[ovs-discuss] Is it possible to create ovs-bridge without MAC address ?
Hi, I am using openvswitch 2.3.0 (on Debian 8) to create VLAN(s) and bridges for Xen VM use. When I connect my computer to office network which has "sticky" bit set, the switch detects multiple MAC addresses (ovs-bridge + physical ethernet) on one port and blocked my access. Is it possible to create ovs-bridge without MAC address, just like physical L2 switch ? Thanks a lot. John Mok ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss
Re: [ovs-discuss] ovs-dpdk performance not stable
Hi Ian, Thank you for you answers! it is correct that i am using ovs-vsctl set Interface dpdk0 options:n_rxq=8 commands for the queues. Could you please expand on the sentence " It will be split between queues based on des tip so it’s important that test traffic varies if you want traffic to be dispersed evenly among the queues at this level." It might be a typo, or i might just not know what you mean by "des tip", could you please clarify for me? Additionally, what do you mean by varying the traffic? do you mean to somehow not have the packets at a constant frame rate? Regarding the Vhost user queues, i am using Openstack and i did not find yet a way to create multiple queues (i updated the image's metadata hw_vif_multiqueue_enabled=true) but i don't know how to set the queue amount especially that in the VM that i am running i do not have ethtool. Regarding the multiple queues while using one core for the PMD: i did get much better performance when i had two cores for the PMD, however, i am not at the luxury to be able to use two cores. It is puzzling for me that when i use multiple queues i do get better performance not enough but much better then when i use only one. I am sorry but this is a confusing for me. As for the core isolation, i have only core zero isolated for the kernel. i checked with htop and i saw that probably the emulatorpin of the VM might be running there so i moved it but it decreased performance. when i use only n_rxq and n_txq=1 i get performance close to 60MB with 64 packets. Thank you again, Michael On Thu, Apr 19, 2018 at 11:10 AM, Stokes, Ianwrote: > Hi Michael, > > > > So there are a few issues here we need to address. > > > > Queues for phy devices: > > > > I assume you have set the queues for dpdk0 and dpdk1 yourself using > > > > ovs-vsctl set Interface dpdk0 options:n_rxq=8 > > ovs-vsctl set Interface dpdk0 options:n_rxq=8 > > > > Receive Side Scaling (RSS) is used to distribute ingress traffic among the > queues on the NIC at a hardware level. It will be split between queues > based on des tip so it’s important that test traffic varies if you want > traffic to be dispersed evenly among the queues at this level. > > > > Vhost user queues: > > > > You do not have to set the number of queues for vhost ports with n_rxq > since OVS 2.6 as done above but you do have to include the number of > supported queues in the QEMU command line argument that launches the VM by > specifying the argument queues=’Num_Queues’ for the vhost port. If using VM > Kernel virtio interfaces within the VM you will need to enable the extra > queues also using ethtool –L. Seeing that there is only 1 queue for your > vhost user port I think you are missing one of these steps. > > > > PMD configuration: > > > > Since your only using 1 PMD I don’t see much point of using multiple > queues. Typically you match the number of PMDs to the number of queues that > you would like to ensure an even distribution. > > If using 1 PMD like in your case the traffic will always be enqueued to > queue 0 of vhost device even if there are multiple queues available. This > is related to the implantation within OVS. > > > > As a starting point it might be easier to start with 2 PMDs and 2 rxqs for > each phy and vhost ports that you have and ensure that works first. > > > > Also are you isolating the cores the PMD runs on? If not then processes > could be scheduled to that core which would interrupt the PMD processing, > this could be related to the traffic drops you see. > > > > Below is a link to a blog that discusses vhost MQ, it uses OVS 2.5 but a > lot of the core concepts still apply even if some of the configuration > commands may have changed > > > > https://software.intel.com/en-us/articles/configure-vhost- > user-multiqueue-for-ovs-with-dpdk > > > > Ian > > > > *From:* michael me [mailto:1michaelmesgu...@gmail.com] > *Sent:* Wednesday, April 18, 2018 2:23 PM > *To:* Stokes, Ian > *Cc:* ovs-discuss@openvswitch.org > *Subject:* Re: [ovs-discuss] ovs-dpdk performance not stable > > > > Hi Ian, > > > > In the deployment i do have vhost user; below is the full output of the > ovs-appctl > dpif-netdev/pmd-rxq-show command. > > root@W:/# ovs-appctl dpif-netdev/pmd-rxq-show > > pmd thread numa_id 0 core_id 1: > > isolated : false > > port: dpdk1 queue-id: 0 1 2 3 4 5 6 7 > > port: dpdk0 queue-id: 0 1 2 3 4 5 6 7 > > port: vhu1cbd23fd-82queue-id: 0 > > port: vhu018b3f01-39queue-id: 0 > > > > what is strange for me and i don't understand is why do i have only one > queue in the vhost side and eight on the dpdk side. i understood that qemue > automatically had the same amount. though, i am using only one core for the > VM and one core for the PMD. > > in this setting i have eight cores in the system, is that the reason that > i see eight possible queues? > > The setup is North/South (VM to Physical
[ovs-discuss] Could be a bug: extremely long latency for ovn-controller to get SB update in scale test
While working on ovn-controller incremental processing, I was testing the end-to-end latency and comparing with master. I found the latency with master branch is extremely long while changing the way of testing makes it much shorter. The master branch was at commit 593e93e5f (ovsdb clustering was not merged yet). I promised in last ovn meeting to send a report and here are the details. *Precondition: * there are 1k sandbox HVs connected, and 10k lports already created and bound on these HVs. There are 8 lrouters, each connected with 5 lswitches, and each lswitch has 250 lports. *Test: * - on a sandbox (HV1), create a ovs-port: # ovs-vsctl add-port br-int lsp1 -- set interface lsp1 type=internal -- set interface lsp1 external-ids:iface-id=lsp1 - create the lport on NB on a lswitch that is not yet in local datapath of HV1, wait for port up and then sync HVs # time ovn-nbctl lsp-add lswitch_b9322b_ecpmKt lsp1 -- lsp-set-addresses lsp1 "aa:aa:aa:aa:aa:aa 172.145.222.222"; time ovn-nbctl wait-until Logical_Switch_Port lsp1 up=true; time ovn-nbctl --wait=hv sync *Result:* real0m0.408s user0m0.316s sys 0m0.008s real0m4.837s user0m0.348s sys 0m0.012s *real7m48.373s* user0m0.348s sys 0m0.016s *Notes: * - the time for the last --wait=hv sync is extremely long: 7m48s. - the test is with the patch for nb_cfg separation ( https://patchwork.ozlabs.org/patch/899608/), so that --wait=hv sync is optimized without notification flooding. - the lswitch where the lsp1 is added doesn't belong to any existing local datapath on HV1, i.e., there is no other existing lport on HV1 in this lswitch or any other lswitch that is connected with this lswitch by the same lrouter. This means when adding the lport the HV1 will change monitor condition to watch on a new set of datapaths and port-bindings on them. (tested with adding the lsp1 on an existing local datapath, there is no such extremely long latency.) - adding a command to sync hvs before waiting for port up changes the behavior: the total latency is only 20+ seconds then: # time ovn-nbctl lsp-add lswitch_b9322b_ecpmKt lsp1 -- lsp-set-addresses lsp1 "aa:aa:aa:aa:aa:aa 172.145.222.222"; *time ovn-nbctl --wait=hv sync;* time ovn-nbctl wait-until Logical_Switch_Port lsp1 up=true; time ovn-nbctl --wait=hv sync *Debug log analysis:* With debug log enabled for ovn-controller, I found that the latency was because ovn-controller didn't get any SB update during the 7+ minutes. Here are the related logs and analysis: // HV1 got notification of lsp1 port binding added by northd after the lsp1 is created in NB 2018-03-20T00:00:07.353Z|51461|jsonrpc|DBG|tcp:192.168.10.10:6640: received notification, method="update2", params=["monid",{"Port_Binding":{"d6f145bf-ac8f-4107-a565-9ef1fcc9ec6c":{"insert":{"tunnel_key":252,"mac":"aa:aa:aa:aa:aa:aa 172.145.222.222","logical_port":"lsp1","datapath":["uuid","bb979633-789e-4ef8-9663-51839f000c0f"]] // Claim the lsp1 on this chassis 2018-03-20T00:00:07.367Z|51465|binding|INFO|Claiming lport lsp1 for this chassis. 2018-03-20T00:00:07.367Z|51466|binding|INFO|lsp1: Claiming aa:aa:aa:aa:aa:aa 172.145.222.222 // Update port binding to this chassis 2018-03-20T00:00:11.381Z|51588|jsonrpc|DBG|tcp:192.168.10.10:6640: send request, method="transact", params=["OVN_Southbound",{"where":[["_uuid","==",["uuid","d6f145bf-ac8f-4107-a565-9ef1fcc9ec6c"]]],"row":{"chassis":["uuid","b2f297f6-cd8b-4429-94f0-274a24d6640a"]},"op":"update","table":"Port_Binding"}], id=*2169* // Send request to change monitor conditions for new local datapaths 2018-03-20T00:00:11.382Z|51594|jsonrpc|DBG|tcp:192.168.10.10:6640: send request, method=“monitor_cond_change” … id=*2170* // Received notification for its own port-binding update 2018-03-20T00:00:11.383Z|51595|jsonrpc|DBG|tcp:192.168.10.10:6640: received notification, method="update2", params=["monid",{"Port_Binding":{"d6f145bf-ac8f-4107-a565-9ef1fcc9ec6c":{"modify":{"chassis":["uuid","b2f297f6-cd8b-4429-94f0-274a24d6640a"]] 2018-03-20T00:00:11.383Z|51596|jsonrpc|DBG|tcp:192.168.10.10:6640: received reply, result=[{"count":1}], id=*2169* */ 7 minutes later ... but why???* // Received the notification for the new monitor condition related data (it's huge data) 2018-03-20T00:07:56.183Z|52643|jsonrpc|DBG|tcp:192.168.10.10:6640: received notification, method="update2", params=["monid",{"Multicast_Group":{"944391c4-7bf9-4d46-9662-966f9a5450e0":{"insert":{"ports":["set",[["uuid","00d22282-ac38-4a4e-8f47-3684fa915051"],["uuid","03010cdb-647b-4a1d-b0ca-7f4b0e7a6862"],["uuid","03ba954f-5c96-47f0-b517-4f30b2e61907"],["uuid","0480cf96-5158-45aa-ba49-6053cdec5ce3"],["uuid","052397f5-ebb3-4e81-9970-...(huge data)... 2018-03-20T00:07:56.386Z|52644|jsonrpc|DBG|tcp:192.168.10.10:6640: received reply, result={}, id= *2170* // Received the nb_cfg change triggered by the --wait=hv sync command, together with the lflow update caused by the port status change: up=true
[ovs-discuss] latest OVS-2.9.0 Build
OVS team, We are working on opendaylight CSIT. Currently we are using OVS-2.9.0 version and their was an issue and got fixed on March- 9- 2018. If you have triggered latest OVS-2.9.x Build after March 9 . Please let us know the path to download since we need only release Build for our testing for example : http://openvswitch.org/releases/openvswitch-2.9.0.tar.gz If not please let us know when its going to be available . Its helps us to proceed further on blocked testing. BRs, Mahesh Loni smime.p7s Description: S/MIME cryptographic signature ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss
Re: [ovs-discuss] Reg IPv6 Neighbor Advertisement Message fields
> Zak is working on that feature; I expect patches will hit the mailing list in > the next week or two. Hi Justin, As part of this feature, will it also enable us to rewrite the options tlv:type field from SLL (1) to TLL (2) Or may be a set-field option to rewrite these fields in Options TLV. This will help in supporting multicast solicitations. Warm Regards, Vishal ___ discuss mailing list disc...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-discuss
Re: [ovs-discuss] ovs-dpdk performance not stable
Hi Michael, So there are a few issues here we need to address. Queues for phy devices: I assume you have set the queues for dpdk0 and dpdk1 yourself using ovs-vsctl set Interface dpdk0 options:n_rxq=8 ovs-vsctl set Interface dpdk0 options:n_rxq=8 Receive Side Scaling (RSS) is used to distribute ingress traffic among the queues on the NIC at a hardware level. It will be split between queues based on des tip so it’s important that test traffic varies if you want traffic to be dispersed evenly among the queues at this level. Vhost user queues: You do not have to set the number of queues for vhost ports with n_rxq since OVS 2.6 as done above but you do have to include the number of supported queues in the QEMU command line argument that launches the VM by specifying the argument queues=’Num_Queues’ for the vhost port. If using VM Kernel virtio interfaces within the VM you will need to enable the extra queues also using ethtool –L. Seeing that there is only 1 queue for your vhost user port I think you are missing one of these steps. PMD configuration: Since your only using 1 PMD I don’t see much point of using multiple queues. Typically you match the number of PMDs to the number of queues that you would like to ensure an even distribution. If using 1 PMD like in your case the traffic will always be enqueued to queue 0 of vhost device even if there are multiple queues available. This is related to the implantation within OVS. As a starting point it might be easier to start with 2 PMDs and 2 rxqs for each phy and vhost ports that you have and ensure that works first. Also are you isolating the cores the PMD runs on? If not then processes could be scheduled to that core which would interrupt the PMD processing, this could be related to the traffic drops you see. Below is a link to a blog that discusses vhost MQ, it uses OVS 2.5 but a lot of the core concepts still apply even if some of the configuration commands may have changed https://software.intel.com/en-us/articles/configure-vhost-user-multiqueue-for-ovs-with-dpdk Ian From: michael me [mailto:1michaelmesgu...@gmail.com] Sent: Wednesday, April 18, 2018 2:23 PM To: Stokes, IanCc: ovs-discuss@openvswitch.org Subject: Re: [ovs-discuss] ovs-dpdk performance not stable Hi Ian, In the deployment i do have vhost user; below is the full output of the ovs-appctl dpif-netdev/pmd-rxq-show command. root@W:/# ovs-appctl dpif-netdev/pmd-rxq-show pmd thread numa_id 0 core_id 1: isolated : false port: dpdk1 queue-id: 0 1 2 3 4 5 6 7 port: dpdk0 queue-id: 0 1 2 3 4 5 6 7 port: vhu1cbd23fd-82queue-id: 0 port: vhu018b3f01-39queue-id: 0 what is strange for me and i don't understand is why do i have only one queue in the vhost side and eight on the dpdk side. i understood that qemue automatically had the same amount. though, i am using only one core for the VM and one core for the PMD. in this setting i have eight cores in the system, is that the reason that i see eight possible queues? The setup is North/South (VM to Physical network) as for pinning the PMD, i always pin the PMD to core 1 (mask=0x2). when i set the n_rxq and n_txq to high values (even 64 or above) i see no drops for around a minute or two and then suddenly bursts of drops as if the cache was filled. Have you seen something similar? i tried to play with the "max-idle", but it didn't seem to help. originally, i had a setup with 2.9 and 17.11 and i was not able to get better, performance but it could be that i didn't tweak as much. However, i am trying to deploy a setup that i can install without needing to MAKE. Thank you for any input, Michael On Tue, Apr 17, 2018 at 6:28 PM, Stokes, Ian > wrote: Hi Michael, Are you using dpdk vhostuser ports in this deployment? I would expect to see them listed in the output of ovs-appctl dpif-netdev/pmd-rxq-show you posted below. Can you describe the expected traffic flow ( Is it North/South using DPDK phy devices as well as vhost devices or east/west between vm interfaces only). OVS 2.6 has the ability to isolate and pin rxq queues for dpdk devices to specific PMDs also. This can help provide more stable throughput and defined behavior. Without doing this I believe the distribution of rxqs was dealt with in a round robin manner which could change between deployments. This could explain what you are seeing i.e. sometimes the traffic runs without drops. You could try to examine ovs-appctl dpif-netdev/pmd-rxq-show when traffic is dropping and then again when traffic is passing without issue. This output along with the flows in each case might provide a clue as to what is happening. If there is a difference between the two you could investigate pinning the rxqs to the specific setup although you will only benefit from this when have at least 2 PMDs instead of 1. Also OVS 2.6 and DPDK