The patch pair seems reasonable to me.
--Justin
On Jun 17, 2011, at 3:02 PM, Ben Pfaff wrote:
> On Fri, Jun 17, 2011 at 01:52:07PM -0700, Ben Pfaff wrote:
>> This way, the xenserver spec file and the upcoming RHEL 5.6 spec file don't
>> have to install it by hand.
>>
>> Signed-off-by: Ben Pfaf
Looks good.
--Justin
On Jun 17, 2011, at 1:52 PM, Ben Pfaff wrote:
> Reported-by: Andrew Evans
> ---
> debian/openvswitch-switch.init |2 +-
> xenserver/etc_init.d_openvswitch |4 ++--
> 2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/debian/openvswitch-switch.init b
On Fri, Jun 17, 2011 at 01:52:07PM -0700, Ben Pfaff wrote:
> This way, the xenserver spec file and the upcoming RHEL 5.6 spec file don't
> have to install it by hand.
>
> Signed-off-by: Ben Pfaff
This also needed the following change, which I've folded in:
diff --git a/Makefile.am b/Makefile.am
This way, the xenserver spec file and the upcoming RHEL 5.6 spec file don't
have to install it by hand.
Signed-off-by: Ben Pfaff
---
vswitchd/automake.mk |1 +
xenserver/openvswitch-xen.spec |2 --
2 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/vswitchd/automake
Reported-by: Andrew Evans
---
debian/openvswitch-switch.init |2 +-
xenserver/etc_init.d_openvswitch |4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/debian/openvswitch-switch.init b/debian/openvswitch-switch.init
index 44ec67c..9c7343b 100755
--- a/debian/openvsw
On Fri, Jun 17, 2011 at 01:29:10PM -0700, Andrew Evans wrote:
> Thanks, this looks good. I have only a few minor comments:
>
> On Thu, 2011-06-16 at 12:25 -0700, Ben Pfaff wrote:
> > diff --git a/INSTALL.RHEL-5.6 b/INSTALL.RHEL-5.6
>
> Should we change the RHEL 5.6 references to RHEL 5? RHEL 5.7
On Fri, 2011-06-17 at 13:37 -0700, Ben Pfaff wrote:
> On Fri, Jun 17, 2011 at 01:34:55PM -0700, Andrew Evans wrote:
> > Looks good. Is there no way to remove the vifs from the database before
> > they disappear?
>
> I don't know a way. By the time the "vif" script starts, they're
> already gone.
On Fri, Jun 17, 2011 at 01:34:55PM -0700, Andrew Evans wrote:
> On Thu, 2011-06-16 at 14:05 -0700, Ben Pfaff wrote:
> > Bug #5844 reports several log messages of the form:
> >
> > netdev_linux|ERR|ioctl(SIOCGIFHWADDR) on vif426.1 device failed: No
> > such device
> >
> > during migrations
On Thu, 2011-06-16 at 14:05 -0700, Ben Pfaff wrote:
> Bug #5844 reports several log messages of the form:
>
> netdev_linux|ERR|ioctl(SIOCGIFHWADDR) on vif426.1 device failed: No
> such device
>
> during migrations. These are normal and unavoidable, because the vifs
> disappear from the k
On Thu, 2011-06-16 at 14:05 -0700, Ben Pfaff wrote:
> get_etheraddr() in netdev-linux.c logs when the Ethernet address cannot be
> obtained so there is no need to log again in the caller.
Looks good to me, thanks.
___
dev mailing list
dev@openvswitch.o
Thanks, this looks good. I have only a few minor comments:
On Thu, 2011-06-16 at 12:25 -0700, Ben Pfaff wrote:
> diff --git a/INSTALL.RHEL-5.6 b/INSTALL.RHEL-5.6
Should we change the RHEL 5.6 references to RHEL 5? RHEL 5.7 is
currently in beta, and there will probably be future RHEL 5 minor
relea
Cool, i'll go ahead and merge.
Thanks for the reviews.
Ethan
On Fri, Jun 17, 2011 at 11:47, Ben Pfaff wrote:
> On Fri, Jun 17, 2011 at 11:33:33AM -0700, Ethan Jackson wrote:
>> > I think that update_iface() will now always call into ovs-vsctl to set
>> > the iface-status. ?I think that it shoul
On Fri, Jun 17, 2011 at 12:46:40PM -0700, Andrew Evans wrote:
> On Thu, 2011-06-16 at 12:25 -0700, Ben Pfaff wrote:
> > This way, the xenserver spec file and the upcoming RHEL 5.6 spec file don't
> > have to install it by hand.
>
> Looks good to me.
Thanks, pushed.
___
On Fri, Jun 17, 2011 at 12:29:30PM -0700, Andrew Evans wrote:
> On Fri, 2011-06-17 at 12:22 -0700, Ben Pfaff wrote:
> > Here's the diff that I applied. I'll test it again and push it.
>
> Thanks, this looks good to me.
Thanks, I pushed it.
___
dev mail
On Thu, 2011-06-16 at 12:25 -0700, Ben Pfaff wrote:
> This way, the xenserver spec file and the upcoming RHEL 5.6 spec file don't
> have to install it by hand.
Looks good to me.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/li
On Fri, 2011-06-17 at 12:22 -0700, Ben Pfaff wrote:
> Here's the diff that I applied. I'll test it again and push it.
Thanks, this looks good to me.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Fri, 2011-06-17 at 09:21 -0700, Ben Pfaff wrote:
> Other than the issues that Reid pointed out, this looks OK to me.
Thanks. I switched the order of the 'verbose' and 'flow' arguments to
ofp_fatal() to disambiguate the signature from that of ovs_fatal() and
fixed the errors that Reid pointed ou
On Thu, Jun 16, 2011 at 03:35:01PM -0700, Andrew Evans wrote:
> [typos]
Thanks, I fixed all those.
> > +case $0 in
> > +*/*) dir0=`echo "$0" | sed 's,/[^/]*$,,'` ;;
> > +*) dir0=./ ;;
> > +esac
>
> Is there some reason not to use 'dirname'?
The Autoconf manual says:
`dirname'
On Fri, Jun 17, 2011 at 11:33:33AM -0700, Ethan Jackson wrote:
> > I think that update_iface() will now always call into ovs-vsctl to set
> > the iface-status. ?I think that it should only do it if it needs to
> > change, as we do for iface-id.
>
> update_iface() is only called on a given interfac
OK.
We should rewrite this code. It's really hard to understand.
On Fri, Jun 17, 2011 at 11:27:46AM -0700, Ethan Jackson wrote:
> This code is so convoluted it's fairly difficult to understand.
>
> The get_iface_id() function is used to figure out what the iface-id
> should be for a given inter
On Fri, 2011-06-17 at 02:53 -0700, Reid Price wrote:
> +ovs_fatal(verbose, str_, "only %d
> registers supported", FLOW_N_REGS);
> Is this on purpose?
No, thanks for pointing that out.
> +ovs_fatal(verbose, str_, "unknown keyword %
>
> I think that update_iface() will now always call into ovs-vsctl to set
> the iface-status. I think that it should only do it if it needs to
> change, as we do for iface-id.
update_iface() is only called on a given interface when there is a
change in it's xs-vif-uuid, iface-id, or iface-status.
This code is so convoluted it's fairly difficult to understand.
The get_iface_id() function is used to figure out what the iface-id
should be for a given interface based on it's xs_vif_uuid. Most
interfaces don't have xs_vif_uuids, and therefore won't get iface-ids.
This patch gives tap devices
Thanks, I pushed this.
On Fri, Jun 17, 2011 at 10:46:29AM -0700, Justin Pettit wrote:
> I'm not familiar enough with OpenSSL to really understand what's
> going on, but I don't see anything obviously wrong with this.
>
> --Justin
>
>
> On Jun 15, 2011, at 11:50 AM, Ben Pfaff wrote:
>
> > If th
I'm not familiar enough with OpenSSL to really understand what's going on, but
I don't see anything obviously wrong with this.
--Justin
On Jun 15, 2011, at 11:50 AM, Ben Pfaff wrote:
> If the CA certificate changed and OVS added the new CA certificate, the
> change was ineffective. Clearing t
On Fri, Jun 17, 2011 at 10:31:49AM -0700, Ben Pfaff wrote:
> On Thu, Jun 16, 2011 at 11:13:24PM -0700, David Tsai wrote:
> > A pool configured for secure fail-mode can block dom0 traffic on hosts
> > joining
> > the pool or if the host reboots while the controller is unavailable. This
> > commit
On Thu, Jun 16, 2011 at 11:13:24PM -0700, David Tsai wrote:
> A pool configured for secure fail-mode can block dom0 traffic on hosts joining
> the pool or if the host reboots while the controller is unavailable. This
> commit sets default flows on a host under these conditions to allow management
On Thu, Jun 16, 2011 at 05:15:15PM -0700, Ethan Jackson wrote:
> The iface-status external id indicates to a controller which device
> it should manage when there are multiple choices for a given vif.
> Currently, it always chooses a tap device if available, but one
> could imagine more sophisticat
On Thu, Jun 16, 2011 at 05:15:14PM -0700, Ethan Jackson wrote:
> In some cases XenServer will give a virtual machine a tap device in
> addition to its usual vif. These tap devices need iface-ids so
> that controllers can figure out which vif they are related to.
>
> Signed-off-by: Ethan Jackson
On Thu, Jun 16, 2011 at 05:15:13PM -0700, Ethan Jackson wrote:
> Found with pylint.
>
> E:167:update_in_band_mgmt: Undefined variable 'bridge'
>
> Signed-off-by: Ethan Jackson
Looks good, thanks.
___
dev mailing list
dev@openvswitch.org
http://openvsw
On Thu, Jun 16, 2011 at 05:08:07PM -0700, Andrew Evans wrote:
> On Thu, 2011-06-16 at 14:12 -0700, Ben Pfaff wrote:
> > On Wed, Jun 15, 2011 at 06:55:38PM -0700, Andrew Evans wrote:
> > > When an error is encountered while parsing flows from a file, ovs-ofctl
> > > doesn't
> > > print the flow it
On Fri, Jun 17, 2011 at 02:53:58AM -0700, Reid Price wrote:
> > +if (verbose) {
> > +fprintf(stderr, "%s:\n", str_);
> > +}
> > +ovs_fatal(verbose, str_, "only %d registers
> > supported", FLOW_N_REGS);
> >
> Is thi
Some notes in passing, please forgive any ignorance
On Thu, Jun 16, 2011 at 5:08 PM, Andrew Evans wrote:
> On Thu, 2011-06-16 at 14:12 -0700, Ben Pfaff wrote:
> > On Wed, Jun 15, 2011 at 06:55:38PM -0700, Andrew Evans wrote:
> > > When an error is encountered while parsing flows from a file, ovs
33 matches
Mail list logo
