Re: [vpp-dev] VNET classifier table/session

2019-01-18 Thread Harish Patil
Hi Andrew/all, I was going thru' some of the code/doc pointers. As I said I'm using relatively older VPP releases (16.09/17.01). With regards to IPFIX (using VNET classifier functionality), I do not see any support for sending real flow statistics via IPFIX templates. I just see that plain IPFIX

[vpp-dev] VPP 19.01 Release Milestone RC2 is Wednesday 23 January 2019

2019-01-18 Thread Andrew Yourtchenko
Hi all, A gentle reminder that VPP 19.01 RC2 is this Wednesday January 23rd. Note: After Wednesday's RC2 Milestone, only critical bug fixes will be merged into stable/1901. Please review open anomalies for candidates to be fixed this week. Also, please remember to open a Jira ticket for all

[vpp-dev] VPP 19.01 RC1 milestone is complete

2019-01-18 Thread Andrew Yourtchenko
Hi all, The VPP 19.01 RC1 milestone is now fully complete. The VPP 19.01 release branch (stable/1901) has been created, along with the associated nexus and packagecloud repo's. vpp master branch is open for all patches slated for VPP 19.04 (and beyond). vpp stable/1901 is now also

[vpp-dev] debian/ubuntu packaging rework

2019-01-18 Thread Damjan Marion via Lists.Fd.Io
Dear all, I just submitted debian/ubuntu packaging rework patch[1], and would like to hear feedback from people: Instead of hardcoded shell scripts it uses CMake components to put file into right package, so for example: install(FILES foo.txt DESTINATION share/vpp COMPONENT vpp-dev) means

Re: [vpp-dev] IpSec API call ikev2_set_profile_auth causes VPP to crash

2019-01-18 Thread Neale Ranns via Lists.Fd.Io
Hi Michal, Thanks for the fixes. I’ll merge them when they verify. /Neale De : au nom de "Michal Cmarada via Lists.Fd.Io" Répondre à : "Michal Cmarada -X (mcmarada - PANTHEON TECHNOLOGIES at Cisco)" Date : vendredi 18 janvier 2019 à 13:35 À : "vpp-dev@lists.fd.io" , "hc2...@lists.fd.io"

Re: [vpp-dev] wireshark vpp dispatch trace dissector merged

2019-01-18 Thread Dave Barach via Lists.Fd.Io
You must have handed wireshark a trace from an older VPP image which used WTAP_ENCAP_USER13 = 160, instead of the official WTAP_ENCAP_VPP = 280. The encap type code that vpp writes changed on or about 1/9. Look for a commit entitled “Use the official libpcap file type” w/ SHA-1

[vpp-dev] IpSec API call ikev2_set_profile_auth causes VPP to crash

2019-01-18 Thread Michal Cmarada via Lists.Fd.Io
Hi, I was testing IpSec API via honeycomb and when we called ikev2_set_profile_auth with "data" and "data_length" set, vpp crashed. When investigating we found out that "data_length" is not processed correctly. I prepared a patch that fixes this error. It was verified using HC via api call.

Re: [vpp-dev] wireshark vpp dispatch trace dissector merged

2019-01-18 Thread Benoit Ganne (bganne) via Lists.Fd.Io
I can confirm : I just cloned the master branch from the wireshark repository, compiled and it worked. A minor possible improvement though: I had to go to Preferences > Protocols > DLT_USER > Edit to add an association between the VPP dissector and DLT_USER=160 so that wireshark recognize &

Re: [vpp-dev] wireshark vpp dispatch trace dissector merged

2019-01-18 Thread Marco Varlese
That's great !!! On Thu, 2019-01-17 at 13:11 +, Dave Barach via Lists.Fd.Io wrote: > > > I’m pleased to announce that our dispatch trace wireshark dissector has been > merged. At some point in the indefinite future, everyone’s favorite distro > will include a copy of wireshark which knows >

Re: [vpp-dev] VPP statistics

2019-01-18 Thread Ole Troan
> Statistics handling did change in !8.10, see: > https://docs.fd.io/vpp/18.10/md_src_vpp_stats_stats.html > > there is no longer an exporter thread, instead there is a shared memory > segment from which your client can read the stats at the period of your > choosing. Yep, although there is

Re: [vpp-dev] VPP statistics

2019-01-18 Thread Neale Ranns via Lists.Fd.Io
Hi Satish, Statistics handling did change in !8.10, see: https://docs.fd.io/vpp/18.10/md_src_vpp_stats_stats.html there is no longer an exporter thread, instead there is a shared memory segment from which your client can read the stats at the period of your choosing. /neale De : au nom