Acked-by: Sorin Vinturis
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean
Sent: Thursday, 2 July, 2015 09:36
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH] Solved datapath-windows: BSOD when initializing
switch context
Add check in case of
When closing opened instances, make sure the user dump state is cleared.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/issues/90
---
This patch should be applied both on master and branch 2.4.
---
datapath-windows/ovsext/Datapat
Add check in case of NdisAllocateRWLock fails to allocate the resources.
Check datapath->lock before trying to free it.
Signed-off-by: Alin Gabriel Serdean
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/issues/92
---
datapath-windows/ovsext/Flow.c | 9 +
Trop chaud ? profitez de cette offre! [ promo.cemonelec...@gmail.com ](
http://r.news.cmonelectro.bjimmogroup.com/zavqr42lracvf.html )
[ ]( http://r.news.cmonelectro.bjimmogroup.com/3x7iz0ga59racvd.html )
Ventilateurs, Rafraichisseurs d'air, Humidificateur...
[ ]( http://r.news.cmonelect
Yes, this would make the parse function to fail and in fact it will access the
policy array beyond its boundary.
Thanks Alin and Ben,
Eitan
-Original Message-
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, July 01, 2015 5:13 PM
To: Ben Pfaff; Eitan Eliahu
Cc
Sorry was relocating home.
Thanks for raising the question, it made me realize another problem :)
To answer your question directly: no this patch just ignores the attribute
completely it does not do any additional operations.
I introduced the attribute because it was failing at
https://github.
On Wed, Jul 01, 2015 at 04:12:12PM -0300, Thadeu Lima de Souza Cascardo wrote:
> Add support for MLDv1 and MLDv2. The behavior is not that different from
> IGMP. Packets to all-hosts address and queries are always flooded,
> reports go to routers, routers are added when a query is observed, and
> a
On Wed, Jul 01, 2015 at 04:12:11PM -0300, Thadeu Lima de Souza Cascardo wrote:
> Use IPv6 internally for storing multicast addresses. IPv4 addresses are
> translated to their IPv4-mapped equivalent.
>
> v2:
> * Print mapped IPv4 addresses like IPv4 addresses.
> * Fix alignment and other issues poi
Rather than use "myself" in the title, I'd put in "Luis Pena" or "Luis E. P".
Not a big deal, but I'd put this patch after the typo fix, since you're not an
author at this point in the commit history.
--Justin
> On Jul 1, 2015, at 11:13 AM, Luis E. P wrote:
>
> Signed-off-by: Luis E. P
> --
> On Jul 1, 2015, at 11:13 AM, Luis E. P wrote:
Looks good. I noticed a few other typos around it, which might be nice to fix.
> - If the kernel's flow key includes more fields than the userspace
> @@ -102,7 +102,7 @@ A wildcarded flow can represent a group of exact match
> flows. Each '1'
Alin?
On Wed, Jul 01, 2015 at 11:04:57PM +, Eitan Eliahu wrote:
> Hi Ben,
> In general it doesn't seem that if the datapath does not support an attribute
> it would fail. (In other words the parsing validation code will not fail if
> an attribute is not included in the policy array).
> I am
On Wed, Jul 01, 2015 at 08:03:01PM +, Sorin Vinturis wrote:
> OvsPrepareFlow() returns an error only when the new flow allocation
> fails. In this case HandleFlowPut() should return error without trying
> to free the flow, thus avoiding the BSOD.
>
> Signed-off-by: Sorin Vinturis
> Reported-b
Hi Ben,
In general it doesn't seem that if the datapath does not support an attribute
it would fail. (In other words the parsing validation code will not fail if an
attribute is not included in the policy array).
I am not sure if the addition of this specific attribute to the policy masks an
iss
LGTM. Thanks joe!
Acked-by: Andy Zhou
On Wed, Jul 1, 2015 at 10:34 AM, Joe Stringer wrote:
> Datapath support for some flow key fields is used inside ofproto-dpif as
> well as odp-util. Share these fields using the same structure.
>
> Signed-off-by: Joe Stringer
> ---
> v3: Reduce shared field
On Wed, Jul 1, 2015 at 3:12 PM, Ben Pfaff wrote:
> On Tue, Jun 30, 2015 at 11:19:08PM -0700, Jesse Gross wrote:
>> The addition of Geneve options to packet metadata significantly
>> expanded its size. It was reported that this can decrease performance
>> for DPDK ports by up to 25% since we need t
On Tue, Jun 30, 2015 at 11:19:08PM -0700, Jesse Gross wrote:
> The addition of Geneve options to packet metadata significantly
> expanded its size. It was reported that this can decrease performance
> for DPDK ports by up to 25% since we need to initialize the whole
> structure on each packet recei
On Wed, Jul 01, 2015 at 06:57:50PM +, Alin Serdean wrote:
> Since commit:
> https://github.com/openvswitch/ovs/commit/2e460098bff351b9fddcb917447caa3b97a35d86
> a new packet attribute was introduced.
>
> This patch adds OVS_PACKET_ATTR_PROBE to nlPktExecPolicy in datapath-windows
> and ignores
On Fri, Jun 26, 2015 at 10:20:02PM +, Alin Serdean wrote:
> Acked-by: Alin Gabriel Serdean
>
> -Mesaj original-
> De la: dev [mailto:dev-boun...@openvswitch.org] În numele Nithin Raju
> Trimis: Friday, June 26, 2015 9:51 PM
> Către: dev@openvswitch.org
> Subiect: [ovs-dev] [PATCH 2/2
Jarno, this patch is probably a good one for you to look at. It's a
possible important bug fix and I know that you're knowledgeable about
the mutex in question.
(If you wanted to look at the rest of the series that would be nice too
but this patch in particular may be important.)
On Wed, Jun 24,
On Wed, Jul 01, 2015 at 06:50:38PM +, Alin Serdean wrote:
> Currently OVS_UNUSED is defined in compiler.h since syslog.h is a
> standalone wrapper remove it from the parameters.
>
> Signed-off-by: Alin Gabriel Serdean
> Acked-by: Eitan Eliahu
> ---
> v2: add acked-by
I applied v1 a minute a
Applied to master (let me know if you want it on branch-2.4). Thanks!
On Wed, Jul 01, 2015 at 02:58:42PM +, Eitan Eliahu wrote:
> Acked-by: Eitan Eliahu
>
> Thanks,
> Eitan
>
>
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean
> Sent:
From: Wei li
Signed-off-by: Wei li
Signed-off-by: Ben Pfaff
---
This was submitted via github at:
https://github.com/openvswitch/ovs/pull/53
I already applied it to master and branch-2.4.
diff --git a/AUTHORS b/AUTHORS
index 2240576..845d20c 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -180,6
On Mon, Jun 29, 2015 at 03:18:45AM -0700, Jarno Rajahalme wrote:
>
> > On Jun 26, 2015, at 8:54 AM, Ben Pfaff wrote:
> >
> > On Thu, Jun 25, 2015 at 12:48:27PM +0900, Simon Horman wrote:
> >> Hi Ben,
> >>
> >> On Tue, Jun 23, 2015 at 11:38:56AM -0700, Ben Pfaff wrote:
> >>> Groups were not dest
On Wed, Jul 01, 2015 at 07:01:44PM +, Alin Serdean wrote:
> syslog_direct_create defined in (lib/syslog-direct.c) uses make_unix_socket
> which is currently undefined on the windows build.
>
> We either can remove the new file from the chain but this patch proposes
> to define a wrapper to mak
OvsPrepareFlow() returns an error only when the new flow allocation
fails. In this case HandleFlowPut() should return error without trying
to free the flow, thus avoiding the BSOD.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/is
On 07/01/2015 03:40 PM, Aaron Rosen wrote:
> Fwiw the neutron side is ready to merge as well:
>
> https://review.openstack.org/#/c/196132/
>
> we also were able to test your patch in the gate
> https://review.openstack.org/#/c/196131/4 with this patchset on top and it
> fixes the race we were see
Now that struct cls_match has 'add_version' the 'version' in cls_match
was largely redundant. Remove 'version' from struct cls_rule, and add
it to function prototypes that need it. This makes versioning more
explicit (or less indirect) in the API.
Suggested-by: Ben Pfaff
Signed-off-by: Jarno Ra
It is cleaner to not use ofp_bundle_entry for non-bundle flow mods.
To address this, the new struct ofproto_flow_mod combines an
ofputil_flow_mod and the necessary execution context for executing the
start, revert, and finish phases of the flow mod, which all were
previously members of struct ofp_b
Fwiw the neutron side is ready to merge as well:
https://review.openstack.org/#/c/196132/
we also were able to test your patch in the gate
https://review.openstack.org/#/c/196131/4 with this patchset on top and it
fixes the race we were seeing before.
Aaron
On Mon, Jun 29, 2015 at 5:21 PM, Ale
Use IPv6 internally for storing multicast addresses. IPv4 addresses are
translated to their IPv4-mapped equivalent.
v2:
* Print mapped IPv4 addresses like IPv4 addresses.
* Fix alignment and other issues pointed out by clang, sparse and Ben Pfaff.
Signed-off-by: Thadeu Lima de Souza Cascardo
Cc:
Add support for MLDv1 and MLDv2. The behavior is not that different from
IGMP. Packets to all-hosts address and queries are always flooded,
reports go to routers, routers are added when a query is observed, and
all MLD packets go through slow path.
v2:
* Avoid an infinite loop when an unknown reco
This series introduces Multicast Listener Discovery support. It converts
internal representation of addresses to IPv6, including mapping IPv4 to mapped
IPv4.
Thadeu Lima de Souza Cascardo (2):
mcast-snooping: Use IPv6 address for MDB
Multicast Listener Discovery support
NEWS
syslog_direct_create defined in (lib/syslog-direct.c) uses make_unix_socket
which is currently undefined on the windows build.
We either can remove the new file from the chain but this patch proposes
to define a wrapper to make_unix_socket in which we return EINVAL.
This will avoid this kind of p
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, July 01, 2015 11:49 AM
To: Eitan Eliahu; dev@openvswitch.org
Subject: RE: [PATCH 2/2] Undefined symbol make_unix_socket
Hi Eitan,
It would be a nice i
Since commit:
https://github.com/openvswitch/ovs/commit/2e460098bff351b9fddcb917447caa3b97a35d86
a new packet attribute was introduced.
This patch adds OVS_PACKET_ATTR_PROBE to nlPktExecPolicy in datapath-windows
and ignores it for the moment to maintain binary compatibility.
Signed-off-by: Alin
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Sorin Vinturis
Trimis: Wednesday, July 1, 2015 8:28 PM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH] datapath-windows: Solved BSOD when handling flows
OvsPrepareFlow() retu
Currently OVS_UNUSED is defined in compiler.h since syslog.h is a
standalone wrapper remove it from the parameters.
Signed-off-by: Alin Gabriel Serdean
Acked-by: Eitan Eliahu
---
v2: add acked-by
---
include/windows/syslog.h | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --g
Hi Eitan,
It would be a nice idea for a future feature. I think this issue is related to
that: https://github.com/openvswitch/ovs-issues/issues/74.
Alin.
-Mesaj original-
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Wednesday, July 1, 2015 6:03 PM
Către: Alin Serdean; dev@op
Hi,
The allowed address pair extension was added to the neutron api to allow
protocols like VRRP to work. All it dictates are
mac_address/ip_address(cidrs) that are allowed to pass through a neutron
port.
rest inline
On Wed, Jul 1, 2015 at 10:47 AM, Ben Pfaff wrote:
> On Wed, Jul 01, 2015 at 1
Signed-off-by: Luis E. P
---
AUTHORS | 1 +
1 file changed, 1 insertion(+)
diff --git a/AUTHORS b/AUTHORS
index 2240576..df6859f 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -293,6 +293,7 @@ Krishna Miriyalakris...@nicira.com
Len Gao l...@vmware.com
Logan Rosen log
Signed-off-by: Luis E. P
---
datapath/README.md | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/datapath/README.md b/datapath/README.md
index 1a4d8e1..5fe4af2 100644
--- a/datapath/README.md
+++ b/datapath/README.md
@@ -41,7 +41,7 @@ flow key that it parsed from the packet
On Wed, Jul 01, 2015 at 12:25:58PM -0500, Kyle Mestery wrote:
> On Wed, Jul 1, 2015 at 12:11 PM, Ben Pfaff wrote:
> > Where's the spec for allowed address pairs? It's probably pretty easy
> > to implement in OVN.
> >
> The API developer documentation is here [1]. The BP with a link to a google
>
Datapath support for some flow key fields is used inside ofproto-dpif as
well as odp-util. Share these fields using the same structure.
Signed-off-by: Joe Stringer
---
v3: Reduce shared fields, as suggested by Andy.
v2: Rebase against master.
---
lib/dpif-netdev.c | 10 +++---
li
OvsPrepareFlow() returns an error only when the new flow allocation
fails. In this case HandleFlowPut() should return error without trying
to free the flow, thus avoiding the BSOD.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/is
On Wed, Jul 1, 2015 at 12:11 PM, Ben Pfaff wrote:
> On Wed, Jul 01, 2015 at 11:11:05AM +0300, Gal Sagie wrote:
> > As you might know, allowed address pairs in neutron is an extension to
> > allow port
> > to have more then a pair of MAC-IP addresses assigned to it.
> > This is useful for cases of
On Wed, Jul 01, 2015 at 03:32:18PM +, Sorin Vinturis wrote:
> Tunnel ports are not initialized with the corresponding default port.
> The newly allocated vport is not yet initialized and the ovsType
> member does not reflect the correct tunnel port type, thus the
> transport port destination wo
On Wed, Jul 01, 2015 at 03:21:03PM +, Sorin Vinturis wrote:
> VXLAN tunnel tx packet stats are incorrectly incremented.
>
> Signed-off-by: Sorin Vinturis
> Reported-by: Sorin Vinturis
> Reported-at: https://github.com/openvswitch/ovs-issues/issues/89
> Acked-by: Eitan Eliahu
> ---
> This pa
On Wed, Jul 01, 2015 at 03:20:27PM +, Sorin Vinturis wrote:
> All pool allocations should have unique tags.
> OVS_STT_POOL_TAG matches the OVS_MEMORY_TAG.
>
> Changed the OVS_STT_POOL_TAG to make it unique.
>
> Signed-off-by: Sorin Vinturis
> Acked-by: Eitan Eliahu
> ---
> This patch should
On Wed, Jul 01, 2015 at 11:11:05AM +0300, Gal Sagie wrote:
> As you might know, allowed address pairs in neutron is an extension to
> allow port
> to have more then a pair of MAC-IP addresses assigned to it.
> This is useful for cases of where few VM's need to share virtual MAC/IP,
> like
> for VRR
On Wed, Jul 01, 2015 at 08:30:20AM -0400, Russell Bryant wrote:
> On 07/01/2015 04:11 AM, Gal Sagie wrote:
> > Hello Everyone,
> >
> > As you might know, allowed address pairs in neutron is an extension to
> > allow port
> > to have more then a pair of MAC-IP addresses assigned to it.
> > This is
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Jul 1, 2015 at 2:20 AM, Loftus, Ciara wrote:
>> On Tue, Jun 30, 2015 at 7:56 AM, Loftus, Ciara
>> wrote:
>> >>
>> >> On Wed, Jun 24, 2015 at 1:17 PM, Ben Pfaff wrote:
>> >> > On Fri, Jun 19, 2015 at 04:13:22PM -0700, Jesse Gross wrote:
>> >> >> Tunnel metadata can be substantially larger
Sorin,
I do understand the issue but the use of the signature to solve it is not
clean. The signature field in the context structure is used for validation
during debug. It could be removed and the code should still work.
We should not use the signature to identify self-allocated NBLs.
Please co
Eitan,
Consider the scenario below.
We receive a packet (NBL), that has multiple NBs, in OvsStartNBLIngress(). The
NBL is not allocated by OVS and the NBL context is allocated and initialized in
OvsInitExternalNBLContext(). After this the OvsCreateNewNBLsFromMultipleNBs()
function is called an
Thx a lot for the review,
Applied to master and branch-2.4
On Wed, Jul 1, 2015 at 6:11 AM, Flavio Leitner wrote:
> On Tue, Jun 30, 2015 at 06:29:02PM -0700, Alex Wang wrote:
> > This commit adds a rpmbuild spec file for building ovs datapath
> > in dkms similar to the one for debian.
> >
> > Si
Tunnel ports are not initialized with the corresponding default port.
The newly allocated vport is not yet initialized and the ovsType
member does not reflect the correct tunnel port type, thus the
transport port destination won't be correctly initialized.
Signed-off-by: Sorin Vinturis
Reported-b
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: Sorin Vinturis [mailto:svintu...@cloudbasesolutions.com]
Sent: Wednesday, July 01, 2015 8:27 AM
To: Eitan Eliahu; dev@openvswitch.org
Subject: RE: [PATCH] datapath-windows: Failed initialization for datapath
tunnel ports
Ei
Eitan,
This code was changed by the STT patch. If the created tunnel has specified the
tunnel destination port, then the issue does not occur. But if the created
tunnel does not have a specified tunnel port, then the default port is not set
and the tunnel destination port will be zero, thus the
Hi Sorin,
Can you explain which change broke this code.
I recall validating connectivity on STT and I believe Nithin validated on STT
and on VXLAN channel as well.
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Sorin Vinturis
Sent: Wednesday,
VXLAN tunnel tx packet stats are incorrectly incremented.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/issues/89
Acked-by: Eitan Eliahu
---
This patch should be applied both on master and branch 2.4.
---
datapath-windows/ovsex
All pool allocations should have unique tags.
OVS_STT_POOL_TAG matches the OVS_MEMORY_TAG.
Changed the OVS_STT_POOL_TAG to make it unique.
Signed-off-by: Sorin Vinturis
Acked-by: Eitan Eliahu
---
This patch should be applied both on master and branch 2.4.
---
datapath-windows/ovsext/Util.h | 2
Hi Sorin,
Thank you for debugging this issue.
I think that the NBL context signature is used for code validation only (think
about a signature we put in an object for the purpose of validating this object
when executing in a different context).
Can you think about another way to fix this issue (e
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Sorin Vinturis
Sent: Wednesday, July 01, 2015 1:04 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH] datapath-windows: Avoid double incrementation of tx
VXLAN packet st
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Sorin Vinturis
Sent: Wednesday, July 01, 2015 1:13 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH] datapath-windows: Duplicate tag for STT pools
All pool allocations
Hi Alin,
Should we implement OVS system logging for Windows through the use of
"Application Event Logging"?
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean
Sent: Tuesday, June 30, 2015 9:28 AM
To: dev@openvswitch.org
Subject: [ovs
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean
Sent: Tuesday, June 30, 2015 9:27 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH 1/2] OVS_UNUSED is defined indirectly
Currently OVS_UNUSED is defined
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean
Sent: Tuesday, June 30, 2015 9:39 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH] datapath-windows: OVS_PACKET_ATTR_PROBE
Since commit:
https://urldefens
Thanks Mark..
I will this and let you know the status.
On Wed, Jul 1, 2015 at 7:25 PM, Gray, Mark D wrote:
> >
> > I'm not using vhost at present. But using macvtap devices. Below screen
> shot
> > for your reference and following are the details on host.
> >
> >
> >
> > [root@artha utilities
This BSOD occurred in the context of a packet (NBL) with multiple
NET_BUFFER(s) (NBs). The reason for the BSOD is due to the marking
of NBLs created by OVS as being external and wrongly completing them.
This patch should be applied both on master and branch 2.4.
Signed-off-by: Sorin Vinturis
Rep
>
> I'm not using vhost at present. But using macvtap devices. Below screen shot
> for your reference and following are the details on host.
>
>
>
> [root@artha utilities]# ifconfig macvtap2
> macvtap2: flags=4163 mtu 1500
> inet6 fe80::5054:ff:fec7:ed4f prefixlen 64 scopeid 0x20
>
On Tue, Jun 30, 2015 at 06:29:02PM -0700, Alex Wang wrote:
> This commit adds a rpmbuild spec file for building ovs datapath
> in dkms similar to the one for debian.
>
> Signed-off-by: Alex Wang
>
> ---
LGTM
Acked-by: Flavio Leitner
___
dev mailing
On 07/01/2015 04:11 AM, Gal Sagie wrote:
> Hello Everyone,
>
> As you might know, allowed address pairs in neutron is an extension to
> allow port
> to have more then a pair of MAC-IP addresses assigned to it.
> This is useful for cases of where few VM's need to share virtual MAC/IP,
> like
> for
Re-added the dev list
>
> Briefly:
>
> I'm using rhel7 as host. On host I created ovs bridge for i350 nic.
>
> Created kvm instance on the host. My guest uses the ovs bridge mactap
> devices (pkt0 & pkt1), on which I will receive media traffic.
I don’t see any ports to the guest? Could you use
New stats for vhost ports are rx_bytes, tx_bytes, multicast, rx_errors and
rx_length_errors. New stats for PMD ports are rx_dropped, rx_length_errors,
rx_crc_errors and rx_missed_errors. DPDK imissed packets are now classified
as dropped instead of errors.
Signed-off-by: Timo Puha
---
lib/netdev
> On Tue, Jun 30, 2015 at 7:56 AM, Loftus, Ciara
> wrote:
> >>
> >> On Wed, Jun 24, 2015 at 1:17 PM, Ben Pfaff wrote:
> >> > On Fri, Jun 19, 2015 at 04:13:22PM -0700, Jesse Gross wrote:
> >> >> Tunnel metadata can be substantially larger than our existing fields
> >> >> (up to 124 bytes in a sing
All pool allocations should have unique tags.
OVS_STT_POOL_TAG matches the OVS_MEMORY_TAG.
Changed the OVS_STT_POOL_TAG to make it unique.
Signed-off-by: Sorin Vinturis
---
datapath-windows/ovsext/Util.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/datapath-windows/ovsext
Hello Everyone,
As you might know, allowed address pairs in neutron is an extension to
allow port
to have more then a pair of MAC-IP addresses assigned to it.
This is useful for cases of where few VM's need to share virtual MAC/IP,
like
for VRRP, Load balancing, NFV use cases and so on...
(Aaron w
VXLAN tunnel tx packet stats are incorrectly incremented.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/issues/89
---
datapath-windows/ovsext/Actions.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/datapath-windows/ovsext/A
Tunnel ports are not initialized with the corresponding default port.
The newly allocated vport is not yet initialized and the ovsType
member does not reflect the correct tunnel port type, thus the
transport port destination won't be correctly initialized.
Signed-off-by: Sorin Vinturis
Reported-b
79 matches
Mail list logo
