On 02/18/11 21:25, Ben Pfaff wrote:
So the whole thing looks good then.
Thanks very much for finishing this off on a Friday night! I pushed this.
-Andrew
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev_openvswitc
On Fri, Feb 18, 2011 at 9:19 PM, Andrew Evans wrote:
> I'm mimicking RPM's convention for expressing shared library dependencies in
> terms of the library names rather than the packages in which they reside.
> Obviously that doesn't translate literally to the kernel, but I thought it
> would be re
On 02/18/11 20:29, Ben Pfaff wrote:
It looks good. The only question in my mind is whether "openvswitch_mod.ko.0"
is a name that matches whatever conventions are used for RPM names. It
wouldn't be considered suitable as a Debian package name, because Debian
package names are not allowed to cont
I pushed this, thanks.
On Fri, Feb 18, 2011 at 5:53 PM, Justin Pettit wrote:
> Looks good to me. Thanks for tracking this down!
>
> Acked-by: Justin Pettit
>
> --Justin
>
>
> On Feb 18, 2011, at 4:10 PM, Ben Pfaff wrote:
>
>> The ovs_skb_cb in 'packet' in this function is initially a clone of t
On Fri, Feb 18, 2011 at 5:27 PM, Andrew Evans wrote:
> I also realized that the 'depmod' command needs to be run in the kernel
> package now, strictly speaking, so I've moved that.
>
> I also forgot to update INSTALL.XenServer in the last commit, so I've
> done that.
Thanks for catching those two
from paulc's yahoo.com
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev_openvswitch.org
Looks good to me. Thanks for tracking this down!
Acked-by: Justin Pettit
--Justin
On Feb 18, 2011, at 4:10 PM, Ben Pfaff wrote:
> The ovs_skb_cb in 'packet' in this function is initially a clone of the
> corresponding area in 'skb', which came from the Netlink layer and thus
> isn't necessar
On 2/18/11 2:58 PM, Andrew Evans wrote:
> On 2/18/11 11:18 AM, Ben Pfaff wrote:
>> I think that the userspace package could (and should) require the
>> kernel package to be installed by putting an appropriate "Provides:"
>> into the kernel package name. No?
>
> I was thinking along the same lines
The ovs_skb_cb in 'packet' in this function is initially a clone of the
corresponding area in 'skb', which came from the Netlink layer and thus
isn't necessarily all-zeros. This commit initializes it properly before
passing it along to execute_actions().
The most common problem caused by failing
On 2/18/11 11:18 AM, Ben Pfaff wrote:
> I think that the userspace package could (and should) require the
> kernel package to be installed by putting an appropriate "Provides:"
> into the kernel package name. No?
I was thinking along the same lines. We could embed a version number
that we bump wh
On 2/18/11 2:42 PM, Ben Pfaff wrote:
> On Fri, Feb 18, 2011 at 02:27:41PM -0800, Andrew Evans wrote:
>> if [ -h "$f" -a ! -e "$f" ]; then
>
> It's best to avoid -a and -o. Please use && instead.
Ok.
Thanks for your review. I pushed this.
-Andrew
___
On Fri, Feb 18, 2011 at 02:27:41PM -0800, Andrew Evans wrote:
> On 2/18/11 10:15 AM, Ben Pfaff wrote:
> > On Thu, Feb 17, 2011 at 05:52:07PM -0800, Andrew Evans wrote:
> > In references to commit numbers, please also give the subject of the
> > commit (the first line of the log message).
>
> Ok, n
On 2/18/11 10:15 AM, Ben Pfaff wrote:
> On Thu, Feb 17, 2011 at 05:52:07PM -0800, Andrew Evans wrote:
> In references to commit numbers, please also give the subject of the
> commit (the first line of the log message).
Ok, now it reads like this:
Commit d66880ee (xenserver: Clean up /usr/sbin/brc
Thanks, merged.
On Fri, Feb 18, 2011 at 11:18 AM, Ben Pfaff wrote:
> On Fri, Feb 18, 2011 at 11:15:10AM -0800, Ethan Jackson wrote:
>> > I have one question, though. ?It looks like flow_push_stats() only
>> > affects the rules that get resubmitted into, not the initial rule. ?Is
>> > that correct
On 2/18/11 11:04 AM, Ben Pfaff wrote:
> The following comment in bridge.c could use an update too:
>
> /* Collect all of the potential targets, as the union of the "managers"
> * column and the "targets" columns of the rows pointed to by
> * "manager_options", excluding any that are
On Fri, Feb 18, 2011 at 11:15:10AM -0800, Ethan Jackson wrote:
> > I have one question, though. ?It looks like flow_push_stats() only
> > affects the rules that get resubmitted into, not the initial rule. ?Is
> > that correct? ?Is there something else that adds stats to the initial
> > rule? ?I mig
On Wed, Feb 16, 2011 at 10:57:46PM -0800, Andrew Evans wrote:
> XenServer has the concept of a "supplemental pack", which is just a collection
> of RPMs rolled into an ISO image for installation as a unit, either during
> XenServer installation or later as an update.
>
> The script used to build s
> I have one question, though. It looks like flow_push_stats() only
> affects the rules that get resubmitted into, not the initial rule. Is
> that correct? Is there something else that adds stats to the initial
> rule? I might just be missing it in the diff; I didn't try to apply
> this and loo
On Tue, Feb 15, 2011 at 02:18:48PM -0800, Ethan Jackson wrote:
> This patch causes statistics to be updated for rules which are
> resubmitted into. Once per second statistics are queried from the
> datapath and pushed along the resubmit graph (calculated on demand
> from the action list). This app
On Fri, Feb 18, 2011 at 10:23:43AM -0800, Andrew Evans wrote:
> We had retained but deprecated the use of the older 'managers' column in the
> 'Open vSwitch' table for compatibility with applications that might still use
> it, but that created more problems than it solved. This commit removes the
>
Erm, that should be 'managers' in the Subject: line. Fixed in my branch.
-Andrew
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev_openvswitch.org
We had retained but deprecated the use of the older 'managers' column in the
'Open vSwitch' table for compatibility with applications that might still use
it, but that created more problems than it solved. This commit removes the
'managers' column from the schema, and removes all references to it f
On Thu, Feb 17, 2011 at 05:52:07PM -0800, Andrew Evans wrote:
> Commit d66880ee attempted to revert the original XenServer-shipped versions of
> scripts replaced by OVS during an RPM upgrade, but the logic was incorrect. It
> assumed that %postun of the package being replaced was run before the %po
23 matches
Mail list logo
