Re: [vpp-dev] [csit-dev] Cannot add route to ipsec0 interface

2016-10-11 Thread Matus Fabian -X (matfabia - PANTHEON TECHNOLOGIES at Cisco)
Hi Neale, IPSec interface is tunnel interface created by IKEv2 code when child SA is successfully established. You can't set IP on it because there is missing "ip4_sw_interface_enable_disable" when interface is created, I will fix it. Regards, Matus > -Original Message- > From: csit-de

Re: [vpp-dev] crash in new QoS code

2016-10-11 Thread Neale Ranns
Hi John, The ARP packets in the bridge-domain are not generated by the VPP control-plane (i.e. thread 0), instead they are either RX from a bridge port, or generated by an ip4_arp_inline, so in both cases, from a worker thread - so they should be OK. The ARP request in the traceback below (w

[vpp-dev] Coverity back online

2016-10-11 Thread Luke, Chris
The Coverity Scan service is back online. 75 new defects detected. Worth noting it now also covers the Java code built with the "build-vpp-api" and "plugins" targets. Allegedly soon it can also include Python stuff, will add then when/if it becomes available. The "fix" was to have my script tha

Re: [vpp-dev] crash in new QoS code

2016-10-11 Thread John Lo (loj)
Hi Neal, I believe David is using GRE tunnel in transparent ethernet bridging (TEB) mode, to carry ethernet packet to a bridge domain. For the TEB case, the GRE tunnel is created like an ethernet interface with calls to register_etehrnet_interface using a MAC that's (d00b:eed0: + sw_if_inde

Re: [vpp-dev] crash in new QoS code

2016-10-11 Thread Neale Ranns
Hi David, On 11/10/16 15:36, David Hotham via vpp-dev wrote: So back on the original crash that started this thread… I’ve been doing some digging, have a decent idea of what’s going wrong, and would appreciate input on what the right approach / next steps towards fixing it should be. When th

Re: [vpp-dev] crash in new QoS code

2016-10-11 Thread David Hotham via vpp-dev
My startup.conf currently contains this: cpu { main-core 0 corelist-workers 1 corelist-hqos-threads 2, 3 } David From: Singh, Jasvinder [mailto:jasvinder.si...@intel.com] Sent: 11 October 2016 16:25 To: David Hotham; vpp-dev@lists.fd.io; Dumitrescu, Cristian Subject: RE: crash in new QoS c

Re: [vpp-dev] crash in new QoS code

2016-10-11 Thread Singh, Jasvinder
As I mentioned earlier, hqos thread runs only qos scheduler. In the configuration below, I don't see any logical core assigned to main/worker thread ? From: David Hotham [mailto:david.hot...@metaswitch.com] Sent: Tuesday, October 11, 2016 3:37 PM To: vpp-dev@lists.fd.io; Singh, Jasvinder ; Dum

Re: [vpp-dev] crash in new QoS code

2016-10-11 Thread David Hotham via vpp-dev
So back on the original crash that started this thread... I've been doing some digging, have a decent idea of what's going wrong, and would appreciate input on what the right approach / next steps towards fixing it should be. When the GRE tunnel is configured, VPP's main control thread causes a

Re: [vpp-dev] Cannot add route to ipsec0 interface

2016-10-11 Thread Neale Ranns
Hi Zdenko, I'm surprised traffic was received on the interface given it does not have an IP address. A couple of questions: - what is an IPSEC interface? - how and why did we bypass the IP input features on that interface? some answers inline... On 11/10/16 13:53, Zdenko Olsovsky -X (zols

[vpp-dev] Cannot add route to ipsec0 interface

2016-10-11 Thread Zdenko Olsovsky -X (zolsovsk - PANTHEON TECHNOLOGIES at Cisco)
Hi everyone, I made some tests for IKEv2 functionality which were working about a month ago. Unfortunately, today I realized that these are no longer working. Running this command : exec ip route add 10.0.10.1/24 via ipsec0 -- worked fine before Wil