> Very peculiar - the  nconnections stuff is incremented / decremented
> by the xenXMOpen & xenXMClose methods. So the change you show below
> should be identical to previous behaviour. Is something calling the
> xenXMClose method too many times maybe ? I guess some judicious use
> of syslog would show it up

  Hum, I really dislike the following:

int xenXMClose(virConnectPtr conn ATTRIBUTE_UNUSED) {
    if (!nconnections--) {

I don't want to have to rely on C operator priorities to try to guess
if the code is correct.

    nconnections--;
    if (nconnections <= 0) { 

is way cleaner IMHO.

> > 
> > @@ -489,7 +487,7 @@
> >    xenXMOpen (virConnectPtr conn ATTRIBUTE_UNUSED,
> >               const char *name ATTRIBUTE_UNUSED, int flags 
> > ATTRIBUTE_UNUSED)
> >    {
> > -    if (nconnections == 0) {
> > +    if (configCache == NULL) {
> >            configCache = virHashCreate(50);
> >            if (!configCache)
> >                return (-1);

  IMHO that patch is wrong because it would not reset nconnections to 0
I would prefer to understand what is really going on. Could you check with
the cleanup of xenXMClose to see if this is sufficient ?

> > But the attached patch also adds proper error messages to
> > xenXMConfigCacheRefresh too.

  That is fine. There is still one thing I don't understand in the original
patch, the replacement of xenDaemonDomainLookupByName() by
xenDaemonLookupByName() which I don't see in the current code ?!?

Daniel

-- 
Red Hat Virtualization group http://redhat.com/virtualization/
Daniel Veillard      | virtualization library  http://libvirt.org/
veillard@xxxxxxxxxx  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine  http://rpmfind.net/

--
Libvir-list mailing list
Libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list

[Index of Archives]     [Virt Tools]     [Libvirt Users]     [Lib OS Info]     [Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [KDE Users]     [Fedora Tools]