This commit extends the vtep module to support creating the
'Ucast_Macs_Remote' table entries in the vtep database for
MAC addresses on the ovn logical ports.
Signed-off-by: Alex Wang
Acked-by: Russell Bryant
---
V8->V9:
- Add Ack from Russell.
- Minor adjustment based on Russell's review.
V7->
This commit adds the vtep module to ovn-controller-vtep. The
module will scan through the Port_Binding table in OVN-SB database,
and update the vtep logcial switches tunnel keys.
Signed-off-by: Alex Wang
Acked-by: Russell Bryant
---
V8->V9:
- Add Ack from Russell.
- Fix minor comments.
V7->V8:
This commit conducts the following documentation changes:
* add a description in ovn-architecture manual for
the life cycle about VTEP gateway.
* add TODOs related to ovn-controller-vtep.
* refine the ovn-sb, ovn-nb schema manual to require
logical 'port' type and 'options' configu
Hey Russell,
Sorry for this very delayed reply,~
Please see my reply inline,
Thanks,
Alex Wang,
On Mon, Aug 24, 2015 at 10:48 AM, Russell Bryant wrote:
> On 08/23/2015 02:06 PM, Alex Wang wrote:
> > This commit extends the vtep module to support creating the
> > 'Ucast_Macs_Remote' table entr
When dpdk configuration changes, all pmd threads are recreated
and rx queues of each port are reloaded. After this process,
rx queue could be mapped to a different pmd thread other than
the one before reconfiguration. However, this is totally
transparent to ofproto layer modules. So, if the ofpr
In the push_ukey_ops__(), when flow_del operation returns error, the 'struct
stats' passed to the operation function will be set to all zero. And we
should not use it to calculate the delta (i.e. minus the zero stats by the
cached stats causes overflow).
Even though this should rarely happen, it
On Thu, Aug 27, 2015 at 7:07 PM, Ramu Ramamurthy
wrote:
> On 2015-08-27 18:26, Jesse Gross wrote:
>>
>> On Thu, Aug 27, 2015 at 5:46 PM, Ramu Ramamurthy
>> wrote:
>>>
>>> We found that enabling UDP checksums for vxlan triggers GRO on the
>>> receiver,
>>> and boosts the vxlan performance to 8+ Gb
On 2015-08-27 18:26, Jesse Gross wrote:
On Thu, Aug 27, 2015 at 5:46 PM, Ramu Ramamurthy
wrote:
We found that enabling UDP checksums for vxlan triggers GRO on the
receiver,
and boosts the vxlan performance to 8+ Gbps.
Hence, we want to enable vxlan csum for OVS vxlan tunnels. This is
already
Currently, when using an OpenFlow 1.0 connection to encode a
tunnel metadata set field action, a series of NXAST_REG_LOADs
are emitted. The result is something like this:
actions=load:0xa->NXM_NX_TUN_METADATA0[0..63],load:0->
NXM_NX_TUN_METADATA0[64..127],load:0->NXM_NX_TUN_METADATA0[128..191],
lo
On Thu, Aug 27, 2015 at 5:46 PM, Ramu Ramamurthy
wrote:
> We found that enabling UDP checksums for vxlan triggers GRO on the receiver,
> and boosts the vxlan performance to 8+ Gbps.
> Hence, we want to enable vxlan csum for OVS vxlan tunnels. This is already
> supported in OVS 2.4, and kernel vers
Looks good to me,
Thx for fixing it~
On Thu, Aug 27, 2015 at 11:11 AM, Ben Pfaff wrote:
> "sed -i" isn't entirely portable, and we can avoid it by using the
> argument to check_logs as intended.
>
> Signed-off-by: Ben Pfaff
> Acked-by: Alex Wang
> ---
> v1->v2: Retain $1 in check_logs call fr
On Thu, Aug 27, 2015 at 5:15 PM, Justin Pettit wrote:
> Sadly:
>
> Acked-by: Justin Pettit
>
>
Thx, I also feel sad
Will be around~
Thanks,
Alex Wang,
> --Justin
>
>
> > On Aug 27, 2015, at 4:59 PM, Alex Wang wrote:
> >
> > Signed-off-by: Alex Wang
> > ---
> > AUTHORS | 2 +-
> > 1 file c
posting again, since the earlier emailer mangles email addresses.
Overview:
We are running openstack kilo with neutron OVS using intel 10G
adapters (br kx4 dual-port, 82599es).
neutron OVS is using vxlan tunnels. We are using RHEL 7.1 (which uses
the 3.x kernel).
Openvswitch kernel modul
Signed-off-by: Nithin Raju
---
AUTHORS | 1 +
1 file changed, 1 insertion(+)
diff --git a/AUTHORS b/AUTHORS
index 254095e..6c683f8 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -157,6 +157,7 @@ Rob Hoesrob.h...@citrix.com
Romain Lenglet romain.leng...@berabera.info
Russell Bry
Overview:
We are running openstack kilo with neutron OVS using intel 10G adapters (br
kx4 dual-port, 82599es).
neutron OVS is using vxlan tunnels. We are using RHEL 7.1 (which uses the 3.x
kernel).
Openvswitch kernel module is from upstream (3.x) and ovs is 2.3.0 based.
Problem:
We find
Sadly:
Acked-by: Justin Pettit
--Justin
> On Aug 27, 2015, at 4:59 PM, Alex Wang wrote:
>
> Signed-off-by: Alex Wang
> ---
> AUTHORS | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/AUTHORS b/AUTHORS
> index 254095e..4be28c4 100644
> --- a/AUTHORS
> +++ b/AUTHORS
> @
Signed-off-by: Alex Wang
---
AUTHORS | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/AUTHORS b/AUTHORS
index 254095e..4be28c4 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -7,7 +7,7 @@ Alexander Duyck alexander.h.du...@redhat.com
Alexandru Copot alex.miha...@gmail.com
> On Aug 27, 2015, at 3:08 PM, Alin Serdean
> wrote:
>
> Alin :) (not Sorin).
Of course :)
> Yes both ports are tunneling ports.
>
> We cannot save the tunkey because it is different the two ports have
> different destination addresses.
>
> OvsOutputBeforeSetAction indeed reinitializes
>
Alin :) (not Sorin).
Yes both ports are tunneling ports.
We cannot save the tunkey because it is different the two ports have different
destination addresses.
OvsOutputBeforeSetAction indeed reinitializes
ovsFwdCtx->fwdDetail->SourcePortId and ovsFwdCtx->fwdDetail->SourceNicIndex
(https://gi
hi Sorin,
Looks good overall.
I just one one comment about the patch which I’ve inlined.
How did you test this BTW?
> VOID
> @@ -1309,7 +1405,7 @@ OvsTunnelFilterThreadUninit(POVS_TUNFLT_THREAD_CONTEXT
> threadCtx)
> * --
On Thu, Aug 27, 2015 at 05:16:19PM -0400, Russell Bryant wrote:
> On 08/27/2015 04:26 PM, Flavio Leitner wrote:
> > On Thu, Aug 27, 2015 at 03:49:13PM -0400, Russell Bryant wrote:
> >> On 08/27/2015 03:26 PM, Flavio Leitner wrote:
> >>> On Wed, Aug 26, 2015 at 05:46:53PM -0400, Russell Bryant wrote
hi Sorin,
OK, if I understand you correctly, what you are saying is that, ‘output2' is
also tunnel port, and when we output the packet to ‘output1’, all of the tunnel
context is lost, and we’ll not be able to output the packet to ‘output2’. Is
this right? If so, the fix you are putting in is no
On 08/27/2015 04:26 PM, Flavio Leitner wrote:
> On Thu, Aug 27, 2015 at 03:49:13PM -0400, Russell Bryant wrote:
>> On 08/27/2015 03:26 PM, Flavio Leitner wrote:
>>> On Wed, Aug 26, 2015 at 05:46:53PM -0400, Russell Bryant wrote:
The previous systemd units for OVN had systemd create a rundir fo
Hi Nithin,
Regarding the question about the flows in the form:
strip_vlan,set_tunnel,output15,output16 the main user behind it is neutron. A
nice explanation why this is done can be found:
http://assafmuller.com/2013/10/14/gre-tunnels-in-openstack-neutron/ (look at
Table 21: cookie=0x0, durati
> Original Message-
> From: Nithin Raju [mailto:nit...@vmware.com]
> Sent: Monday, 3 August, 2015 20:22
> To: Sorin Vinturis
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH] datapath-windows: Enable failure after
> restarting extension
>
>> On Jul 15, 2015, at 7:50 AM, Sorin Vi
> On Jul 20, 2015, at 12:00 PM, Sorin Vinturis
> wrote:
>
> In order to support IRP cancelling mechanism for pending IRPs, all
> tunnel filter requests, VXLAN create/delete tunnel, need to be
> processed iteratively.
>
> Signed-off-by: Sorin Vinturis
hi Sorin,
I had a question regarding laten
On Thu, Aug 27, 2015 at 03:49:13PM -0400, Russell Bryant wrote:
> On 08/27/2015 03:26 PM, Flavio Leitner wrote:
> > On Wed, Aug 26, 2015 at 05:46:53PM -0400, Russell Bryant wrote:
> >> The previous systemd units for OVN had systemd create a rundir for
> >> each daemon (ovn-northd and ovn-controller
On Thu, Aug 27, 2015 at 12:53:32PM -0700, Gurucharan Shetty wrote:
> > That would break OSP which creates a bridge only in OVS, so it must
> > preserve the configuration between reboots.
>
> What is OSP?
OpenStack Platform
fbl
___
dev mailing list
dev
> This patch surpresses flow attribute probing in the windows datapath.
>
> Signed-off-by: Alin Gabriel Serdean
Alin,
The change look good. Only thing I’d avoid is the OVS_LOG_ERROR() statement.
The documentation for ‘OVS_FLOW_ATTR_PROBE’ explicitly mentions that logging
should be suppressed.
> That would break OSP which creates a bridge only in OVS, so it must
> preserve the configuration between reboots.
What is OSP?
>
> However, OSP also creates tons of temporary ports which after a reboot
> they simply don't make sense and just cause problems. We can safely
> remove them since the
On Thu, Aug 27, 2015 at 11:28:21AM -0700, Ben Pfaff wrote:
> Actually, that makes me wonder whether --delete-bridges is the right
> solution. It makes a lot of sense to start fresh from an empty set of
> bridges at boot time because it provides a "clean slate" at each boot on
> which to reproducib
On 08/27/2015 03:26 PM, Flavio Leitner wrote:
> On Wed, Aug 26, 2015 at 05:46:53PM -0400, Russell Bryant wrote:
>> The previous systemd units for OVN had systemd create a rundir for
>> each daemon (ovn-northd and ovn-controller). This "worked", in that
>> the services did start successfully. Howe
hi Alin,
I am trying to reason with what this patch is trying to accomplish. I might be
missing something here, but here’s what I’m thinking:
OvsOutputBeforeSetAction() is typically called for an action such as:
outport1, set_tunnel(), outport2
Here ‘outport2’ is a tunnel port, and output1 is a
On Wed, Aug 26, 2015 at 05:46:55PM -0400, Russell Bryant wrote:
> Most real deployments will need to customize the database locations
> for ovn-controller and ovn-controller-vtep. Instead of making them
> override the entire command used to start the daemons, provide and
> document some environmen
Hi,
I love Gal's idea of having IP field, which make it possible for a
pure L3 overlay (i.e. no L2 header encapsulated), with the help of ARP
responders on each transport node.
It would be efficient in cloud env where L3 is sufficient.
Han
On Tue, Jul 14, 2015 at 9:08 PM, Gal Sagie wrote:
> Thi
>
> Gurucharan, I think that solves your concern about restoring flows. Is that OK
> for you?
Yes.
>
> Thanks.
> Cascardo.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Aug 26, 2015 at 05:46:54PM -0400, Russell Bryant wrote:
> Signed-off-by: Russell Bryant
> ---
Acked-by: Flavio Leitner
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Aug 26, 2015 at 05:46:53PM -0400, Russell Bryant wrote:
> The previous systemd units for OVN had systemd create a rundir for
> each daemon (ovn-northd and ovn-controller). This "worked", in that
> the services did start successfully. However, they didn't actually
> work. In practice, bot
On Thu, Aug 27, 2015 at 11:28:21AM -0700, Ben Pfaff wrote:
> On Thu, Aug 27, 2015 at 12:07:32PM -0300, Thadeu Lima de Souza Cascardo wrote:
> > When using virtualization, new ports are created and removed all the time.
> > These
> > ports do not persist after a system reboot, for example. They may
On Wednesday 26 August 2015 06:14 PM, Timo Puha wrote:
Update relevant artifacts to add support for DPDK v2.1.0
- INSTALL.DPDK.md
- acinclude.m4: Change DPDK library name
- netdev-dpdk: Add 16 bytes extra padding to mbuf size to adapt to DPDK bug
fix that changes the treatment of the re
On Thu, Aug 27, 2015 at 12:07:32PM -0300, Thadeu Lima de Souza Cascardo wrote:
> When using virtualization, new ports are created and removed all the time.
> These
> ports do not persist after a system reboot, for example. They may be created
> again by the virtualization manager, but that will ha
On Thu, Aug 27, 2015 at 12:48:41AM -0700, Alex Wang wrote:
> Since the check_log() only uses $1, this still does not totally fix the
> problem,
Oops.
I posted a v3 which uses a slightly different approach from your suggestion:
http://openvswitch.org/pipermail/dev/2015-August/059307.html
_
"sed -i" isn't entirely portable, and we can avoid it by using the
argument to check_logs as intended.
Signed-off-by: Ben Pfaff
Acked-by: Alex Wang
---
v1->v2: Retain $1 in check_logs call from OVN_CONTROLLER_VTEP_STOP.
Also drop other, now-unneeded, call to "sed -i".
v2->v3: Merge multiple ch
The patch looks good, thanks.
Two comments below
On 26/08/2015 13:44, "Timo Puha" wrote:
>Update relevant artifacts to add support for DPDK v2.1.0
> - INSTALL.DPDK.md
> - acinclude.m4: Change DPDK library name
> - netdev-dpdk: Add 16 bytes extra padding to mbuf size to adapt to DPDK
>bug
> fi
> On Aug 26, 2015, at 7:31 PM, Ben Pfaff wrote:
>
> On Wed, Aug 26, 2015 at 05:42:46PM -0700, Jarno Rajahalme wrote:
>> Without an explicit bounds check GCC 4.9 issues an array out of bounds
>> error. This patch adds explicit checks which will however be
>> optimized away as the relevant parame
> On Aug 26, 2015, at 6:56 PM, Jesse Gross wrote:
>
> On Wed, Aug 26, 2015 at 5:42 PM, Jarno Rajahalme
> wrote:
>> Initialize the whole 'exact_match_mask' in
>> mf_mask_field_and_prereqs(), as geneve options can be bigger than 16
>> bytes.
>>
>> Requested-by: Jesse Gross
>> Signed-off-by: Ja
Thank you all for the comments and inputs.
we will cross-port this thread on openstack mailing list and see what needs to
be done.
Another change we proposed in this patch set is adding support for new
bridge(br-phy) in OVN.
Please refer the “[PATCH 3/4] ovn-controller: Controller man page modi
On Thu, Aug 27, 2015 at 12:07:31PM -0300, Thadeu Lima de Souza Cascardo wrote:
> Signed-off-by: Thadeu Lima de Souza Cascardo
Thanks, applied to master.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On 08/27/2015 11:07 AM, Thadeu Lima de Souza Cascardo wrote:
> Signed-off-by: Thadeu Lima de Souza Cascardo
> ---
> utilities/ovs-ctl.in | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/utilities/ovs-ctl.in b/utilities/ovs-ctl.in
> index 918be2c..9bbbe0d 100755
> --- a/ut
On Thu, Aug 27, 2015 at 8:07 AM, Thadeu Lima de Souza Cascardo
wrote:
> When using virtualization, new ports are created and removed all the time.
> These
> ports do not persist after a system reboot, for example. They may be created
> again by the virtualization manager, but that will happen aft
Signed-off-by: Thadeu Lima de Souza Cascardo
---
utilities/ovs-ctl.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/utilities/ovs-ctl.in b/utilities/ovs-ctl.in
index 918be2c..9bbbe0d 100755
--- a/utilities/ovs-ctl.in
+++ b/utilities/ovs-ctl.in
@@ -216,7 +216,7 @@ start_ovsdb
When using virtualization, new ports are created and removed all the time. These
ports do not persist after a system reboot, for example. They may be created
again by the virtualization manager, but that will happen after the vswitch is
already running, and the virtualization manager will add them
This message was not delivered due to the following reason:
Your message could not be delivered because the destination server was
not reachable within the allowed queue period. The amount of time
a message is queued before it is returned depends on local configura-
tion parameters.
Most likely t
Since the check_log() only uses $1, this still does not totally fix the
problem,
How about something like this?
diff --git a/tests/ofproto-macros.at b/tests/ofproto-macros.at
index 0dc208e..f0a58a5 100644
--- a/tests/ofproto-macros.at
+++ b/tests/ofproto-macros.at
@@ -128,7 +128,12 @@ check_logs
Hi Pravin,
On Mon, Aug 17, 2015 at 11:33:59AM -0700, Pravin Shelar wrote:
> On Thu, Aug 13, 2015 at 6:30 PM, Simon Horman
> wrote:
> > When an error occurs skipping IPv6 extension headers retain the already
> > parsed IP protocol and IPv6 addresses in the flow. Also assume that the
> > packet is
55 matches
Mail list logo
