Re: [ovs-dev] [PATCH] [ovn-controller] [RFC] Poor man's lflow incremental processing

"dev" wrote on 02/05/2016 05:09:44 PM: > From: Ryan Moats/Omaha/IBM@IBMUS > To: Ben Pfaff > Cc: dev@openvswitch.org > Date: 02/05/2016 05:10 PM > Subject: Re: [ovs-dev] [PATCH] [ovn-controller] [RFC] Poor man's > lflow incremental processing > Sent

[ovs-dev] [PATCH 2/3] dpif-netdev: Move rxq management into functions.

Signed-off-by: Ilya Maximets --- lib/dpif-netdev.c | 163 +++--- 1 file changed, 105 insertions(+), 58 deletions(-) diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c index c477f2b..e62f5f6 100644 --- a/lib/dpif-netdev.c

Re: [ovs-dev] [PATCH] manpages.mk: remove ovs-benchmark.

On Mon, Feb 08, 2016 at 10:31:48AM +0300, Ilya Maximets wrote: > Fixes: e75291417990 ("ovs-benchmark: Remove.") > Signed-off-by: Ilya Maximets Applied, thanks. ___ dev mailing list dev@openvswitch.org

Re: [ovs-dev] [PATCH 1/1] ovn/utilities: Separating OVN NB and SBdatabases. Run dbs in separate

Russell Bryant wrote on 02/08/2016 10:24:29 AM: > From: Russell Bryant > To: Ryan Moats/Omaha/IBM@IBMUS, Michael > Cc: dev@openvswitch.org > Date: 02/08/2016 10:24 AM > Subject: Re: [ovs-dev] [PATCH 1/1] ovn/utilities:

Re: [ovs-dev] [PATCH 1/1] ovn/utilities: Separating OVN NB and SBdatabases. Run dbs in separate

For devstack, we need a few more things: 1) the ability to start and stop the ovsdb processes separate from ovn-northd, so this patch needs to break them out as separate commands. 2) the logfile settings for vconsole and vfile need to use environment variables so that we can control their level

Re: [ovs-dev] [PATCH 1/1] ovn/utilities: Separating OVN NB and SBdatabases. Run dbs in separate

On 02/08/2016 11:17 AM, Ryan Moats wrote: > For devstack, we need a few more things: Context: devstack is an OpenStack development environment. Our OVN integration pre-dated ovn-ctl, but we'd like to start using it. > 1) the ability to start and stop the ovsdb processes separate from >

Re: [ovs-dev] [ovs-discuss] Somebody making --user and dpdk compatible again?

Quoting Ansis Atteka (ansisatt...@gmail.com): > On 2 February 2016 at 23:33, Serge Hallyn wrote: > > > Quoting Ansis Atteka (ansisatt...@gmail.com): > > > On 29 January 2016 at 12:10, Serge Hallyn > > wrote: > > > > Sorry I've not really had

Re: [ovs-dev] [PATCH 1/1] ovn/utilities: Separating OVN NB and SB databases. Run dbs in separate

The Subject turns into the first line of the commit message. It looks like it got cut off. It's a bit long anyway. Ideally it should be kept to 72 characters or so. On 02/08/2016 10:41 AM, Michael wrote: > OVN NB & SB DB's should be separated and should run with ovn-ctl start_northd > /

Re: [ovs-dev] [PATCH V2] Add tunnel key to Physical_Locator

Thanks, applied to master. On Sun, Feb 07, 2016 at 06:29:32AM +, Ofer Ben-Yacov wrote: > It does. > > Thanks, > Ofer. > > ‫בתאריך יום ו׳, 5 בפבר׳ 2016 ב-22:57 מאת ‪Ben Pfaff‬‏ <‪b...@ovn.org‬‏>:‬ > > > On Thu, Feb 04, 2016 at 09:49:35AM +, Ofer Ben-Yacov wrote: > > > The tunnel key in

Re: [ovs-dev] [OVS-dev]: OVN: RFC re: logical and physical endpoint separation proposal

Hi Russell Thanks for your comments Darrell On 2/5/16, 12:23 PM, "Russell Bryant" wrote: >On 01/29/2016 09:36 PM, Darrell Lu wrote: >> The following is a proposal regarding how to allow logical >> and physical endpoint contexts for OVN to stay separated, from >> a CMS POV.

[ovs-dev] [PATCH] build: Remove ovs-benchmark from manpages.mk.

Fixes: e75291417990 ("ovs-benchmark: Remove.") Signed-off-by: Russell Bryant --- manpages.mk | 6 -- 1 file changed, 6 deletions(-) diff --git a/manpages.mk b/manpages.mk index 6141242..bf96e9f 100644 --- a/manpages.mk +++ b/manpages.mk @@ -110,12 +110,6 @@

Re: [ovs-dev] [PATCH] build: Remove ovs-benchmark from manpages.mk.

On Mon, Feb 08, 2016 at 11:21:31AM -0500, Russell Bryant wrote: > Fixes: e75291417990 ("ovs-benchmark: Remove.") > Signed-off-by: Russell Bryant Oops, I already applied Ilya's version of this patch. ___ dev mailing list

Re: [ovs-dev] [PATCH v3] netdev_dpdk.c: Add QoS functionality.

> -Original Message- > From: Ben Pfaff [mailto:b...@ovn.org] > Sent: Friday, February 05, 2016 9:19 PM > To: Stokes, Ian > Cc: dev@openvswitch.org > Subject: Re: [ovs-dev] [PATCH v3] netdev_dpdk.c: Add QoS functionality. > > On Thu, Feb 04, 2016 at 09:30:59AM +, Stokes, Ian wrote: > >

Re: [ovs-dev] [PATCH v3] netdev_dpdk.c: Add QoS functionality.

> -Original Message- > From: Daniele Di Proietto [mailto:diproiet...@vmware.com] > Sent: Friday, February 05, 2016 9:45 PM > To: Ben Pfaff; Stokes, Ian > Cc: dev@openvswitch.org > Subject: Re: [ovs-dev] [PATCH v3] netdev_dpdk.c: Add QoS functionality. > > > > On 05/02/2016 13:18, "Ben

Re: [ovs-dev] [PATCH 1/3] ovn: ovn-controller changes for qos settings

The qos settings are managed using the 'options' fields in the "Port_Binding" table. Signed-off-by: Babu Shanmugam This seems more complicated than necessary. In the SBREC_PORT_BINDING_FOR_EACH loop in binding_run(), I think that we only need to be able to find the

Re: [ovs-dev] [PATCH V3] netdev-dpdk: Add Jumbo Frame Support.

> >Hi, > >Some comments below. > Hi Flavio, Thanks for your review! I've responded to your comments inline - I've also addressed the cosmetic changes that you suggested in V3 of the patch. Thanks again, Mark > >On Mon, 1 Feb 2016 10:18:54 + >Mark Kavanagh

Re: [ovs-dev] [PATCH V3] netdev-dpdk: Add Jumbo Frame Support.

> >Hi Mark, > Hi Daniele, Thanks for the review! Responses inline below. Cheers, Mark >This patch besides adding Jumbo Frame support also cleans up >the mbuf initialization (by changing the macros, adding >dpdk_buf_size, and rewriting __ovs_rte_pktmbuf_init), so thanks >for this. I think it

Re: [ovs-dev] [PATCH] CONTRIBUTING: Strengthen testing guidelines

On 02/08/2016 06:18 PM, Justin Pettit wrote: > >> On Feb 8, 2016, at 10:54 AM, Russell Bryant wrote: >> >> On 02/08/2016 01:03 PM, Justin Pettit wrote: >> >> There are solutions to this problem, but it requires letting a CI system >> do the merging of patches for you. >> >>> The

Re: [ovs-dev] [PATCH 2/3] sset: Make sset iteration check the type of the nodes it's iterating.

On 02/08/2016 08:45 PM, Russell Bryant wrote: > On 02/08/2016 07:52 PM, Ben Pfaff wrote: >> Without this additional check, SSET_FOR_EACH is happy to iterate over any >> type that has a struct hmap member named 'map'. We have a number of these: >> hmapx, shash, simap, and smap, in addition to

[ovs-dev] [PATCH] netdev: Free packets in netdev_send() for devices that don't support send.

This manifested as a memory leak in test 898 "ofproto-dpif - sFlow packet sampling - tunnel set", which included an output to a tunnel vport that doesn't have an implementation of netdev_send(). Reported-by: William Tu Reported-at:

Re: [ovs-dev] netdev-dummy memory leak: ofproto-dpif - sFlow packet sampling - tunnel set

On Mon, Feb 08, 2016 at 10:33:39AM -0800, William Tu wrote: > I spent some time fixing this memory leak but couldn't get it right. Please > give me some advise how to fix it. Thank you. I figured it out. It was tricky, and a genuine bug. See:

[ovs-dev] [PATCH 2/2] dist-docs: Install plaintext files properly.

We shouldn't try to treat every file as Markdown. Signed-off-by: Ben Pfaff --- build-aux/dist-docs | 25 - 1 file changed, 20 insertions(+), 5 deletions(-) diff --git a/build-aux/dist-docs b/build-aux/dist-docs index 1aeefa7..93709c8 100755 ---

[ovs-dev] [PATCH 1/2] dist-docs: Properly pass "make" options into sub-make.

In particular this ensures that the sub-make that dist-docs runs has the benefit of any parallelism flags passed into the top-level make. Signed-off-by: Ben Pfaff --- Makefile.am | 2 +- build-aux/dist-docs | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff

Re: [ovs-dev] [PATCH 2/3] sset: Make sset iteration check the type of the nodes it's iterating.

On Mon, Feb 08, 2016 at 08:46:44PM -0500, Russell Bryant wrote: > On 02/08/2016 08:45 PM, Russell Bryant wrote: > > On 02/08/2016 07:52 PM, Ben Pfaff wrote: > >> Without this additional check, SSET_FOR_EACH is happy to iterate over any > >> type that has a struct hmap member named 'map'. We have

Re: [ovs-dev] [PATCH] CONTRIBUTING: Strengthen testing guidelines

> On Feb 8, 2016, at 5:38 PM, Russell Bryant wrote: > > On 02/08/2016 06:18 PM, Justin Pettit wrote: >> >> The "shame" comment was meant tongue-in-cheek, but I didn't want to >> overwhelm the message with smiley faces. Sorry that didn't come >> through. > > Ah, well I'm

Re: [ovs-dev] [OVN] Applying ACL changes to existing connections

On Fri, Feb 5, 2016 at 1:30 PM, Russell Bryant wrote: > > On 02/05/2016 02:22 AM, Justin Pettit wrote: > > Joe and I spent some time today discussing our options. This is > > pretty tricky to get right and most of the options that come > > immediately to mind have subtle corner

Re: [ovs-dev] ovsdb-server servicing large number of idle connections

On Mon, Feb 8, 2016 at 3:44 PM, Andy Zhou wrote: > > Hi, Han, > > This is to follow up on the Thursday OVN IRC chat. After the meeting, I was able replicate the test condition you mentioned where a single ovsdb-server services 1000 ovsdb-client connections. When there is no

Re: [ovs-dev] [PATCH 1/3] ovn: ovn-controller changes for qos settings

Ben, I tested your solution and it works well. It is in fact simpler than my approach. I will squash the other two commits along with this for v2. Thank you, Babu On Tuesday 09 February 2016 05:44 AM, Ben Pfaff wrote: On Mon, Feb 08, 2016 at 03:20:11PM +0530, Babu Shanmugam wrote: The qos

Re: [ovs-dev] [PATCH 1/3] util: New macro BUILD_ASSERT_TYPE.

On Mon, Feb 8, 2016 at 4:52 PM, Ben Pfaff wrote: > I intend to make further use of this in upcoming commits. > > Signed-off-by: Ben Pfaff > --- > lib/util.h | 20 ++-- > 1 file changed, 18 insertions(+), 2 deletions(-) > This is clever! Acked-by:

Re: [ovs-dev] [PATCH 3/3] hmap: Add extra build-time iteration checks for types derived from hmap.

On Mon, Feb 08, 2016 at 04:52:45PM -0800, Ben Pfaff wrote: > Some of our data structures derived from hmap use the same member names. > This means it's possible to confuse them in iteration, e.g. to iterate a > shash with SIMAP_FOR_EACH. Of course this will crash at runtime, but it > seems even

Re: [ovs-dev] [PATCH 1/3] ovn: ovn-controller changes for qos settings

That sounds great. I like simple. On Tue, Feb 09, 2016 at 08:19:57AM +0530, Babu Shanmugam wrote: > Ben, I tested your solution and it works well. It is in fact simpler than my > approach. I will squash the other two commits along with this for v2. > > > Thank you, > Babu > > On Tuesday 09

Re: [ovs-dev] [PATCH monitor_cond V3 01/10] ovsdb: create column index mapping between ovsdb row to monitor row

Andy Zhou wrote on 09/02/2016 01:33:12 AM: > On Mon, Feb 8, 2016 at 12:43 PM, Liran Schour wrote: > Andy Zhou wrote on 08/02/2016 09:41:47 PM: > > > On Sun, Feb 7, 2016 at 12:08 AM, Liran Schour wrote: > > Andy Zhou

Re: [ovs-dev] [PATCH 1/3] util: New macro BUILD_ASSERT_TYPE.

On Mon, Feb 08, 2016 at 06:49:33PM -0800, Andy Zhou wrote: > On Mon, Feb 8, 2016 at 4:52 PM, Ben Pfaff wrote: > > > I intend to make further use of this in upcoming commits. > > > > Signed-off-by: Ben Pfaff > > --- > > lib/util.h | 20 ++-- > > 1

Re: [ovs-dev] [PATCH 2/3] sset: Make sset iteration check the type of the nodes it's iterating.

On 02/08/2016 07:52 PM, Ben Pfaff wrote: > Without this additional check, SSET_FOR_EACH is happy to iterate over any > type that has a struct hmap member named 'map'. We have a number of these: > hmapx, shash, simap, and smap, in addition to sset. > > I guess another approach would be give each

Re: [ovs-dev] [PATCH 3/3] hmap: Add extra build-time iteration checks for types derived from hmap.

On Mon, Feb 8, 2016 at 4:52 PM, Ben Pfaff wrote: > Some of our data structures derived from hmap use the same member names. > This means it's possible to confuse them in iteration, e.g. to iterate a > shash with SIMAP_FOR_EACH. Of course this will crash at runtime, but it > seems

Re: [ovs-dev] [PATCH 1/2] dist-docs: Include Documentation/ in dist-docs.

On Mon, Feb 08, 2016 at 12:25:13PM -0500, Russell Bryant wrote: > Signed-off-by: Russell Bryant Thanks. I noticed two other issues while testing this, so I sent patches for those too: https://patchwork.ozlabs.org/patch/580625/

[ovs-dev] [PATCH 3/3] dpif-netdev: Destroy pmd threads only if pmd-cpu-mask changed.

Since 5f2ccb1c0d3b ("dpif: Allow adding ukeys for same flow by different pmds.") there is the possibility to reassign queues among pmd threads without restarting them and deleting the megaflow cache. So, reconfiguration can be performed without destroying of PMD threads if pmd-cpu-mask not

[ovs-dev] [PATCH 1/1] ovn/utilities: Separating OVN NB and SB databases. Run dbs in separate

OVN NB & SB DB's should be separated and should run with ovn-ctl start_northd / stop_northd From 351cb81298c5b5393427958d1663cf1d3ddb370b Mon Sep 17 00:00:00 2001 From: Michael Arnaldi Date: Mon, 8 Feb 2016 16:29:26 +0100 Subject: [PATCH 1/1] ovn/utilities: Separating OVN NB

Re: [ovs-dev] [PATCH v8 0/5] Convert DPDK configuration from command line to DB based

Aaron Conole writes: > Apologies for the top post on this one. > > I am going to rebase and resubmit this series in 14 - 18 hours. At that > time I am doing the following: > > Changing Kevin and Panu's Cc: lines to ACKs > Adding: > Tested-by: RobertX Wojciechowicz

[ovs-dev] [PATCH 0/3] rxq management enchancement

Ilya Maximets (3): dpif-netdev: Reload each thread only once in do_add_port. dpif-netdev: Move rxq management into functions. dpif-netdev: Destroy pmd threads only if pmd-cpu-mask changed. lib/dpif-netdev.c | 215 +- 1 file changed, 147

Re: [ovs-dev] [PATCH] dpif-netdev: Reload each thread only once in do_add_port.

Resent as a part of patch-set: http://openvswitch.org/pipermail/dev/2016-February/065846.html On 04.02.2016 16:41, Ilya Maximets wrote: > While adding of pmd interface with multiple queues several queues > may be assigned to one thread and this thread will be reloaded > one time for each added

[ovs-dev] [PATCH 1/3] dpif-netdev: Reload each thread only once in do_add_port.

While adding of pmd interface with multiple queues several queues may be assigned to one thread and this thread will be reloaded one time for each added queue. Signed-off-by: Ilya Maximets --- lib/dpif-netdev.c | 10 ++ 1 file changed, 10 insertions(+) diff

[ovs-dev] [PATCH] Separating ovn nb databases

Title fixed and updated ovn-northd to the new db paths. From 9052644fd5ed07583540292fcd0cb5ddc47d633e Mon Sep 17 00:00:00 2001 From: Michael Arnaldi Date: Mon, 8 Feb 2016 18:56:51 +0100 Subject: [PATCH] Separating ovn nb databases Signed-off-by: Michael Arnaldi

Re: [ovs-dev] [PATCH] CONTRIBUTING: Strengthen testing guidelines

> On Feb 8, 2016, at 9:34 AM, Kavanagh, Mark B > wrote: > >> I think the example that you show is actually a pretty good example of >> how quickly we catch problems. > > Sure. At the same time though, this is a reactive approach, rather than a > proactive one,

[ovs-dev] netdev-dummy memory leak: ofproto-dpif - sFlow packet sampling - tunnel set

Hi, I spent some time fixing this memory leak but couldn't get it right. Please give me some advise how to fix it. Thank you. Testcase: 898: ofproto-dpif.at:5172 ofproto-dpif - sFlow packet sampling - tunnel set Call stacks: malloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)

[ovs-dev] [PATCH 2/2] dist-docs: Fix relative links.

There are some relative links designed to work on github. These links are broken within dist-docs. Adding a symbolic link from the original filename to the plain text version makes these links work within dist-docs. Signed-off-by: Russell Bryant --- build-aux/dist-docs | 1 +

[ovs-dev] [PATCH 1/2] dist-docs: Include Documentation/ in dist-docs.

Signed-off-by: Russell Bryant --- Documentation/automake.mk | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/automake.mk b/Documentation/automake.mk index 9b10df7..5903c22 100644 --- a/Documentation/automake.mk +++ b/Documentation/automake.mk @@

Re: [ovs-dev] [PATCH] [ovn-controller] [RFC] Poor man's lflow incremental processing

On Mon, Feb 08, 2016 at 10:34:05AM -0600, Ryan Moats wrote: > > > I think this is because lflow_run() marks a logical datapath "clean" > > > after processing any one logical flow in the datapath. I don't know > why > > > it does that, but on the face of it it seems wrong. > > > > > > > I'll

Re: [ovs-dev] [PATCH] build: Remove ovs-benchmark from manpages.mk.

On 02/08/2016 11:30 AM, Ben Pfaff wrote: > On Mon, Feb 08, 2016 at 11:21:31AM -0500, Russell Bryant wrote: >> Fixes: e75291417990 ("ovs-benchmark: Remove.") >> Signed-off-by: Russell Bryant > > Oops, I already applied Ilya's version of this patch. > No worries, I didn't see

[ovs-dev] [PATCH] ovn-controller: Assign zone-ids consistently

Currently, conntrack zone-id is assigned to lport by ovn-controller, but the ovn-controller does not remember what was earlier assigned to the same lport (possibly in an earlier run across restart). So, after ovn-controller restart, the zone-ids may get set inconsistently on lports,

Re: [ovs-dev] [RFC] ovn-controller: Only process lflows for local datapaths.

On Wed, Feb 03, 2016 at 02:15:48PM -0600, Ryan Moats wrote: > I've talked to Kyle about this (as it's becoming a blocker for our scale > testing) and so I'm willing to spend some time on the following "poor > man's" approach: > > (1) Leave the current ovsdb-server process on the default port to

Re: [ovs-dev] [PATCH] CONTRIBUTING: Strengthen testing guidelines

> >On Mon, Jan 25, 2016 at 03:15:23PM +, Mark Kavanagh wrote: >> Current testing guidelines do not take account of two distinct >> OVS builds: >>a) 'standard' OVS >>b) OVS with DPDK integration >> >> It is critical that all patches are tested against both builds; if not, >> there is a

[ovs-dev] PacketOut support in OF Bundles

Hi. From the ovs design page, bundles only support FlowMod and PortMod messages, right? I would like to propose an extension to ovs to support the use of PacketOut messages in OpenFlow Bundles. I think PacketOut in bundles would be a good addition because it allows OF controllers/applications to

Re: [ovs-dev] [PATCH monitor_cond V3 01/10] ovsdb: create column index mapping between ovsdb row to monitor row

Andy Zhou wrote on 08/02/2016 09:41:47 PM: > On Sun, Feb 7, 2016 at 12:08 AM, Liran Schour wrote: > Andy Zhou wrote on 04/02/2016 11:54:06 PM: > > > On Sun, Jan 31, 2016 at 11:03 PM, Liran Schour > wrote: > > Columns

Re: [ovs-dev] [PATCH] Separating ovn nb databases

I will do this stuff tomorrow, I will take the way ovsdb is started using ovs-ctl and refactor the logic in a shared library. I was also thinking about the database issue for HA and I’m currently thinking why not to leverage the current ecosystem of distributed dbs. In specific why not to make

Re: [ovs-dev] [PATCH] Separating ovn nb databases

On 02/08/2016 04:09 PM, Michael wrote: > I will do this stuff tomorrow, I will take the way ovsdb is started > using ovs-ctl and refactor the logic in a shared library. > What I will do for start is make some flags to have the option of > running dbs separately. OK thanks, > I was also thinking

Re: [ovs-dev] [PATCH] ovn-controller: Assign zone-ids consistently

On 02/08/2016 02:00 PM, Ramu Ramamurthy wrote: > > Resend (apologies) as the mailer mangled spaces on the patch. Can you try posting with the git-send-email utility? -- Russell Bryant ___ dev mailing list dev@openvswitch.org

Re: [ovs-dev] [OVS-dev]: OVN: RFC re: logical and physical endpoint separation proposal

On 02/08/2016 12:05 PM, Darrell Ball wrote: > On 2/5/16, 12:23 PM, "Russell Bryant" wrote: >> I agree with this sort of separation in principle. Some specific >> examples would help me understand the proposal, though. You mention >> that this applies to both localnet and

Re: [ovs-dev] [PATCH] [ovn-controller] [RFC] Poor man's lflow incremental processing

Aha! that makes a lot more sense, but unfortunately, I don't think I'm grokking the code yet: 1. When using the baseling of SBREC_LOGICAL_FLOW_FOR_EACH, at startup, each lflow_run pass handles 12 flows and 12 matches... 2. When I tried using ovsdb_idl_track_add_all and

Re: [ovs-dev] [PATCH] Separating ovn nb databases

On 02/08/2016 01:02 PM, Michael wrote: > Title fixed and updated ovn-northd to the new db paths. Some minor tips on conventions for emailing patches: When you send another revision, you don't need to specify what changed in the commit message itself. Notes about what has changed can go after

Re: [ovs-dev] [PATCH 1/1] ovn/utilities: Separating OVN NB and SBdatabases. Run dbs in separate

I will now revert only in the following points: > How about using $rundir/ovnnb_db.sock and $rundir/ovnsb_db.sock instead > of the new directory? For High Availability we need to have the dbs on a directory that we can easily share between hosts. We can do this by default or make it optional

Re: [ovs-dev] [PATCH] CONTRIBUTING: Strengthen testing guidelines

On 02/08/2016 01:03 PM, Justin Pettit wrote: > >> On Feb 8, 2016, at 9:34 AM, Kavanagh, Mark B >> wrote: >> >>> I think the example that you show is actually a pretty good example of >>> how quickly we catch problems. >> >> Sure. At the same time though, this is a

Re: [ovs-dev] [PATCH] ovn-controller: Assign zone-ids consistently

Resend (apologies) as the mailer mangled spaces on the patch. Signed-off-by: Ramu Ramamurthy --- ovn/controller/binding.c | 70 +++- 1 file changed, 63 insertions(+), 7 deletions(-) diff --git

Re: [ovs-dev] [PATCH 2/3] ofproto-dpif-xlate: Don't consider mirrors used when excluded by VLAN.

> On Feb 5, 2016, at 7:27 PM, Ben Pfaff wrote: > > On Fri, Feb 05, 2016 at 04:41:29PM -0800, Jarno Rajahalme wrote: >> >>> On Feb 5, 2016, at 3:30 PM, Ben Pfaff wrote: >>> >>> Mirroring is supposed to happen at most once for any destination on a given >>> packet,

Re: [ovs-dev] [PATCH monitor_cond V3 01/10] ovsdb: create column index mapping between ovsdb row to monitor row

On Sun, Feb 7, 2016 at 12:08 AM, Liran Schour wrote: > Andy Zhou wrote on 04/02/2016 11:54:06 PM: > > > On Sun, Jan 31, 2016 at 11:03 PM, Liran Schour > wrote: > > Columns indexing is different in ovsdb_row then in ovsdb_monitor_row. > > We

[ovs-dev] Rackspace Users

Hi, Would you be interested in all *Rackspace *users for your email campaigns? We can provide you Database from North America, Latin America, EMEA, and APAC. *Information Fields* – Name, Title, Email, Phone Numbers, Company Name, and Company Details like Physical Address, Web Address,

Re: [ovs-dev] [PATCH] ovn-controller: Assign zone-ids consistently

On 2016-02-08 11:12, Russell Bryant wrote: On 02/08/2016 02:00 PM, Ramu Ramamurthy wrote: Resend (apologies) as the mailer mangled spaces on the patch. Can you try posting with the git-send-email utility? I am posting again from a better behaved mailer

[ovs-dev] [PATCH] - ovn-controller: Assign zone-ids consistently

Currently, conntrack zone-id is assigned to lport by ovn-controller, but the ovn-controller does not remember what was earlier assigned to the same lport (possibly in an earlier run across restart). So, after ovn-controller restart, the zone-ids may get set inconsistently on lports, resulting

Re: [ovs-dev] [PATCH 1/3] ovn: ovn-controller changes for qos settings

On Mon, Feb 08, 2016 at 03:20:11PM +0530, Babu Shanmugam wrote: > >>The qos settings are managed using the 'options' fields in the > >>"Port_Binding" table. > >> > >>Signed-off-by: Babu Shanmugam > >This seems more complicated than necessary. In the >

Re: [ovs-dev] [PATCH] CONTRIBUTING: Strengthen testing guidelines

> On Feb 8, 2016, at 10:54 AM, Russell Bryant wrote: > > On 02/08/2016 01:03 PM, Justin Pettit wrote: > > There are solutions to this problem, but it requires letting a CI system > do the merging of patches for you. > >> The Linux kernel does something similar. >> Shame is

Re: [ovs-dev] [PATCH monitor_cond V3 01/10] ovsdb: create column index mapping between ovsdb row to monitor row

On Mon, Feb 8, 2016 at 12:43 PM, Liran Schour wrote: > Andy Zhou wrote on 08/02/2016 09:41:47 PM: > > > On Sun, Feb 7, 2016 at 12:08 AM, Liran Schour wrote: > > Andy Zhou wrote on 04/02/2016 11:54:06 PM: > > > > > On Sun, Jan

[ovs-dev] ovsdb-server servicing large number of idle connections

Hi, Han, This is to follow up on the Thursday OVN IRC chat. After the meeting, I was able replicate the test condition you mentioned where a single ovsdb-server services 1000 ovsdb-client connections. When there is no ovsdb transactions going on, large number of CPU cycles are used ( on my

[ovs-dev] [PATCH 3/3] hmap: Add extra build-time iteration checks for types derived from hmap.

Some of our data structures derived from hmap use the same member names. This means it's possible to confuse them in iteration, e.g. to iterate a shash with SIMAP_FOR_EACH. Of course this will crash at runtime, but it seems even better to catch it at compile time. An alternative would be to use

[ovs-dev] [PATCH 1/3] util: New macro BUILD_ASSERT_TYPE.

I intend to make further use of this in upcoming commits. Signed-off-by: Ben Pfaff --- lib/util.h | 20 ++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/lib/util.h b/lib/util.h index 0ad75f2..afd8e37 100644 --- a/lib/util.h +++ b/lib/util.h @@

[ovs-dev] [PATCH 2/3] sset: Make sset iteration check the type of the nodes it's iterating.

Without this additional check, SSET_FOR_EACH is happy to iterate over any type that has a struct hmap member named 'map'. We have a number of these: hmapx, shash, simap, and smap, in addition to sset. I guess another approach would be give each of these a unique member name, but the short name