[vpp-dev] CSIT switchover to Ubuntu 16.04.1

2016-10-14 Thread Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at Cisco)
Hello, After planning and preparation phase, the infrastructure and code is now in final phase of making the switch over to new Ubuntu 16.04.1. CSIT commit for changes in bootstrap/topology files as well as necessary changes in libraries is: https://gerrit.fd.io/r/#/c/3290/ VIRL job (csit-vpp-

[vpp-dev] CSIT switchover to Ubuntu 16.04.1

2016-10-14 Thread Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at Cisco)
Hello, After planning and preparation phase, the infrastructure and code is now in final phase of making the switch over to new Ubuntu 16.04.1. CSIT commit for changes in bootstrap/topology files as well as necessary changes in libraries is: https://gerrit.fd.io/r/#/c/3290/ VIRL job (csit-vpp-

Re: [vpp-dev] [csit-dev] CSIT switchover to Ubuntu 16.04.1

2016-10-14 Thread Dave Wallace
Peter, I was going to complete the review of 3290 and merge it, but I'm currently getting an error from https://gerrit.fd.io: "502 Bad Gateway". I'll try again in a few hours when I'm back from an appointment. Thanks, -daw- On 10/14/2016 08:51 AM, Peter Mikus -X (pmikus - PANTHEON TECHNOLOG

Re: [vpp-dev] [csit-dev] CSIT switchover to Ubuntu 16.04.1

2016-10-14 Thread Dave Barach (dbarach)
I've pinged the Linux Foundation on their "emergency" alias. Gerrit.fd.io is inaccessible. Thanks... Dave From: vpp-dev-boun...@lists.fd.io [mailto:vpp-dev-boun...@lists.fd.io] On Behalf Of Dave Wallace Sent: Friday, October 14, 2016 9:29 AM To: Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at

Re: [vpp-dev] [csit-dev] CSIT switchover to Ubuntu 16.04.1

2016-10-14 Thread Maciek Konstantynowicz (mkonstan)
Thanks Peter - Great job ! Ed, CJ, Could you assist us with coordinated switch-over in both vpp and ci-mgmt projects per Peter’s points 2) and 3) as they have to be applied concurrently. Would like to do it live on Monday AM EDT/PDT and immediately verify operation to see if any issues. -Macie

[vpp-dev] dpdk/vpp and cross-version migration for vhost

2016-10-14 Thread Michael S. Tsirkin
Hi! So it looks like we face a problem with cross-version migration when using vhost. It's not new but became more acute with the advent of vhost user. For users to be able to migrate between different versions of the hypervisor the interface exposed to guests by hypervisor must stay unchanged. T

Re: [vpp-dev] bond interfaces in VPP

2016-10-14 Thread Bernier, Daniel (520165)
Thanks As any work been done on the other modes (ie lacp) ? Thanks Daniel Bernier Sent from my iphone On Oct 6, 2016, at 4:03 PM, John Lo (loj) mailto:l...@cisco.com>> wrote: Yes, bonded interface is supported in VPP using the DPDK Link Bonding PMD library. One can specify slave links to th

[vpp-dev] Media processing in VPP

2016-10-14 Thread Mandeep Singh via vpp-dev
Given the VPP plugin extensible framework it looks very plausible to add media processing plugin into VPP itself. That is, instead of processing media in VMs move it into VPP layer. The media control plane can stay as VM and can interact with SDN controller via remote programmability option. Is

Re: [vpp-dev] Media processing in VPP

2016-10-14 Thread Keith Burns
On Fri, Oct 14, 2016, 6:03 PM Mandeep Singh via vpp-dev wrote: Given the VPP plugin extensible framework it looks very plausible to add media processing plugin into VPP itself. That is, instead of processing media in VMs move it into VPP layer. The media control plane can stay as VM and can inter

Re: [vpp-dev] Media processing in VPP

2016-10-14 Thread Mandeep Singh via vpp-dev
Keith, I was thinking more of SIP and RTP. However, traditional protocols such as Megaco/MGCP seems to have no support in SDNController. Are there other means of passing media description to VPP? I am not very familiar with OpenFlow but at face it doesn't look like protocol to carry media info

Re: [vpp-dev] bond interfaces in VPP

2016-10-14 Thread John Lo (loj)
We have only tested bonded interface in mode 2 with VPP and not with other modes. For mode 4 (which I assume is what you are asking about for LACP), I am not sure VPP behavior fit the DPDK mode 4 requirement for polling interval and buffer size. Regards, John From: Bernier, Daniel (520165) [m