We're actually currently using the VDE peripheral programs (including
vdeq) for the vswitch and just replacing the core. However, this is a
good incentive to fix/replace vqeq as well.
I can come up with a patch pretty quickly to prevent vdeq from dying on
error. I don't have a repro of the problem though.
Jesse
Martin Casado wrote:
This will all go away when we release VDE compatible vswitch ... :)
Brandon Heller <[email protected]> writes:
Before installing the latest VDE from Subversion source, does
anyone have any suggestions or notes about bugs that have come
up?
I've occasionally seen this problem too, but for me it is
somewhat difficult to reproduce: when I see it, I just rerun my
test and everything generally works that time.
I think that this is really a bug in the vdeq program. At least,
it's involved: vdeq has a loop that reads from one socket and
writes to another, and any error return causes it to exit, which
in turn kills the VM that was using it, which in turn causes the
monitor script (if I recall correctly) to kill off all the other
VMs.
Fixes welcomed.
_______________________________________________
nox-dev mailing list
[email protected]
http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
_______________________________________________
nox-dev mailing list
[email protected]
http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
