Hi Nagp,
No, the current implementation is not multi-fib capable, i.e., all connections
go through the main fib. Transport endpoint structs used in the session api and
the application interface already carry vrfs but we’re not yet using them in
tcp and for session lookup. We do however plan to
Hi Neale,
I forgot to say ,I neither created tunnel nor used LDP.I used the static
command configuration.
Thanks,
xyxue
From: 薛欣颖
Date: 2017-05-25 13:36
To: nranns; vpp-dev
Subject: Re: [vpp-dev] MPLS L3VPN PING FAILED
Hi Neale,
The ping still failed .Here is the specific information：
DBGvp
Hi Neale,
The ping still failed .Here is the specific information：
DBGvpp# show trace
CLIB unknown format `%#' x label 0 eos 1024
17:23:38:439098: lookup-mpls-dst
fib-index:0 hdr:[1023:85:0:eos] load-balance:29
17:23:38:439159: ip4-mpls-label-disposition
disp:0
17:23:38:439198:
I am asking this because the session lookup functions in tcp_input.c does
not seem to be taking a fib index or a table id to look up the session -
just the usual 4-tuple.
If it is done in a different way, please help me see it.
Thanks,
-nagp
___
vpp-dev
I am guessing the fix will work long term, but if not, perhaps an
alternative is to add --oldpackage to the rpm install options.
Burt
On Fri, May 19, 2017 at 8:34 AM, Thomas F Herbert
wrote:
>
>
> On 05/18/2017 04:37 PM, Florin Coras wrote:
>
> Well, uri.am is a known issue, apparently not the
That specific trace formatting error was fixed on master earlier today. Could
you give it a try?
Chris.
From: vpp-dev-boun...@lists.fd.io [mailto:vpp-dev-boun...@lists.fd.io] On
Behalf Of ???
Sent: Wednesday, May 24, 2017 22:03
To: vpp-dev@lists.fd.io
Subject: [vpp-dev] Some error in L3VPN
Hi
Hi guys,
There is a ping failed occured in L3vpn,and the trace had incorrect print
information below:
DBGvpp# show trace
CLIB unknown format `%#' x label 0 eos 1024
17:23:38:439098: lookup-mpls-dst
fib-index:0 hdr:[1023:85:0:eos] load-balance:29
17:23:38:439159: ip4-mpls-label-dis
Hi Matt,
Glad to hear it. And thank you for the patch.
Regards,
neale
-Original Message-
From: Matthew Smith
Date: Wednesday, 24 May 2017 at 22:24
To: "Neale Ranns (nranns)"
Cc: "vpp-dev@lists.fd.io"
Subject: Re: [vpp-dev] IPsec interface handling in FIB
Hi Neale,
Hi Neale,
I set that flag and have been testing with it and it seems to have solved the
problem.
Thanks!
-Matt
> On May 20, 2017, at 2:18 AM, Neale Ranns (nranns) wrote:
>
> Hi Matt,
>
> No ARP lookup is needed for interfaces that are point-2-point. The FIB will
> link entries reachable
Dear Neale,
your guess was right.
With ping it works fine and vpp does not crash.
I was using prove because is suggested by the official wiki in order to
test the interface, therefore I would suggest to replace it also in the
wiki.
Thanks for the prompt answer.
Best regards,
Alessio
On Wed, M
Hi Alessio,
My guess is the probe command caused VPP to crash, and since your running it as
a deamon, it restarted with your previous configs gone.
The probe command sends an ARP request to the address in question, so don’t
probe your own address, instead probe a peer’s. These days we also hav
Dear all,
I use the vagrant installation.
I can start the VM, build the packages and start vpp process.
I am stuck at the step (Step 1: Configure and enable an interface) of the
guide available at
https://wiki.fd.io/view/VPP/Build,_install,_and_test_images.
At the first boot I have:
vagrant@lo
Hi Xyxue,
VPP does not support VP{WL]S.
However, I’ve made a patch to enable it;
https://gerrit.fd.io/r/#/c/6861/1
please experiment and let me know how you get on.
Changes to your configs below:
- Create the mpls tunnel in L2 mode
mpls tunnel l2-only via 1.1.1.2 host-eth0 out-label
On Wed, May 24, 2017 at 6:21 PM, Jon Loeliger wrote:
> On Wed, May 24, 2017 at 2:00 AM, wrote:
> > Hi Jon,
>
> Hi Ole,
>
> > Thanks for the poetry! ;-)
>
> Most welcome.
>
> > This is me in 01384fe
> > Apologies for that. Next time I see you let me bring both band aid and
> whiskey.
>
> Apology
well sure enough…i doubled the cpu reservation and it all passed clean…
things are pretty quiet right now so ill throw it up again when things are busy
but right now I’m just happy to see a full test-all-debug pass.
thanks,
Ed
> On May 24, 2017, at 8:42 AM, Klement Sekera -X (ksekera - PANTHEO
It may be extreme, but perhaps detect if any namespace is configured (whether
we’re running in a namespace or in the default) and then fall back to the
safest behavior (with a note in the startup output saying so).
Is it reasonable? Could add a flag that always binds every device found for
peop
Hi,
Seems that your last patch fixed the issue. I was now able to bind both unbound
interfaces and also interfaces that were bound to kernel but were down. Thanks
_____ _ ___
__/ __/ _ \ (_)__| | / / _ \/ _ \
_/ _// // / / / _ \ | |/ / ___/ ___/
/_/ /(_)_/
On Wed, May 24, 2017 at 2:00 AM, wrote:
> Hi Jon,
Hi Ole,
> Thanks for the poetry! ;-)
Most welcome.
> This is me in 01384fe
> Apologies for that. Next time I see you let me bring both band aid and
> whiskey.
Apology accepted!
> To my excuse, there has been this list of "broken" APIs maint
Hi Sergio,
Right, without the change to dpdk.am, it builds correctly with the
crypto libraries, but the plugin fails to load.
After the change, i'm able to load the plugin, but the enqueue_crypto
call does not return, no crypto error ops counters are incremented
either
Building it using the "mak
I know that the functional BFD tests passed so unless there is a bug in
the tests, the failures are pretty much timing issues. From my
experience the load is the culprit as the BFD tests test interactive
sessions, which need to be kept alive. The timings currently are set at
300ms and for most test
I think i fixed the issue. New version is in gerrit. If you still see the crash
please try to capture backtrace.
Thanks,
Damajn
> On 24 May 2017, at 16:29, Damjan Marion (damarion) wrote:
>
>
> Any chance you can capture backtrace?
>
> just "gdb —args vpp unix interactive”
>
> Thanks,
>
Any chance you can capture backtrace?
just "gdb —args vpp unix interactive”
Thanks,
Damjan
> On 24 May 2017, at 13:10, Michal Cmarada -X (mcmarada - PANTHEON TECHNOLOGIES
> at Cisco) wrote:
>
> Hi,
>
> I tried your patch, I built rpms from it and then reinstalled vpp with those
> rpms. I
right now its a VERY intentional mix…but depending on loading I could easily
see this coming up if those timings are strict.
To not dodge your question max loading on my slowest node would be 3 concurrent
builds on an Xeon™ E3-1240 v3 (4 cores @ 3.4Ghz)
yeah yeah stop laughing…..Do you have s
Hi,
I tried your patch, I built rpms from it and then reinstalled vpp with those
rpms. I ensured that the interface was not bound to kernel or dpdk. But I got
Segmentation fault. See output:
[root@overcloud-novacompute-1 ~]# dpdk-devbind --status
Network devices using DPDK-compatible driver
==
Hi Xyxue,
The lookup was performed in FIB index 1– you must have used ‘set int ip table
host-XXX YYY’ - but the route you added is in the default table.
If you want the routes in the same table as the interface do;
Ip route add table YYY 192.168.3.0/24 via mpls-tunnel0 out-label 1023
Regards,
Yeah, I’m aware of this issue but I don’t have any good idea how to address it.
We can disable unbind if we detect that we are running inside namespace, but
that will not fix problem in opposite direction, when specific interface is
mapped to namespace and vpp running in global.
any ideas?
>
Hi guys,
I have the following configuration:
mpls tunnel add via 2.1.1.1 host-eth1 out-label 1024
set int state mpls-tunnel0 up
ip route add 192.168.3.0/24 via mpls-tunnel0 out-label 1023
Ping from CE to PE ,and the PE drop it.
That is the fib :
192.168.3.0/24
unicast-ip4-chain
[@0]: dpo-
Yes, I can try that. Will send you more details after it is done.
Michal
-Original Message-
From: Damjan Marion (damarion)
Sent: Tuesday, May 23, 2017 6:39 PM
To: Michal Cmarada -X (mcmarada - PANTHEON TECHNOLOGIES at Cisco)
Cc: Dave Barach (dbarach) ; Marco Varlese
; Kinsella, Ray ;
Hi Jon,
Thanks for the poetry! ;-)
This is me in 01384fe
Apologies for that. Next time I see you let me bring both band aid and whiskey.
To my excuse, there has been this list of "broken" APIs maintained by the
Python and Java language binding implementors. It's been on the todo list for a
lon
29 matches
Mail list logo
