[ovs-dev] log flooded with: cannot get STP status on nonexistent port 7

Hi, I am using Ubuntu 12.04 with the default OVS packages based on the 1.4 release in combination with Xen. I have observed a problem with stubdom domains causing the logs to be flooded with messages such as: cannot get STP status on nonexistent port 7 The cause is that Xen removes a stubdo

[ovs-dev] [Patch] datapath: fix checksum after putting vlan tag

After inserting the VLAN tag with __vlan_put_tag(), we should re-calculate the skb checksum if it is done. Cc: Jesse Gross Signed-off-by: Cong Wang --- diff --git a/datapath/vport-netdev.c b/datapath/vport-netdev.c index 727194e..99a6eca 100644 --- a/datapath/vport-netdev.c +++ b/datapath/vport

[ovs-dev] "Els Cenci", d'A. Dumas

"La Lluna en un Cove" presenta "Els Cenci" d'Alexandre Dumas amb traducció de Carles Llorach [...] Havent vist ara la tomba, si desitgeu una impressió més vívida dels actors principals d’aquesta tragèdia de la que hagueu pogut tenir amb la narració dels fets, feu una visita a la Galleria Ba

Re: [ovs-dev] How to set VXLAN VNI?

On Feb 25, 2013, at 10:00 PM, Changbin Liu wrote: > > Hi Kyle, > > Thanks so much for your reply! > > To clarify, I noticed the code was broken recently and have been using > "3b6f2889400fd340b851c2d36356457559ae6e81", where you fixed one VXLAN bug in > early Jan. > > I tried your first way:

Re: [ovs-dev] How to set VXLAN VNI?

On Tue, Feb 26, 2013 at 9:54 AM, Kyle Mestery (kmestery) wrote: > On Feb 25, 2013, at 10:00 PM, Changbin Liu wrote: > > > > Hi Kyle, > > > > Thanks so much for your reply! > > > > To clarify, I noticed the code was broken recently and have been using > > "3b6f2889400fd340b851c2d36356457559ae6e81

Re: [ovs-dev] How to set VXLAN VNI?

On Feb 26, 2013, at 9:21 AM, Changbin Liu wrote: > > > On Tue, Feb 26, 2013 at 9:54 AM, Kyle Mestery (kmestery) > wrote: > On Feb 25, 2013, at 10:00 PM, Changbin Liu wrote: > > > > Hi Kyle, > > > > Thanks so much for your reply! > > > > To clarify, I noticed the code was broken recently and h

[ovs-dev] branch-1.7 and kernel 3.4

Do you have plans to support branch-1.7 on kernels 3.4 ? Patch below allows to compile datapath module. After loading module gre tunnels and pure l2 switching is working. I'am just wondering how you decide which kernel version is supported ? --- openvswitch-1.7.3-dist/datapath/datapath.c 2013-01-

Re: [ovs-dev] branch-1.7 and kernel 3.4

I found commit on head for 3.4: http://openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=commitdiff;h=8b9a3d50b433c7fbed7b31b3d6f2bda966440b41 for 3.7 http://openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=commitdiff;h=df2652105fd562ebd66590276cc939c67d37b3e5 and for 3.8 http://openvswitc

Re: [ovs-dev] log flooded with: cannot get STP status on nonexistent port 7

On Tue, Feb 26, 2013 at 09:10:49AM +, James Dingwall wrote: > I am using Ubuntu 12.04 with the default OVS packages based on the > 1.4 release in combination with Xen. I have observed a problem with > stubdom domains causing the logs to be flooded with messages such > as: > > cannot get STP s

Re: [ovs-dev] branch-1.7 and kernel 3.4

On Tue, Feb 26, 2013 at 8:30 AM, Tomasz Paszkowski wrote: > I found commit on head for 3.4: > > http://openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=commitdiff;h=8b9a3d50b433c7fbed7b31b3d6f2bda966440b41 > > for 3.7 > > http://openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=commitdiff;h=df2

Re: [ovs-dev] branch-1.7 and kernel 3.4

Hi Jesse, Thank for your response. So if I'll make this module running on newer kernels is there a chance for those changes to be merged in ? On Tue, Feb 26, 2013 at 5:48 PM, Jesse Gross wrote: > On Tue, Feb 26, 2013 at 8:30 AM, Tomasz Paszkowski wrote: >> I found commit on head for 3.4: >> >>

Re: [ovs-dev] branch-1.7 and kernel 3.4

If it helps, we're planning to release OVS 1.9 in the next day or two, which includes support for kernels up to 3.8. This will also be a long-term supported branch. --Justin On Feb 26, 2013, at 9:00 AM, Tomasz Paszkowski wrote: > Hi Jesse, > > Thank for your response. So if I'll make this

Re: [ovs-dev] branch-1.7 and kernel 3.4

Great :-) So I'll focus on 1.9 ! On Tue, Feb 26, 2013 at 6:19 PM, Justin Pettit wrote: > If it helps, we're planning to release OVS 1.9 in the next day or two, which > includes support for kernels up to 3.8. This will also be a long-term > supported branch. > > --Justin > > > On Feb 26, 2013

Re: [ovs-dev] log flooded with: cannot get STP status on nonexistent port 7

On Feb 26, 2013, at 8:46 AM, Ben Pfaff wrote: > Justin, this seems like a reasonable enough request to me. But since > we're generally pretty reluctant to change branch-1.4 without good > reason, I thought I'd get a second opinion from you. What do you think? It sounds reasonable to me, too.

Re: [ovs-dev] log flooded with: cannot get STP status on nonexistent port 7

On Tue, Feb 26, 2013 at 09:40:42AM -0800, Justin Pettit wrote: > On Feb 26, 2013, at 8:46 AM, Ben Pfaff wrote: > > > Justin, this seems like a reasonable enough request to me. But since > > we're generally pretty reluctant to change branch-1.4 without good > > reason, I thought I'd get a second

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

Looks good to me with the following incremental which I've applied. I'll merge shortly. Thanks for doing this work Kyle. Ethan --- ofproto/ofproto-dpif.c |4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index 8009c1c..9f7

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

Thanks for the review Ethan, one question on this below. On Feb 26, 2013, at 1:10 PM, Ethan Jackson wrote: > Looks good to me with the following incremental which I've applied. I'll > merge > shortly. Thanks for doing this work Kyle. > > Ethan > > --- > ofproto/ofproto-dpif.c |4 +--- >

[ovs-dev] [PATCH] Declare the version as "1.4.6".

Signed-off-by: Justin Pettit --- NEWS |4 configure.ac |2 +- debian/changelog |6 ++ 3 files changed, 11 insertions(+), 1 deletions(-) diff --git a/NEWS b/NEWS index a318b15..54bd50a 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,7 @@ +v1.4.6 - 26 Feb 2013 +---

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

On Tue, Feb 26, 2013 at 11:10 AM, Ethan Jackson wrote: > Looks good to me with the following incremental which I've applied. I'll > merge > shortly. Thanks for doing this work Kyle. We should make sure that whatever patches are necessary for getting VXLAN to work again also get applied to bran

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

> I looked at dpif_port_add(), and if you don't set odp_port to UINT32_MAX > here, > the dpif_class->port_add() function will use some undefined value here to > assign the port number. This is the reason I made this change. Did I > misunderstand > this? Ah yes, I think you're right. This interfa

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

> Ah yes, I think you're right. This interface is confusing, I assumed that > assignment was just unnecessary. I'll send out a patch. > > Ah nevermind, I haven't merged it yet. Good catch, thanks. Ethan ___ dev mailing list dev@openvswitch.org http://

Re: [ovs-dev] [PATCH] Declare the version as "1.4.6".

On Tue, Feb 26, 2013 at 11:16:49AM -0800, Justin Pettit wrote: > Signed-off-by: Justin Pettit Looks good, thanks. ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

[ovs-dev] [PATCH] Set dates for 1.9.0 release.

This also sets the dates for 1.8.0, even though it was an internal-only release. Signed-off-by: Justin Pettit --- NEWS |7 --- debian/changelog |3 ++- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/NEWS b/NEWS index aa909e5..0d892a3 100644 --- a/NEWS +++

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

On Feb 26, 2013, at 1:19 PM, Jesse Gross wrote: > On Tue, Feb 26, 2013 at 11:10 AM, Ethan Jackson wrote: >> Looks good to me with the following incremental which I've applied. I'll >> merge >> shortly. Thanks for doing this work Kyle. > > We should make sure that whatever patches are necessar

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

On Feb 26, 2013, at 1:21 PM, Ethan Jackson wrote: > > > Ah yes, I think you're right. This interface is confusing, I assumed that > assignment was just unnecessary. I'll send out a patch. > > > Ah nevermind, I haven't merged it yet. Good catch, thanks. > > Ethan Cool, glad we caught it

Re: [ovs-dev] [PATCH] Set dates for 1.9.0 release.

On Tue, Feb 26, 2013 at 11:26:29AM -0800, Justin Pettit wrote: > This also sets the dates for 1.8.0, even though it was an internal-only > release. > > Signed-off-by: Justin Pettit OK, fair enough. ___ dev mailing list dev@openvswitch.org http://openvs

Re: [ovs-dev] [PATCH] Declare the version as "1.4.6".

On Feb 26, 2013, at 1:16 PM, Justin Pettit wrote: > Signed-off-by: Justin Pettit Looks good to me. ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

Re: [ovs-dev] [ovs-git] Open vSwitch: in-band: Use "internal" netdev type for local ports. (branch-1.10)

On Tue, Feb 26, 2013 at 02:26:02PM -0500, dev@openvswitch.org wrote: > This is an automated email from the git hooks/post-receive script. It was > generated because a ref change was pushed to the repository containing > the project "Open vSwitch". > > The branch, branch-1.10 has been created >

[ovs-dev] Bug#701758: openvswitch: new version upstream

Package: openvswitch Severity: wishlist Hi, there's two new versions of openvswitch which are 1.4.5 for LTS and 1.7.3 is the most recent release. Ubuntu also seems to have a even newer version that I guess is taken directly from GIT. I also see some other bugs reports from last year and it let me

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

I'm doing the backport now. Whats the name of Pravin's patch you're referring to? Is that on master, or still in review? Ethan On Tue, Feb 26, 2013 at 11:27 AM, Kyle Mestery (kmestery) < kmest...@cisco.com> wrote: > On Feb 26, 2013, at 1:19 PM, Jesse Gross wrote: > > On Tue, Feb 26, 2013 at

[ovs-dev] Bug#701758: openvswitch: new version upstream

On Tue, Feb 26, 2013 at 04:31:22PM -0300, Ernesto Domato wrote: > Hi, there's two new versions of openvswitch which are 1.4.5 for LTS and 1.7.3 > is the most recent release. Ubuntu also seems to have a even newer version > that > I guess is taken directly from GIT. > > I also see some other bugs

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

It's still in review. I think it's only necessary because it changes the name of the dest port attribute and there's no point in causing compatibility problems if we can avoid them. I think everything should be functional with the existing patches on master though. On Tue, Feb 26, 2013 at 11:36

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

On Feb 26, 2013, at 1:36 PM, Ethan Jackson wrote: > > I'm doing the backport now. Whats the name of Pravin's patch you're > referring to? Is that on master, or still in review? > Still in review, here is the title: [ovs-dev] [PATCH v2] Tunnel: Cleanup old tunnel infrastructure. > Ethan >

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

On Feb 26, 2013, at 1:39 PM, Jesse Gross wrote: > It's still in review. I think it's only necessary because it changes > the name of the dest port attribute and there's no point in causing > compatibility problems if we can avoid them. I think everything > should be functional with the existing

[ovs-dev] Bug#701760: openvswitch-switch: problem interacting with libvirt

Package: openvswitch-switch Version: 1.4.2+git20120612-9 Severity: important Hi, I'm using openvswitch with libvirt and it works as expected. The problem that I'm having is that if you reboot or shutdown the host server, openvswitch is going down before libvirt virual machines finish going down t

[ovs-dev] Bug#701758: openvswitch: new version upstream

On Tue, Feb 26, 2013 at 4:36 PM, Ben Pfaff wrote: > The package isn't orphaned. Good to hear this :-) > Debian is currently frozen for release, > and the outstanding bug reports aren't severe enough to warrant > breaking the freeze. > > After the release, we'll upload whatever is the current LTS

[ovs-dev] Bug#701758: openvswitch: new version upstream

On Tue, Feb 26, 2013 at 04:49:46PM -0300, Ernesto Domato wrote: > On Tue, Feb 26, 2013 at 4:36 PM, Ben Pfaff wrote: > > Debian is currently frozen for release, > > and the outstanding bug reports aren't severe enough to warrant > > breaking the freeze. > > > > After the release, we'll upload whate

Re: [ovs-dev] ovs-dev] [vxlan] ofproto: Create and delete tnl_backers in type_run()

On Tue, Feb 26, 2013 at 11:41 AM, Kyle Mestery (kmestery) wrote: > On Feb 26, 2013, at 1:39 PM, Jesse Gross > wrote: >> It's still in review. I think it's only necessary because it changes >> the name of the dest port attribute and there's no point in causing >> compatibility problems if we can

[ovs-dev] [PATCH] netdev-linux: Fix netdev_linux_send() return value in corner case.

A negative 'sock' means there was an error but netdev_linux_send() returns a positive errno value on error. Signed-off-by: Ben Pfaff --- lib/netdev-linux.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c index 4ea7a36..c0e0c40 100

[ovs-dev] Bug#701758: openvswitch: new version upstream

On Tue, Feb 26, 2013 at 5:23 PM, Ben Pfaff wrote: > On Tue, Feb 26, 2013 at 04:49:46PM -0300, Ernesto Domato wrote: >> Yeah, I know the Wheezy is already frozen, but I thought that maybe >> the last version in Sid could be nice try it. > > The normal path to a getting a bug fix for a severe bug in

Re: [ovs-dev] [Patch] datapath: fix checksum after putting vlan tag

On Tue, Feb 26, 2013 at 1:46 AM, Cong Wang wrote: > After inserting the VLAN tag with __vlan_put_tag(), we should > re-calculate the skb checksum if it is done. > > Cc: Jesse Gross > Signed-off-by: Cong Wang This code path is supposed to be emulate dev_hard_start_xmit() on older kernels. Howev

[ovs-dev] Bug#701758: openvswitch: new version upstream

On Tue, Feb 26, 2013 at 05:43:49PM -0300, Ernesto Domato wrote: > On Tue, Feb 26, 2013 at 5:23 PM, Ben Pfaff wrote: > > On Tue, Feb 26, 2013 at 04:49:46PM -0300, Ernesto Domato wrote: > >> Yeah, I know the Wheezy is already frozen, but I thought that maybe > >> the last version in Sid could be nic

Re: [ovs-dev] [PATCH 10/15] datapath: Add basic MPLS support to kernel

On Fri, Feb 15, 2013 at 1:56 AM, Simon Horman wrote: > Allow datapath to recognize and extract MPLS labels into flow keys > and execute actions which push, pop, and set labels on packets. > > Based heavily on work by Leo Alterman and Ravi K. > > Cc: Ravi K > Cc: Leo Alterman > Reviewed-by: Isaku

Re: [ovs-dev] [PATCH] netdev-linux: Fix netdev_linux_send() return value in corner case.

Looks good. --Justin On Feb 26, 2013, at 12:35 PM, Ben Pfaff wrote: > A negative 'sock' means there was an error but netdev_linux_send() returns > a positive errno value on error. > > Signed-off-by: Ben Pfaff > --- > lib/netdev-linux.c |2 +- > 1 files changed, 1 insertions(+), 1 deletion

Re: [ovs-dev] log flooded with: cannot get STP status on nonexistent port 7

On Feb 26, 2013, at 11:09 AM, Ben Pfaff wrote: > You mean 1.4.6 also? That seems reasonable, we've built up a few bug > fixes since 1.4.5. > > I applied this to branch-1.4 We released 1.4.6, which has this fix. --Justin ___ dev mailing list dev@o

Re: [ovs-dev] branch-1.7 and kernel 3.4

Just in case you missed it, we released 1.9.0. --Justin On Feb 26, 2013, at 9:26 AM, Tomasz Paszkowski wrote: > Great :-) So I'll focus on 1.9 ! > > > > On Tue, Feb 26, 2013 at 6:19 PM, Justin Pettit wrote: >> If it helps, we're planning to release OVS 1.9 in the next day or two, which >>

Re: [ovs-dev] [PATCH v3] Tunnel: Cleanup old tunnel infrastructure.

On Tue, Feb 26, 2013 at 1:39 PM, Pravin B Shelar wrote: > diff --git a/datapath/tunnel.c b/datapath/tunnel.c > index a05cf54..47cecb3 100644 > --- a/datapath/tunnel.c > +++ b/datapath/tunnel.c > @@ -543,6 +283,9 @@ int ovs_tnl_send(struct vport *vport, struct sk_buff *skb) > u8 ttl; >

Re: [ovs-dev] [PATCH v3] Tunnel: Cleanup old tunnel infrastructure.

On Tue, Feb 26, 2013 at 3:55 PM, Jesse Gross wrote: > On Tue, Feb 26, 2013 at 1:39 PM, Pravin B Shelar wrote: >> diff --git a/datapath/tunnel.c b/datapath/tunnel.c >> index a05cf54..47cecb3 100644 >> --- a/datapath/tunnel.c >> +++ b/datapath/tunnel.c >> @@ -543,6 +283,9 @@ int ovs_tnl_send(struct

Re: [ovs-dev] [PATCH v3] Tunnel: Cleanup old tunnel infrastructure.

On Tue, 26 Feb 2013 15:55:45 -0800 Jesse Gross wrote: > > + if (!OVS_CB(skb)->tun_key) > > Can you add an unlikely() here? > > > + goto error_free; As far as I know goto's are always considered unlikely by Gcc. ___ dev mailing l

Re: [ovs-dev] [PATCH v4] Tunnel: Cleanup old tunnel infrastructure.

On Tue, Feb 26, 2013 at 2:14 PM, Pravin B Shelar wrote: > diff --git a/datapath/tunnel.c b/datapath/tunnel.c > index a05cf54..29752df 100644 > --- a/datapath/tunnel.c > +++ b/datapath/tunnel.c > @@ -543,15 +247,8 @@ int ovs_tnl_send(struct vport *vport, struct sk_buff > *skb) > u8 ttl; >

[ovs-dev] [CudaMailTagged] Last chance... Your invitation from Apurva Patil is about to expire

Re: [ovs-dev] [Patch] datapath: fix checksum after putting vlan tag

On Tue, 2013-02-26 at 12:51 -0800, Jesse Gross wrote: > On Tue, Feb 26, 2013 at 1:46 AM, Cong Wang wrote: > > After inserting the VLAN tag with __vlan_put_tag(), we should > > re-calculate the skb checksum if it is done. > > > > Cc: Jesse Gross > > Signed-off-by: Cong Wang > > This code path is

[ovs-dev] [PATCH] Add Nicira vendor extension actions NXAST_STACK_PUSH/POP

nicira-ext: Add Nicira actions NXAST_STACK_PUSH and NXAST_STACK_POP. The Push action takes a single parameter. Any source allowed by NXAST_REG_MOVE is allowed to be pushed onto the stack. When the source is a bit field, its value will be right shifted to bit zero before being pushed onto the s

[ovs-dev] [PATCH 1/2] ofproto-dpif: Always maintain subfacet key.

Due to flow based tunneling, we can no longer assume that it's possible to reconstruct a subfacet's key from its facet's flow. The flow's in_port may be stale due to tunnel configuration changes. Signed-off-by: Ethan Jackson --- ofproto/ofproto-dpif.c | 83 ++---

[ovs-dev] [PATCH 2/2] ofproto-dpif: Handle tunnel config changes in facet_revalidate().

For most of the history of Open vSwitch, one could assume that a given datapath flow key would consistently translate into the same userspace struct flow representation. However, with the switch to flow based tunneling, we now have a situation where a database configuration change can cause a data

[ovs-dev] Bug#701760: Bug#701760: openvswitch-switch: problem interacting with libvirt

Hey, I have also gone through this problem in my early days with open vswitch. In this condition, you can do one thing: Just create a init script for open vswitch and add it into the services, from init script you can handle the start/stop priority for open vswitch server during restart/reboot of

Re: [ovs-dev] [PATCH 10/15] datapath: Add basic MPLS support to kernel

On Tue, Feb 26, 2013 at 02:49:47PM -0800, Jesse Gross wrote: > On Fri, Feb 15, 2013 at 1:56 AM, Simon Horman wrote: > > Allow datapath to recognize and extract MPLS labels into flow keys > > and execute actions which push, pop, and set labels on packets. > > > > Based heavily on work by Leo Alterm

[ovs-dev] [PATCH] nx-match: Correct writing of value and length in set_field_to_ofast()

ofpbuf_put_* may reallocate the underlying buffer of the ofpbuf and thus writing data after a ofpbuf_put_* call must write to memory relative to the pointer returned by the call. Prior to this change the length and trailing value would not be written to the set_field action if ofpbuf_put_* may rea