Hey, On Wed, Mar 25, 2015 at 05:42:03PM +0100, Fabiano Fidêncio wrote: > Christophe, > > On Wed, Mar 25, 2015 at 4:47 PM, Christophe Fergeau <cfergeau@xxxxxxxxxx> wrote: > > On Wed, Mar 25, 2015 at 03:54:08PM +0100, Fabiano Fidêncio wrote: > >> When --direct fails the expected behavior is to report an error and > >> close the dialog, instead of the current behavior, that shows a dialog > >> and waits for the guest to connect. > > > > Is it specific to --direct though? It looks to me that we should never > > try to wait if virt_viewer_update_display() fails. I think we used to do > > that before commits 6100d08 / 9e2f9ea0 > > Hmmm. Interesting, at least :-) > I have tested the behavior before that commits and looks like it must > be specific to --direct. The other cases should wait for a VM. git master with/without your patch still shows the problematic behaviour with qemu+tcp:// and a remote host listening on 127.0.0.1 so it's not specific to --direct. From a quick glance at extract_connect_info() and update_display(), the "return FALSE" situations I saw are situations when we won't be able to connect regardless of what changes on the remote host, so I don't think --direct should be special-cased there. Christophe
Attachment:
pgpJwZDRtFgSl.pgp
Description: PGP signature
_______________________________________________ virt-tools-list mailing list virt-tools-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/virt-tools-list
