Still failing: https://jenkins.fd.io/job/vpp-verify-master-centos7/2810/console
Ed On Mon, Dec 12, 2016 at 10:33 AM, Ed Warnicke <hagb...@gmail.com> wrote: > Rebased my patch to restore make test on top of yours: > > https://gerrit.fd.io/r/#/c/4194/ > > Lets see if it verifies... > > Ed > > On Mon, Dec 12, 2016 at 10:01 AM, Matej Klotton -X (mklotton - PANTHEON > TECHNOLOGIES at Cisco) <mklot...@cisco.com> wrote: > >> Hi, >> >> I fixed issue with VXLAN in https://gerrit.fd.io/r/#/c/4194/ >> >> Matej. >> >> > -----Original Message----- >> > From: csit-dev-boun...@lists.fd.io [mailto:csit-dev-boun...@lists.fd.io] >> On >> > Behalf Of Ole Troan >> > Sent: 12. decembra 2016 17:34 >> > To: Maciek Konstantynowicz (mkonstan) <mkons...@cisco.com> >> > Cc: csit-...@lists.fd.io; vpp-dev@lists.fd.io >> > Subject: Re: [csit-dev] [vpp-dev] vpp make test for verify - are we >> there yet ? >> > >> > Maciek, >> > >> > No. >> > There is a breakage with VXLAN. Here is the analysis from Neale: >> > >> > >> > ========================================================== >> > ============ >> > ERROR: Encapsulation test >> > ---------------------------------------------------------------------- >> > Traceback (most recent call last): >> > File "/vpp/test/template_bd.py", line 91, in test_encap >> > self.check_encapsulation(pkt) >> > File "/vpp/test/test_vxlan.py", line 53, in check_encapsulation >> > self.assertEqual(pkt[VXLAN].vni, type(self).vni) File "/vpp/build- >> > root/python/virtualenv/local/lib/python2.7/site-packages/sca >> py/packet.py", >> > line 817, in __getitem__ >> > raise IndexError("Layer [%s] not found" % lname) >> > IndexError: Layer [VXLAN] not found >> > >> > >> > >> > Is more subtle. >> > The VxLAN encap test passes every time when just the VxLAN suite is run. >> > Here is the VPP trace with the VxLAN encapped packet: >> > >> > 00:00:02:549965: pg0-tx >> > 192 bytes >> > current data -50, length 192, free-list 6, trace 0x0 >> > 00000000: >> > 0000000000000000000000000000000000000000000000000000000000000000 >> > 00000020: >> > 0000000000000000000000000000000000000000000000000000000000000000 >> > 00000040: >> > 000000000000000000000000000002010000ff0202fe9a2f300a0800450000b2 >> > 00000060: 00000000fd116317ac100101ac100102dc4012b5009e0000 <<0800000 >> > 000000100>> >> > 00000080: >> > 0000000000010000000000020800450000800001000040117063040302010102 >> > 000000a0: >> > 03044e202710006c2582a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > 000000c0: >> > a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > 000000e0: >> > a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > 00000100: a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > >> > Scapy interprets this as: >> > >> > ###[ Ethernet ]### >> > dst = 02:01:00:00:ff:02 >> > src = 02:fe:9a:2f:30:0a >> > type = 0x800 >> > ###[ IP ]### >> > … >> > ###[ UDP ]### >> > sport = 56384 >> > dport = 4789 >> > len = 158 >> > chksum = 0x0 >> > ###[ VXLAN ]### >> > flags = 134217728L >> > vni = 0x1 >> > reserved = 0x0 >> > ###[ Ethernet ]### >> > …. >> > >> > Note the flags field is a 4 byte integer with value 0x08000000. As was >> in the >> > header. >> > >> > >> > When the VxLAN test runs along with all the other tests we get the same >> > packet in the VPP trace. >> > >> > >> > >> > 00:00:02:726784: pg0-tx >> > 192 bytes >> > current data -50, length 192, free-list 6, trace 0x0 >> > 00000000: >> > 0000000000000000000000000000000000000000000000000000000000000000 >> > 00000020: >> > 0000000000000000000000000000000000000000000000000000000000000000 >> > 00000040: >> > 000000000000000000000000000002010000ff0202fe994605ca0800450000b2 >> > 00000060: 00000000fd116317ac100101ac100102dc4012b5009e0000 >> > <<08000000 00000100>> >> > 00000080: >> > 0000000000010000000000020800450000800001000040117063040302010102 >> > 000000a0: >> > 03044e202710006c2582a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > 000000c0: >> > a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > 000000e0: >> > a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > 00000100: a5a5a5a5a5a5a5a5a5a5a5a5a5a5 >> > >> > >> > >> > But this time Scapy interprets this as: >> > >> > ###[ Ethernet ]### >> > … >> > ###[ IP ]### >> > … >> > ###[ UDP ]### >> > ... >> > ###[ VXLAN ]### >> > flags = Instance >> > reserved1 = 0 >> > vni = 0x1 >> > reserved2 = 0x0 >> > >> > Where according to the .py definition of a VxLAN header: >> > >> > class VXLAN(Packet): >> > name = "VXLAN" >> > >> > fields_desc = [ >> > FlagsField("flags", 0x8, 8, >> > ['OAM', 'R', 'NextProtocol', 'Instance', >> > 'V1', 'V2', 'R', 'G']), >> > >> > The flag ‘Instance’ refers to 0x10 not 0x08. Bit shift left one. >> > And then, according to the error, it cannot find the VxLAN header. >> > >> > FIIK what that’s about. >> > >> > Neale >> > >> > > On 12 Dec 2016, at 17:19, Maciek Konstantynowicz (mkonstan) >> > <mkons...@cisco.com> wrote: >> > > >> > > Hello, Does anyone know if vpp make test is back on track to be ready >> to be >> > used for vpp make verify jobs on a per patch basis? >> > > >> > > Being lazy I know - cause I could run it myself :) >> > > >> > > -Maciek >> > > >> > > _______________________________________________ >> > > vpp-dev mailing list >> > > vpp-dev@lists.fd.io >> > > https://lists.fd.io/mailman/listinfo/vpp-dev >> > >> > _______________________________________________ >> > csit-dev mailing list >> > csit-...@lists.fd.io >> > https://lists.fd.io/mailman/listinfo/csit-dev >> _______________________________________________ >> csit-dev mailing list >> csit-...@lists.fd.io >> https://lists.fd.io/mailman/listinfo/csit-dev >> > >
_______________________________________________ vpp-dev mailing list vpp-dev@lists.fd.io https://lists.fd.io/mailman/listinfo/vpp-dev
