Re: [vpp-dev] #vpp loadbalancer #vpp

2019-06-28 Thread Pierre Pfister (ppfister) via Lists.Fd.Io 
Hello,

The VPP LB is pretty much an implementation of Maglev 
(https://ai.google/research/pubs/pub44824).

Maglev is the consistent hashing algorithm used to generate the 
’new_flow_table’.
That table is used when a packet for a flow that is not known by the LB is 
forwarded.
When that happens, the flow is added to the sticky_ht (sticky hash table), such 
that later packets from the same flow go to the same application server (AS).

This consistent-hashing approach is the key to having the ability to have 
multiple LBs running in parallel, without need for synchronization.

VPP LB then supports multiple encaps, like IPv4-GRE or IPv6-GRE for IPv4 or 
IPv6 data packet, NAT for IPv4 or IPv6 or L3DSR for IPv4 (see lb_vip_type_t 
enum).

Cheers,

- Pierre


> Le 26 juin 2019 à 10:46, yangjialu...@gmail.com a écrit :
> 
> Hi guys! Does anyone know about how the sticky_ht, buckets and the 
> new_flow_table work?<屏幕快ç…> § 2019-06-26 16.35.44.png><屏幕快ç…> § 
> 2019-06-26 16.35.12.png> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> 
> View/Reply Online (#13377): https://lists.fd.io/g/vpp-dev/message/13377
> Mute This Topic: https://lists.fd.io/mt/32212960/675306
> Mute #vpp: https://lists.fd.io/mk?hashtag=vpp=1480735
> Group Owner: vpp-dev+ow...@lists.fd.io
> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [ppfis...@cisco.com]
> -=-=-=-=-=-=-=-=-=-=-=-

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#13391): https://lists.fd.io/g/vpp-dev/message/13391
Mute This Topic: https://lists.fd.io/mt/32212960/21656
Mute #vpp: https://lists.fd.io/mk?hashtag=vpp=1480452
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] question about load balance

2018-01-10 Thread Pierre Pfister (ppfister) 
Hello,

You will not be able to get an exact match on 1/3 - 2/3 traffic sharing since 3 
is not a power of two.
If you want to get something closer, you may try to increase the values to get 
a better approximation.

For instance, 43 and 85 have a sum of 128 with 43/128 = 0.3359375, which is 
close to the 0.3 you are looking for.

- Pierre


Le 10 janv. 2018 à 07:36, 薛欣颖 > 
a écrit :


Hi Neale,

I know what you mean.
What should I do to match exactly I configured?  And ignore constraints 'a 
load-balance is always a power of 2'.

Thanks,
Xyxue

From: Neale Ranns (nranns)
Date: 2018-01-09 20:58
To: 薛欣颖; vpp-dev
Subject: Re: [vpp-dev] question about load balance
Hi Xyxue,

It does match your configuration. You requested a 2:1 ratio un-equal cost load 
balancing, you got a 5:3, which VPP considers to be close enough. It’s not 
exact because the number of buckets in a load-balance is always a power of 2.

/neale

From: > on 
behalf of 薛欣颖 >
Date: Tuesday, 9 January 2018 at 12:26
To: vpp-dev >
Subject: [vpp-dev] question about load balance


Hi guys,

I'm testing routing load balance. There is a question : When I configured 1:2 , 
the actual effect 3:5. My configuration and more info is shown below:

configuration:
  ip route add 4.1.1.1/24 via ip4-address 1.1.1.2 interface host-eth0 weight 1
  ip route add 4.1.1.1/24 via ip4-address 2.1.1.2 interface host-eth4 weight 2

The actual effect:
4.1.1.0/24
  unicast-ip4-chain
  [@0]: dpo-load-balance: [proto:ip4 index:22 buckets:8 uRPF:29 to:[0:0]]
[0] [@5]: ipv4 via 2.1.1.2 host-eth4: 000c296db08c2c534a0393540800
[1] [@5]: ipv4 via 2.1.1.2 host-eth4: 000c296db08c2c534a0393540800
[2] [@5]: ipv4 via 2.1.1.2 host-eth4: 000c296db08c2c534a0393540800
[3] [@5]: ipv4 via 2.1.1.2 host-eth4: 000c296db08c2c534a0393540800
[4] [@5]: ipv4 via 2.1.1.2 host-eth4: 000c296db08c2c534a0393540800
[5] [@5]: ipv4 via 1.1.1.2 host-eth0: 000c296db078fc4dd4dad8af0800
[6] [@5]: ipv4 via 1.1.1.2 host-eth0: 000c296db078fc4dd4dad8af0800
[7] [@5]: ipv4 via 1.1.1.2 host-eth0: 000c296db078fc4dd4dad8af0800

How can I make  the actual effect the  same as my configuration?

Thanks,
Xyxue

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] gerrit http authentication

2017-11-21 Thread Pierre Pfister (ppfister) 
Ha, I will try. I always used the long password from gerrit UI.

- Pierre

Le 21 nov. 2017 à 09:40, Andrew Yourtchenko 
<ayour...@gmail.com<mailto:ayour...@gmail.com>> a écrit :

Did you try the http password you use for logging into the gerrit on the web ? 
(Aka “Linux foundation” one ?) that one worked fine for me for the past weeks...

--a

On 21 Nov 2017, at 09:31, Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>> wrote:

Hello all,

Heads up on this one since it still does not work for me.

I was using https auth on servers that are shared between multiple people (any 
ssh private key left there could be accessed by other root users).
It seems safer to rely on copy/pasting a password for every push than 
multiplying private keys.

Cheers,

- Pierre



Le 19 oct. 2017 à 20:25, John Lo (loj) <l...@cisco.com<mailto:l...@cisco.com>> 
a écrit :

Yes, I can confirm using LF password does work while https password failed.  
-John

From: vpp-dev-boun...@lists.fd.io<mailto:vpp-dev-boun...@lists.fd.io> 
[mailto:vpp-dev-boun...@lists.fd.io] On Behalf Of Andrew Yourtchenko
Sent: Thursday, October 19, 2017 1:29 PM
To: Luke, Chris <chris_l...@comcast.com<mailto:chris_l...@comcast.com>>
Cc: vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
Subject: Re: [vpp-dev] gerrit http authentication

I got a reply from Vanessa on my ticket - apparently after upgrade the Linux 
foundation password should work... I will test tomorrow but sending mail here 
in case anyone else is interested to try before that.
--a

On 19 Oct 2017, at 16:14, Luke, Chris 
<chris_l...@comcast.com<mailto:chris_l...@comcast.com>> wrote:
…and certainly not on port != 22.

Chris.

From: Andrew Yourtchenko [mailto:ayour...@gmail.com]
Sent: Thursday, October 19, 2017 9:16
To: Gabriel Ganne <gabriel.ga...@enea.com<mailto:gabriel.ga...@enea.com>>
Cc: Luke, Chris 
<chris_l...@cable.comcast.com<mailto:chris_l...@cable.comcast.com>>; 
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
Subject: Re: [vpp-dev] gerrit http authentication

I sent the mail to help desk, which opened the ticket 47239. Not sure if there 
is a way to “pile up” a “me too” it’s pretty much a showstopper for me, 
since same as Gabriel - not all of the environments allow SSH...
--a

On 19 Oct 2017, at 13:56, Gabriel Ganne 
<gabriel.ga...@enea.com<mailto:gabriel.ga...@enea.com>> wrote:

Yes, I do.

Pulling works fine, only the review action is an issue.



This is what shoud correspond to (previously) "HTTP Password" in the gerrit 
settings menu.

The gerrit link seems to be still active : 
https://gerrit.fd.io/r/#/settings/http-password

However, the functionnality itself is deactivated



--

Gabriel Ganne


From: Luke, Chris <chris_l...@comcast.com<mailto:chris_l...@comcast.com>>
Sent: Thursday, October 19, 2017 1:50:30 PM
To: Gabriel Ganne; vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
Subject: RE: gerrit http authentication

Just to be unambiguous for the archives, you mean HTTP authentication when 
pushing patches to Gerrit with Git, and not interactive browsing of the UI?

Chris.

From: vpp-dev-boun...@lists.fd.io<mailto:vpp-dev-boun...@lists.fd.io> 
[mailto:vpp-dev-boun...@lists.fd.io] On Behalf Of Gabriel Ganne
Sent: Thursday, October 19, 2017 4:13
To: vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
Subject: [vpp-dev] gerrit http authentication


Hi,



Unless I'm mistaken, it seems http authentication has been removed from gerrit.

It was useful to me, as I work in a company where any non-http traffic is 
blocked.



Do you think it's possible to restore it ?



Best regards,



--

Gabriel Ganne

___
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev
___
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] gerrit http authentication

2017-11-21 Thread Pierre Pfister (ppfister) 
Hello all,

Heads up on this one since it still does not work for me.

I was using https auth on servers that are shared between multiple people (any 
ssh private key left there could be accessed by other root users).
It seems safer to rely on copy/pasting a password for every push than 
multiplying private keys.

Cheers,

- Pierre



Le 19 oct. 2017 à 20:25, John Lo (loj) > 
a écrit :

Yes, I can confirm using LF password does work while https password failed.  
-John

From: vpp-dev-boun...@lists.fd.io 
[mailto:vpp-dev-boun...@lists.fd.io] On Behalf Of Andrew Yourtchenko
Sent: Thursday, October 19, 2017 1:29 PM
To: Luke, Chris >
Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] gerrit http authentication

I got a reply from Vanessa on my ticket - apparently after upgrade the Linux 
foundation password should work... I will test tomorrow but sending mail here 
in case anyone else is interested to try before that.
--a

On 19 Oct 2017, at 16:14, Luke, Chris 
> wrote:
…and certainly not on port != 22.

Chris.

From: Andrew Yourtchenko [mailto:ayour...@gmail.com]
Sent: Thursday, October 19, 2017 9:16
To: Gabriel Ganne >
Cc: Luke, Chris 
>; 
vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] gerrit http authentication

I sent the mail to help desk, which opened the ticket 47239. Not sure if there 
is a way to “pile up” a “me too” it’s pretty much a showstopper for me, 
since same as Gabriel - not all of the environments allow SSH...
--a

On 19 Oct 2017, at 13:56, Gabriel Ganne 
> wrote:

Yes, I do.

Pulling works fine, only the review action is an issue.



This is what shoud correspond to (previously) "HTTP Password" in the gerrit 
settings menu.

The gerrit link seems to be still active : 
https://gerrit.fd.io/r/#/settings/http-password

However, the functionnality itself is deactivated



--

Gabriel Ganne


From: Luke, Chris >
Sent: Thursday, October 19, 2017 1:50:30 PM
To: Gabriel Ganne; vpp-dev@lists.fd.io
Subject: RE: gerrit http authentication

Just to be unambiguous for the archives, you mean HTTP authentication when 
pushing patches to Gerrit with Git, and not interactive browsing of the UI?

Chris.

From: vpp-dev-boun...@lists.fd.io 
[mailto:vpp-dev-boun...@lists.fd.io] On Behalf Of Gabriel Ganne
Sent: Thursday, October 19, 2017 4:13
To: vpp-dev@lists.fd.io
Subject: [vpp-dev] gerrit http authentication


Hi,



Unless I'm mistaken, it seems http authentication has been removed from gerrit.

It was useful to me, as I work in a company where any non-http traffic is 
blocked.



Do you think it's possible to restore it ?



Best regards,



--

Gabriel Ganne

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Load balancer plugin for VPP

2017-10-20 Thread Pierre Pfister (ppfister) 
Can you do a trace (trace add dpdk-input 4 , show trace ) to see what VPP is 
doing ?
Did you configure the gre tunnel on backend servers ?
Can you show the full tcpdump capture on the backend server (with max 
verbosity) ?

Thanks,

- pierre

Le 20 oct. 2017 à 04:25, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

Hi Pierre,
Thanks for the info, I was able to bring up the VPP LB node, with 3 backend 
servers. I used ping and telnet to verify that backend servers can be reached 
using telnet or ping on the VPP LB node. I'll deal with DPDK later.
Then I tried to do the same thing on a client, VPP LB node got the request, but 
did not perform the forwarding to backend server.
Below are the relevant information, let me know if you can see where did I do 
wrong.

VIP is on 10.99.99.0/24<http://10.99.99.0/24> network

VPP LB Client:
 10.145.207.151
  I have setup route to reach VIP subnet through VPP LB node
ip route add ${VIP}/24 dev eth0
  eth0 is on 10.145.207.0 (both client, VPP LB, and backend servers are all 
on this subnet)
VPP LB
10.145.207.166
 lb conf ip4-src-address 10.145.207.1 timeout 10

 # configure VIP
 lb vip ${VIP}/24 encap gre4 new_len 1024

 # 3 backend servers
 lb as ${VIP}/24 10.145.207.141
 lb as ${VIP}/24 10.145.207.142
 lb as ${VIP}/24 10.145.207.168

On backend servers (141, 142, 168)
In addition to its own host IP. I have one of the VIP (10.99.99.101) 
configured on dummy device.

As mentioned, on VPP LB node, I can telnet or ping 10.99.99.101. But, I was not 
able to do that on the client node.
On the VPP LB node, I used "tcpdump -n -i eth0 port 23" and can see the income 
request from 10.145.207.151, wanting to reach 10.99.99.101.
But. no further response.
I wonder why VPP LB is not doing the forwarding from request coming from the 
outside the node?

John









On Wed, Oct 18, 2017 at 11:40 PM, Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>> wrote:
Hello !

Great that you managed to answer your prior questions.
Did you see this little doc: https://docs.fd.io/vpp/16.12/lb_plugin_doc.html 
(Not very thorough, but it should help).

LB does not require specific NIC. The fact that the lb plugin is loaded does 
not interfere with the NICs.
You should look at DPDK's configuration in startup.conf file (uncomment and 
fill the dpdk section).
Also make sure that the DPDK plugin is loaded (I wild guess based on your 
previous question might be that you moved the lb plugin into a different 
folder, and the dpdk plugin is not loaded by VPP).

"lb conf " is required to tell VPP which IPv4 source address 
to use when sending tunnelled packets.
Since there is no return traffic with Maglev, this address does not need to be 
one of VPP's addresses. But it should be properly routable toward VPP to not 
cause drops due to ingress-filtering.

Cheers,

- Pierre





Le 19 oct. 2017 à 04:34, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

Hi Pierre,
Sorry for the spam, I made further progress and don't want to waste your time 
answering questions that I already found answer.
- Once I got lb pluging loaded properly, I was able to configure it using vppctl
Remaining questions:
- why I am got just "local0" device from "vppctl show int"? Does VPP LB require 
special NIC? How about DPDK?
-  Do I need to set "vppctl lb conf "? Should that be the IP 
of the host where VPP LB is run?

John


On Wed, Oct 18, 2017 at 6:08 PM, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> wrote:
Hi Pierre,
I made further progress.
I did the "rpm -i vpp-plugins-17.07.01-release.x86_64.rpm", and got plugin 
installed.
I updated the /etc/vpp/startup.conf file, and how I can see plugins using 
"vppctl show plugins".
My questions remains are:
- how to configure lb and what is the syntax
- instead of load all the plugins in /usr/lib/vpp_plugins/, how to configure 
and load just one plugin?

John



On Wed, Oct 18, 2017 at 5:35 PM, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> wrote:
Hi Pierre,
I was using "yum install vpp and vpp-lib" and there was no plugins.
Now I build the whole vpp source, including plugins, and added
plugin_path 
//vpp/extras/rpm/vpp-17.07.01/build-root/build-vpp-native/vpp/plugins
to /etc/vpp/startup.conf file. Then restarted vpp.service.
Now
- how to do know that VPP lb plugin is loaded
- I still cant find the "lb" command. I guess it is really not a command. But, 
where I need to add these "lb" configuration and what is the syntax?
- In the monitoring section, it mentioned "show lb, show lv vips, .." Are they 
subcommands for vppctl? I tried and it does not work.

John


On Tue, Oct 3, 2017 at 11:19 PM, Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>&g

Re: [vpp-dev] Load balancer plugin for VPP

2017-10-19 Thread Pierre Pfister (ppfister) 
Hello !

Great that you managed to answer your prior questions.
Did you see this little doc: https://docs.fd.io/vpp/16.12/lb_plugin_doc.html 
(Not very thorough, but it should help).

LB does not require specific NIC. The fact that the lb plugin is loaded does 
not interfere with the NICs.
You should look at DPDK's configuration in startup.conf file (uncomment and 
fill the dpdk section).
Also make sure that the DPDK plugin is loaded (I wild guess based on your 
previous question might be that you moved the lb plugin into a different 
folder, and the dpdk plugin is not loaded by VPP).

"lb conf " is required to tell VPP which IPv4 source address 
to use when sending tunnelled packets.
Since there is no return traffic with Maglev, this address does not need to be 
one of VPP's addresses. But it should be properly routable toward VPP to not 
cause drops due to ingress-filtering.

Cheers,

- Pierre





Le 19 oct. 2017 à 04:34, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

Hi Pierre,
Sorry for the spam, I made further progress and don't want to waste your time 
answering questions that I already found answer.
- Once I got lb pluging loaded properly, I was able to configure it using vppctl
Remaining questions:
- why I am got just "local0" device from "vppctl show int"? Does VPP LB require 
special NIC? How about DPDK?
-  Do I need to set "vppctl lb conf "? Should that be the IP 
of the host where VPP LB is run?

John


On Wed, Oct 18, 2017 at 6:08 PM, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> wrote:
Hi Pierre,
I made further progress.
I did the "rpm -i vpp-plugins-17.07.01-release.x86_64.rpm", and got plugin 
installed.
I updated the /etc/vpp/startup.conf file, and how I can see plugins using 
"vppctl show plugins".
My questions remains are:
- how to configure lb and what is the syntax
- instead of load all the plugins in /usr/lib/vpp_plugins/, how to configure 
and load just one plugin?

John



On Wed, Oct 18, 2017 at 5:35 PM, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> wrote:
Hi Pierre,
I was using "yum install vpp and vpp-lib" and there was no plugins.
Now I build the whole vpp source, including plugins, and added
plugin_path 
//vpp/extras/rpm/vpp-17.07.01/build-root/build-vpp-native/vpp/plugins
to /etc/vpp/startup.conf file. Then restarted vpp.service.
Now
- how to do know that VPP lb plugin is loaded
- I still cant find the "lb" command. I guess it is really not a command. But, 
where I need to add these "lb" configuration and what is the syntax?
- In the monitoring section, it mentioned "show lb, show lv vips, .." Are they 
subcommands for vppctl? I tried and it does not work.

John


On Tue, Oct 3, 2017 at 11:19 PM, Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>> wrote:
Do you start VPP with the 'make run' command ?

It looks like the way you start VPP, it doesn't specify the righ plugins 
directory.
I'd suggest your try with adding "plugin_path " to 
VPP's command-line arguments.

- Pierre


Le 3 oct. 2017 à 20:40, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

Pierre,
Thanks for the response.
I follow instructions to install VPP:
https://wiki.fd.io/view/VPP/Installing_VPP_binaries_from_packages
That gives me vpp utility. There is no lb command.
I did a build of vpp by follow the instructions below to build vpp on linux
https://wiki.fd.io/view/VPP/Pulling,_Building,_Running,_Hacking_and_Pushing_VPP_Code#Linux
I found "lb" in the plugins directory.

But, how to bring up vpp with lb plugin? I hoping to see some instructions that 
I can follow to try this VPP lb.

John


On Sun, Oct 1, 2017 at 11:55 PM, Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>> wrote:
Hello John,

The LB plugin comes with VPP by default (no need for additional download, 
sources are in src/plugins/lb).

As far as instructions are concerned, I am afraid the link you pointed to is 
all there is for now.
But please ask if you feel like something in particular is missing.

- Pierre



Le 30 sept. 2017 à 00:07, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

I saw a VPP Load Balancer mentioned in this document:

https://docs.fd.io/vpp/17.07/lb_plugin_doc.html

But, I did not see installation information.
Can someone point me to where I can get download and setup instructions?

Thanks
John

___
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev







___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Multiple VRFs in 1609

2017-10-10 Thread Pierre Pfister (ppfister) 
Thanks a ton Jeff.

I merged 8717, 8718 and proposed a different fix to 8716 
(https://gerrit.fd.io/r/#/c/8729/).

I also noticed netlink plugin does not compile on 1710 and master due to a 
change in VPP (Mostly renaming which happened here 
https://gerrit.fd.io/r/#/c/8366/2).
I have a patch ready to fix that, but will wait for the 1710 release before 
merging it.

- Pierre


Le 9 oct. 2017 à 19:18, Shaw, Jeffrey B 
<jeffrey.b.s...@intel.com<mailto:jeffrey.b.s...@intel.com>> a écrit :

I’ve submitted changes 8716, 8717, and 8718, which should fix the issues we’ve 
discussed in this thread.

Thanks,
Jeff

From: Michael Borokhovich [mailto:michael...@gmail.com]
Sent: Monday, October 09, 2017 9:07 AM
To: Ni, Hongjun <hongjun...@intel.com<mailto:hongjun...@intel.com>>
Cc: Neale Ranns (nranns) <nra...@cisco.com<mailto:nra...@cisco.com>>; Shaw, 
Jeffrey B <jeffrey.b.s...@intel.com<mailto:jeffrey.b.s...@intel.com>>; 
vppsb-...@lists.fd.io<mailto:vppsb-...@lists.fd.io>; vpp-dev 
<vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>>
Subject: Re: [vpp-dev] Multiple VRFs in 1609

Thanks, Hongjun,

I will check this once I see that the routing plugin works with VPP1707.

On Fri, Oct 6, 2017 at 7:25 PM, Ni, Hongjun 
<hongjun...@intel.com<mailto:hongjun...@intel.com>> wrote:
Hi Michael and Neale,

As per my previous fixing experience on router plugin, I guess that the cause 
may locate at:
Function tap_inject_enabel(): add IPv4 multicast route.
Please take a look at this function and multicast route involved.

-Hongjun

From: vpp-dev-boun...@lists.fd.io<mailto:vpp-dev-boun...@lists.fd.io> 
[mailto:vpp-dev-boun...@lists.fd.io<mailto:vpp-dev-boun...@lists.fd.io>] On 
Behalf Of Michael Borokhovich
Sent: Friday, October 6, 2017 11:22 PM
To: Neale Ranns (nranns) <nra...@cisco.com<mailto:nra...@cisco.com>>
Cc: Shaw, Jeffrey B 
<jeffrey.b.s...@intel.com<mailto:jeffrey.b.s...@intel.com>>; 
vppsb-...@lists.fd.io<mailto:vppsb-...@lists.fd.io>; vpp-dev 
<vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>>

Subject: Re: [vpp-dev] Multiple VRFs in 1609

I will look into it. It looks like the calling function is in the file 
"turbotap/turbotap/device.c" in vppsb. However, I'm not familiar with the logic 
there...

I will try to fix it temporary locally directly in the 
"vlib_increment_simple_counter" function and see if it works.

Thanks!



On Fri, Oct 6, 2017 at 11:08 AM, Neale Ranns (nranns) 
<nra...@cisco.com<mailto:nra...@cisco.com>> wrote:
Hi Michael,

It would be best if you fixed the caller to not pass ~0. In this case the ~0 is 
coming from the packet VLIB_RX interface. So best to find and fix the input 
node that should have set this correctly.

/neale

From: Michael Borokhovich <michael...@gmail.com<mailto:michael...@gmail.com>>
Date: Friday, 6 October 2017 at 15:49
To: "Shaw, Jeffrey B" 
<jeffrey.b.s...@intel.com<mailto:jeffrey.b.s...@intel.com>>
Cc: "Pierre Pfister (ppfister)" 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>>, Florin Coras 
<fcoras.li...@gmail.com<mailto:fcoras.li...@gmail.com>>, vpp-dev 
<vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>>, "Neale Ranns (nranns)" 
<nra...@cisco.com<mailto:nra...@cisco.com>>, 
"vppsb-...@lists.fd.io<mailto:vppsb-...@lists.fd.io>" 
<vppsb-...@lists.fd.io<mailto:vppsb-...@lists.fd.io>>

Subject: Re: [vpp-dev] Multiple VRFs in 1609

Hi Jeff,

Regarding the workaround for the problem [3] (segfault). Do you think it is 
enough just to add a simple check e.g., if(index != ~0)? Or we have a problem 
somewhere else since the ~0 index should not be passed to this function?

Thanks,
Michael.


On Thu, Oct 5, 2017 at 1:03 PM, Shaw, Jeffrey B 
<jeffrey.b.s...@intel.com<mailto:jeffrey.b.s...@intel.com>> wrote:
Hi Michael, do you really need the router plugin?  Based on the commands you’re 
using it seems you’re adding routes using the vpp cli interface, which is not 
how the router plugin is supposed to work.  Those routes won’t show up on the 
control side (i.e. Linux).  Nor is there any support for multiple routing 
tables.  The routes should be added from the control side, then they are 
mirrored to VPP.

@Pierre,
I was able to get the router plugin to work with VPP 1707 (348edb1 
[origin/stable/1707] Set MAC address needs the HW interface index).
I used vppsb commit 529ec1b [origin/master] Fix name of VCL LD_PRELOAD lib dir 
env var.

A few things went wrong:

1)  Had to disable testrtnl_plugin in vppsb/netlink[1] to get netlink to 
compile

2)  Delete the reference to vlib_buffer_chain_validate() in vppsb/router 
after symbol lookup failure[2]

3)  Work-around a segmentation fault[3] in 
vlib_increment_simple_counter()due to an index being set to ~0

Re: [vpp-dev] Multiple VRFs in 1609

2017-10-10 Thread Pierre Pfister (ppfister) 
Thanks a ton Jeff.

I merged 8717, 8718 and proposed a different fix to 8716 
(https://gerrit.fd.io/r/#/c/8729/).

I also noticed netlink plugin does not compile on 1710 and master due to a 
change in VPP (Mostly renaming which happened here 
https://gerrit.fd.io/r/#/c/8366/2).
I have a patch ready to fix that, but will wait for the 1710 release before 
merging it.

- Pierre

Le 9 oct. 2017 à 19:18, Shaw, Jeffrey B 
> a écrit :

I’ve submitted changes 8716, 8717, and 8718, which should fix the issues we’ve 
discussed in this thread.

Thanks,
Jeff


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Multiple VRFs in 1609

2017-10-05 Thread Pierre Pfister (ppfister) 
Hello Michael,

Can you be more specific about what doesn't work ?

Jeff, do you have some time to look at the router plugin ?

Thanks,

- Pierre


Le 5 oct. 2017 à 05:22, Michael Borokhovich 
> a écrit :

Hi Florin,

Thank you for the response. The reason I stick to 1609 is the router plugin 
(vppsb) that works with 1609 but does not with any other VPP version I tried. 
So, I'm trying to get multiple VRFs works with 1609 since I also need the 
router plugin for the project...

Thanks,
Michael.


On Wed, Oct 4, 2017 at 10:33 PM, Florin Coras 
> wrote:
Michael,

I would recommend you switched to a newer release (17.07 or the soon to be 
release 17.10) since the fib code has been completely reworked in 17.01.

Florin

On Oct 4, 2017, at 7:13 PM, Michael Borokhovich 
> wrote:

Hi,

I'm trying to configure the following setup.

GigabitEthernet0/4/0 - Table 1
GigabitEthernet0/5/0 - Table 2
GigabitEthernet0/6/0 - Table 0 (default)

If a packet with DST_IP="10.5.1.0/24" received at 
GigabitEthernet0/4/0 or GigabitEthernet0/5/0 it needs to be sent via 
GigabitEthernet0/6/0.

The following works for VPP 1710:

vppctl ip route add 10.5.1.0/24 table 1 via 10.5.4.11 
GigabitEthernet0/6/0


vppctl ip route add 10.5.1.0/24 table 2 via 10.5.4.11 
GigabitEthernet0/6/0

But for 1609 it does not work. The routes are added to the default table 0 
instead of tables 1 and 2.

If I do just this (just interface name, without the next ho IP):

vppctl ip route add 10.5.1.0/24 table 1 via 
GigabitEthernet0/6/0

Then the route is installed correctly to table 1 but the outgoing packets are 
messed up - sent without ethernet header.


Any ideas how to make multiple VRFs woks in 1609?


Thanks,

Michael.


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Load balancer plugin for VPP

2017-10-04 Thread Pierre Pfister (ppfister) 
Do you start VPP with the 'make run' command ?

It looks like the way you start VPP, it doesn't specify the righ plugins 
directory.
I'd suggest your try with adding "plugin_path " to 
VPP's command-line arguments.

- Pierre


Le 3 oct. 2017 à 20:40, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

Pierre,
Thanks for the response.
I follow instructions to install VPP:
https://wiki.fd.io/view/VPP/Installing_VPP_binaries_from_packages
That gives me vpp utility. There is no lb command.
I did a build of vpp by follow the instructions below to build vpp on linux
https://wiki.fd.io/view/VPP/Pulling,_Building,_Running,_Hacking_and_Pushing_VPP_Code#Linux
I found "lb" in the plugins directory.

But, how to bring up vpp with lb plugin? I hoping to see some instructions that 
I can follow to try this VPP lb.

John


On Sun, Oct 1, 2017 at 11:55 PM, Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>> wrote:
Hello John,

The LB plugin comes with VPP by default (no need for additional download, 
sources are in src/plugins/lb).

As far as instructions are concerned, I am afraid the link you pointed to is 
all there is for now.
But please ask if you feel like something in particular is missing.

- Pierre



Le 30 sept. 2017 à 00:07, John Wei 
<johnt...@gmail.com<mailto:johnt...@gmail.com>> a écrit :

I saw a VPP Load Balancer mentioned in this document:

https://docs.fd.io/vpp/17.07/lb_plugin_doc.html

But, I did not see installation information.
Can someone point me to where I can get download and setup instructions?

Thanks
John

___
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev



___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Load balancer plugin for VPP

2017-10-02 Thread Pierre Pfister (ppfister) 
Hello John,

The LB plugin comes with VPP by default (no need for additional download, 
sources are in src/plugins/lb).

As far as instructions are concerned, I am afraid the link you pointed to is 
all there is for now.
But please ask if you feel like something in particular is missing.

- Pierre



Le 30 sept. 2017 à 00:07, John Wei 
> a écrit :

I saw a VPP Load Balancer mentioned in this document:

https://docs.fd.io/vpp/17.07/lb_plugin_doc.html

But, I did not see installation information.
Can someone point me to where I can get download and setup instructions?

Thanks
John

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Build vpp on openwrt

2017-09-11 Thread Pierre Pfister (ppfister) 
Hello.

I didn't try it.
But the hardest is probably not to build VPP in an OpenWrt system.
I'd dare say it would be easy if OpenWrt is compiled for x86 target.

The real question is what platform do you want to compile OpenWrt for ?

- Pierre


Le 9 sept. 2017 à 03:18, yug...@telincn.com a écrit :

Hi all,
Is there anyone try to build vpp on openwrt?

Regards,
Ewan


yug...@telincn.com
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Load Balancer like HAProxy or AWS ELB?

2017-07-24 Thread Pierre Pfister (ppfister) 
Hello Choonho,

MagLev is a layer 4 load-balancer. It load balances TCP connexions to a set of 
targets.
This is done by simply forwarding packets on a per-5-tuple basis with a flow 
table to remember decisions.
Since the destination address of packets is not modified, a tunnel is necessary.

HAProxy is an application layer load balancer, performing an HTTP proxy 
function.
HAProxy terminates all TCP flows, and creates upstream TCP flows with targets.
This approach induces a significant overhead due to the computational and 
memory cost of maintaining two TCP connexions.
It also has consequences on TCP options support and passthrough, as well as 
performances (delay and/or throughput).

Using an application layer load balancer has advantages, but avoiding GRE 
tunnel is not a good one, unless you have absolutely no control over the target 
servers.

Since VPP has a TCP stack, it is possible to build an HTTP Proxy with VPP, and 
I would be curious to see the performances of such a thing.

- Pierre


> Le 24 juil. 2017 à 05:43, Choonho Son  a écrit :
> 
> Hi all,
> 
> Current LB plugin is similar with Google's MagLev which connects LB and 
> Target(Application Server) with GRE tunnel.
> 
> But most of LBs like HAProxy or AWS ELB do not have GRE tunnel between LB and 
> target.
> 
> How about making new Plugin like HAProxy.
> 
> Thanks,
> Choonho
> 
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Requirement on Load Balancer plugin for VPP

2017-04-25 Thread Pierre Pfister (ppfister) 

Le 25 avr. 2017 à 09:52, Ni, Hongjun 
<hongjun...@intel.com<mailto:hongjun...@intel.com>> a écrit :

Hi Pierre,

For LB distribution case, I think we could assign a node IP for each LB box.
When received packets from client, LB will do both SNAT and DNAT. i.e. source 
IP -> LB’s Node IP, destination IP -> AS’s IP.
When returned packets from AS, LB also do both DNAT and SNAT. i.e. source IP -> 
AS’s IP, destination IP -> Client’s IP.

I see.
Doing so you completely hide the client's source address from the application.
You also require per-connexion binding at the load balancer (MagLev does 
per-connexion binding, but in a way which allows for hash collisions, because 
it is not a big deal if two flows use the same entry in the hash table. This 
allows for smaller and fixed size hash table, which also provides a performance 
advantage to MagLev).

In my humble opinion, using SNAT+DNAT is a terribly bad idea, so I would advise 
you to reconsider finding a way to either:
- Enable any type of packet tunneling protocol in your ASs (IPinIP, L2TP, 
whatever-other-protocol, and extend VPP's LB plugin with the one you pick).
- Put some box closer to the ASs (bump in the wire) for decap.
- If your routers support MPLS, you could also use it as encap.

If you really want to use SNAT+DNAT (god forbid), and are willing to suffer (or 
somehow like suffering), you may try to:
- Use VPP's SNAT on the client-facing interface. The SNAT will just change 
clients source addresses to one of LB's source addresses.
- Extend VPP's LB plugin to support DNAT "encap".
- Extend VPP's LB plugin to support return traffic and stateless SNAT base on 
LB flow table (And find a way to make that work on multiple cores...).
The client->AS traffic, in VPP, would do ---> client-facing-iface --> SNAT --> 
LB(DNAT) --> AS-facing-iface
The AS->client traffic, in VPP, would do ---> AS-facing-iface --> LB(Stateless 
SNAT) --> SNAT Plugin (doing DNAT-back) --> client-facing-iface

Now the choice is all yours.
But I will have warned you.

Cheers,

- Pierre


Thanks,
Hongjun

From: Pierre Pfister (ppfister) [mailto:ppfis...@cisco.com]
Sent: Tuesday, April 25, 2017 3:12 PM
To: Zhou, Danny <danny.z...@intel.com<mailto:danny.z...@intel.com>>
Cc: Ni, Hongjun <hongjun...@intel.com<mailto:hongjun...@intel.com>>; Ed 
Warnicke <hagb...@gmail.com<mailto:hagb...@gmail.com>>; Li, Johnson 
<johnson...@intel.com<mailto:johnson...@intel.com>>; 
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
Subject: Re: [vpp-dev] Requirement on Load Balancer plugin for VPP

Hello all,

As mentioned by Ed, introducing return traffic would dramatically reduce the 
performance of the solution.
-> Return traffic typically consists of data packets, whereas forward traffic 
mostly consists of ACKs. So you will have to have significantly more LB boxes 
if you want to support all your return traffic.
-> Having to deal with return traffic also means that we need to either make 
sure return traffic goes through the same core, or add locks to the structures 
(for now, everything is lockless, per-core), or steer traffic for core to core.

There also is something that I am not sure to understand. You mentioned DNAT in 
order to steer the traffic to the AS, but how do you make sure the return 
traffic goes back to the LB ? My guess is that all the traffic coming out of 
the ASs is routed toward one LB, is that right ? How do you make sure the 
return traffic is evenly distributed between LBs ?

It's a pretty interesting requirement that you have, but I am quite sure the 
solution will have to be quite far from MagLev's design, and probably less 
efficient.

- Pierre


Le 25 avr. 2017 à 05:11, Zhou, Danny 
<danny.z...@intel.com<mailto:danny.z...@intel.com>> a écrit :

Share  my two cents as well:

Firstly, introducing GRE or whatever other tunneling protocols to LB introduces 
performance overhead (for encap and decap) to both the load balancer as well as 
the network service. Secondly, other mechanism on the network service node not 
only needs to decap the GRE but also needs to perform a DNAT operation in order 
to change the destination IP of the original frame from LB’s IP to the service 
entity’s IP, which introduces the complexity to the network service.

Existing well-known load balancers such as Netfilter or Nginx do not adopt this 
tunneling approach, they just simply do a service node selection followed by a 
NAT operation.

-Danny

From: vpp-dev-boun...@lists.fd.io<mailto:vpp-dev-boun...@lists.fd.io> 
[mailto:vpp-dev-boun...@lists.fd.io] On Behalf Of Ni, Hongjun
Sent: Tuesday, April 25, 2017 11:05 AM
To: Ed Warnicke <hagb...@gmail.com<mailto:hagb...@gmail.com>>
Cc: Li, Johnson <johnson...@intel.com<mailto:johnson...@intel.com>>; 
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
Subject: Re: [vpp-dev] Requi

Re: [vpp-dev] Tap IF Names

2017-03-28 Thread Pierre Pfister (ppfister) 

Le 28 mars 2017 à 13:54, Jon Loeliger 
<j...@netgate.com<mailto:j...@netgate.com>> a écrit :

On Tue, Mar 28, 2017 at 11:59 AM, Jon Loeliger 
<j...@netgate.com<mailto:j...@netgate.com>> wrote:


On Tue, Mar 28, 2017 at 11:50 AM, Damjan Marion (damarion) 
<damar...@cisco.com<mailto:damar...@cisco.com>> wrote:

> On 28 Mar 2017, at 12:15, Kinsella, Ray 
> <ray.kinse...@intel.com<mailto:ray.kinse...@intel.com>> wrote:
>
> +1 to Jon's comments.
>
>
> On 24/03/2017 14:07, Pierre Pfister (ppfister) wrote:
>> Hello Jon,
>>
>> No strong opinion on my side, but I'd just like to notice that there might 
>> be cases where multiple interfaces, in linux, have the same name, if they 
>> are in different network namespaces.
>> VPP could literally control the back-end of thousands of containers' 
>> interfaces, all called eth0.
>
> Well your backend and your frontend device names are typically named 
> different. You are correct in that frontend device in the container is always 
> eth0, the backend device for each container's eth0 is uniquely named in the 
> default network namespace.

Can somebody come up with the patch proposal?

Thanks!

Happy to do so.  One question, though: what is the purpose of
the "renumber" and "custom_dev_instance" fields?

Thanks,
jdl

OK.  I've read through bunch of that custom_dev_instance code now.
I was initially not paying attention to it, as it stumbled into the so-called
"tapcli" data, and we're not using the VPP CLI.

I think the dev_custom_instance will do what we want (Allow the user
to specify the number that will be used as the "show" name for the
corresponding tap interface.  Good.).

However, I now have a different issue with this code.  Almost all the
code in the tapcli.c file is implementation that should be shared by
both the CLI and the API.  RIght now, the API sort of cannibalizes
into the CLI code.

Instead, there should be three files here: the CLI, the API and the
underlying implementation of Tap data structures.

tapcli.c implements all tap nodes, plus the CLI calls.
tapapi.c only implements api calls and then calls tapcli functions.

Admittedly tapcli is not the perfect name, but file renaming in git makes the 
history of changes way harder to follow.
So I don't think it is necessary to rename tapcli.c

tuntap.c, on the other hand, I think is legacy code. There probably is some 
cleaning to be done there.

- Pierre


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Building router plugin

2017-03-27 Thread Pierre Pfister (ppfister) 
You basically have two options:
1. Teach your routing protocol daemon to directly talk to VPP for configuring 
the FIB.
2. Do not modify the routing daemon and let VPP map routes written in the 
kernel.

For 1, you will need the router plugin, but for packet punting only.

For 2., you will need both netlink+router plugins. The netlink plugin (which 
actually more like a library), shouldn't require so much changes though.

- Pierre


Le 26 mars 2017 à 16:31, Łukasz Chrustek 
> a écrit :

Hello,

thank all of You for clarification of status router plugin.

I'm trying to use vpp+dpdk as bare metal linux ruter with bgp and ospf routing 
protocols, goal is to make very fast router with dynamic routing protocols. Can 
You advice is there any other way to exchange dynamic routing information 
between os and vpp ?

Regards
Lukasz


The mailman did not let my last message go through because it was too 
big so I am removing the previous threads to keep this message under 100KB 
limit.

Hello,

Dave and Pierre are correct, the router plugin is no longer being maintained. 
There was not much interest in the community for this project at the time and I 
have since moved on to other things.  I’d be happy to help a new maintainer 
with any questions they may have.

If you really want to just test it out, I suggest attempting to checkout a 
commit of VPP close to the last commit on the router plugin. Somewhere closely 
following VPP commit 0bfe5d8c792abcdb, “A Protocol Independent Hierarchical FIB 
(VPP-352)” will work (you might have to git bisect to get something that 
compiles right away).

If necessary, I can submit a patch to remove the router plugin from the VPP 
sandbox with a note on the wiki page mentioning the finalization of the 
project. I recall there being an 8-month time limit for these the sandbox 
projects.

-Jeff





___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Building router plugin

2017-03-26 Thread Pierre Pfister (ppfister) 
Hello Lukasz,

I am sorry I won't be able to look in details into your problem this week, but 
I can still quickly dump some information.

netlink and router plugin are really maintained. It is not surprising that they 
do not compile without at bit of effort.
Commits are very welcome though, as I know that VPP integration into Linux is 
of interest to many people.

My guess is that netlink should just need some changes in the way it is 
compiled, and that's all.
router on the other way... I think it hasn't been updated since last FIB change 
in VPP, and will therefore probably need a little bit of work to compile and 
function properly.

Thanks,

- Pierre

Le 24 mars 2017 à 19:38, Łukasz Chrustek 
> a écrit :

Hi,

I'm trying to compile vpp from git sources (git clone 
https://gerrit.fd.io/r/vpp) + router plugin (also from git, but for vppsb: git 
clone https://gerrit.fd.io/r/vppsb). Compilation of vpp is working, but I would 
like to test tap inject from router plugin from vppsb (vpp sandbox). 
Compilation procedure from router/README.mk in vppsb directory doesn't work:


## Build/Install

The router is implemented as a plugin to inject, e.g. arp, icmp4, traffic
from data plane devices. To use it, you must build the plugin and put it
in VPPs runtime plugin directory. The plugin depends on vpp and the netlink
repository from the vppsb project. This README assumes familiarity with the
build environment for both projects.

Build vpp, netlink, and router all at once by creating symbolic links in the
top level vpp directory to the netlink and router directories as well as
symbolic links to the respective .mk files in 'build-data/packages'.

```
$ cd /git/vpp
$ ln -sf /git/vppsb/netlink
$ ln -sf /git/vppsb/router
$ ln -sf ../../netlink/netlink.mk build-data/packages/
$ ln -sf ../../router/router.mk build-data/packages/
```

Now build everything and create a link to the plugin in vpp's plugin path.

```
$ cd build-root
$ ./bootstrap.sh
$ make V=0 PLATFORM=vpp TAG=vpp_debug router-install
$ ln -sf 
/git/vpp/build-root/install-vpp_debug-native/router/lib64/router.so.0.0.0 \
/usr/lib/vpp_plugins/router.so
```

Once VPP is running and the plugin is loaded, data plane interfaces can
be tapped.

```
$ vppctl tap inject arp,icmp4 from TenGigabitEthernet2/0/0 as vpp0
```



Regards
Luk


Luk,

Are you installing the vpp-dev (or vpp-devel if Centos) package?
I would imagine you'd need that for headers etc.

Ed

On Fri, Mar 24, 2017 at 8:14 AM, Łukasz Chrustek 
> wrote:

Hi Dave,

but this plugin isn't part of core vpp code tree, it is from vppsb.

Regards
Luk
> Dear Luk,

> The "vpp-install," "install-packags," "install-deb" etc. targets
> will build the set of plugins configured in 
> src/configure.ac:

> For example:

> $ cd build-root
> $ make PLATFORM=vpp TAG=vpp vpp-install

> HTH. Dave

> -Original Message-
> From: vpp-dev-boun...@lists.fd.io
> [mailto:vpp-dev-boun...@lists.fd.io] On 
> Behalf Of Lukasz Chrustek
> Sent: Friday, March 24, 2017 4:55 AM
> To: vpp-dev@lists.fd.io
> Subject: [vpp-dev] Building router plugin

> Hi,

> Can You advice what is the proper way for building router plugin now ?

> I'm trying to build this plugin as stated in README, but I get:

> # ./bootstrap.sh

> Saving PATH settings in /git/vpp/build-root/path_setup
> Source this file later, as needed
> Compile native tools
>  Arch for platform 'native' is native 
>  Finding source for tools 
>  Makefile fragment found in
> /git/vpp/build-root/packages/tools.mk 
>  Source found in /git/vpp/src 
>  Configuring tools: nothing to do 
>  Building tools: nothing to do 
>  Installing tools: nothing to do 

> # make V=0 PLATFORM=vpp TAG=vpp_debug router-install
>  Arch for platform 'vpp' is native 
>  Finding source for vppinfra 
>  Package vppinfra not found with path /git/vpp 
> Makefile:780: recipe for target 'vppinfra-find-source' failed
> make: *** [vppinfra-find-source] Error 1

> compilation  of  vpp is working fine, but not from /git/vpp/build-root
> but from /git/vpp.

> Regards
> Luk

> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev



--
Pozdrawiam,
Łukasz Chrustek

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev




--
Pozdrowienia,
Łukasz Chrustek
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing

Re: [vpp-dev] Tap IF Names

2017-03-24 Thread Pierre Pfister (ppfister) 
Hello Jon,

No strong opinion on my side, but I'd just like to notice that there might be 
cases where multiple interfaces, in linux, have the same name, if they are in 
different network namespaces.
VPP could literally control the back-end of thousands of containers' 
interfaces, all called eth0.

I know this is a rather specific use-case, so I understand it might be helpful 
to change the default, but don't make the assumption that linux interface names 
are unique.

- Pierre


> Le 23 mars 2017 à 20:55, Jon Loeliger  a écrit :
> 
> Guys,
> 
> Why do tap interfaces use the name "tap-%d" internally?  The user has
> given an actual name (that corresponds to the underlying Linux tap name),
> and yet VPP makes up a non-correlatable different name in the form "tap-%d"
> by making up some number and assigning it to the IF.
> 
> This issue is precisely the same problem that loopback IF names had,
> and for which I submitted a patch to at least allow deterministically 
> predictable
> IF names for user-assigned names on loopbacks.
> 
> Later, there is some form of ability to "renumber" tap interfaces.  Will this
> affect the numbering of the IF names the user sees?
> 
> Specifically, I'm wondering if we can somehow unify the tap name prefix
> with the other interface prefixing styles in a more standard, cohesive and
> predictable way.  In the case of tap IFs, having VPP instead at least use
> the name "tap-%s", formatted with the user-supplied name would make
> it predictable and somewhat similar to the other IF types.
> 
> I *think* this table summarizes the current state of IF names for many
> of the interface types.
> 
> Thanks,
> jdl
> 
> 
> 
> DPDK:
> User says : TenGigabitEthernet6/0/0
> VPP uses  : TenGigabitEthernet6/0/0
> Linux uses: 
> 
> Loopback:
> User says :- for "create" and "delete" commands
> VPP uses  : loop   - for any other reference to the IF
> Linux : 
> 
> Host AF_PACKET
> User says : - for "create" and "delete" commands
> VPP uses  : host-   - for any other reference to the IF
> Linux uses: 
> Note: Leftover(?) use of "veth" in all VPP Wiki pages still?
> 
> Tap:
> User says : - for "create" and "delete" commands
> VPP uses  : tap-   - for any other reference to the IF
> Linux uses: 
> Note: No relationship between  and 
> 
> netmap:
> User says : - for "create" and "delete" commands
> VPP uses  : netmap- - for any other reference to the IF
> Linux uses: 
> 
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] classify and redirect to tap

2017-02-08 Thread Pierre Pfister (ppfister) 
Hello John,

IIRC tap inject is in VPP Sandbox "router" sub-project.
It is an experimental plugin allowing for doing what you are looking for.
It was intended to provide a way to run a routing protocol in linux and reflect 
the result into VPP.

I don't know if it still compiles, but it definitely would be useful to have 
someone updating it.

Cheers,

- Pierre


> Le 7 févr. 2017 à 06:56, John Pearson  a écrit :
> 
> Hi, I am trying to send/redirect outgoing tcp port 5000 traffic to tap device.
> 
> I found another thread mention “tap inject” but couldn’t find any 
> documentation on it and “inject” is an unknown input in vpp CLI.
> 
> Any advice?
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] how to setup topology to test lb plugin?

2017-01-11 Thread Pierre Pfister (ppfister) 


Le 11 janv. 2017 à 14:42, 黄登辉 
<denghuihu...@163.com<mailto:denghuihu...@163.com>> a écrit :

Yes, I read the doc and maglev, one question , do we need to create two 
interfaces? One is for VIP , the other is for communicating as?

No. It's all encap/decap. So it should work with a single interface.

- Pierre



发自网易邮箱手机版


在2017年01月11日 20:16 ，Pierre Pfister (ppfister)<mailto:ppfis...@cisco.com>写道：

Hello,

Are you familiar with MagLev ?
Did you take a look at 
https://git.fd.io/vpp/tree/src/plugins/lb/lb_plugin_doc.md ?

Cheers,

- Pierre

Le 11 janv. 2017 à 12:52, 黄登辉 
<denghuihu...@163.com<mailto:denghuihu...@163.com>> a écrit :

Hi  all
 There is lb plugin available in vpp. I want to test it, could you please 
to tell me how to setup topology for testing?


发自网易邮箱手机版
___
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] how to setup topology to test lb plugin?

2017-01-11 Thread Pierre Pfister (ppfister) 
Hello,

Are you familiar with MagLev ?
Did you take a look at 
https://git.fd.io/vpp/tree/src/plugins/lb/lb_plugin_doc.md ?

Cheers,

- Pierre

Le 11 janv. 2017 à 12:52, 黄登辉 
> a écrit :

Hi  all
 There is lb plugin available in vpp. I want to test it, could you please 
to tell me how to setup topology for testing?


发自网易邮箱手机版
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] vHost user test scenarios for CSIT - TWS meeting

2016-12-12 Thread Pierre Pfister (ppfister) 
to discuss next steps on csit weekly call tomorrow, details here:
https://wiki.fd.io/view/CSIT/Meeting

-Maciek

On 28 Nov 2016, at 07:37, Thomas F Herbert 
<<mailto:therb...@redhat.com>therb...@redhat.com<mailto:therb...@redhat.com>> 
wrote:


All,

At last week's CSIT meeting, Maciek 
(mkons...@cisco.com<mailto:mkons...@cisco.com>) offered to compile a summary 
suggestions on this mailing list.

On 11/22/2016 11:34 AM, Pierre Pfister (ppfister) wrote:
Hello Thomas,

Sorry I haven't reached out faster, I was travelling.

Please have a look at vppsb/vhost-test
It includes a standalone script which provides VPP and VM configuration for PVP 
tests.
- Runs testpmd in the VM
- Supports various CPU configuration for VPP
- Can run with or without gdb, debug or release

Not committed yet:
- Supports VM restart
- Support for VPP restart
- Support for multiple additional (dead) vhost interface

I did that outside of the context of CSIT so people can:
- Look at it and see what are the optimisations that are used
- Use it without CSIT

I will keep using and improving it because I use it for my own development and 
testing purposes.

Rest of this inline.

Le 8 nov. 2016 à 22:25, Thomas F Herbert 
<<mailto:therb...@redhat.com>therb...@redhat.com<mailto:therb...@redhat.com>> a 
écrit :


All:

Soliciting opinions from people as to vhost-user testing scenarios and guest 
modes in fd.io<http://fd.io/> CSIT testing of VPP - vhost-user.

I will forward to this mailing list as well as summarize any additional 
feedback.

I asked some people that happen to be here at OVSCON as well as some Red Hat 
and Intel people. I am also including some people that are involved in upstream 
vhost-user work in DPDK.

So far, I have the following feedback with an attempt to condense feedback and 
to keep the list small. If I left out anything, let me know.

In addition to the PVP tests done now with small packets.

Testpmd in guest is OK for now.

1 Add multiple VMs (How many?)

Makes sense to me. 2 is enough (4 would be good number).

2 Both multi-queue and single-queue

Yes. Ideally, 1-2-4 queues.
With different number of workers (0 workers, i.e. single VPP thread, 1 worker, 
queues*2 workers).

3 Tests that cause the equivalent of multiple flows in OVS. Varying variety of 
traffic including layer 2 and layer 3 traffic.

Yes. Should test with L2 and L3.

4 Multiple IF's (Guest or Host or Both?)

Possibly.
But more importantly, I think, we need to have VM restart and interface restart 
(delete - create).
OpenStack integration generates a significant amount of delete-recreate of 
vhost interface.

The following might not be doable by 17.01 and if not consider the following as 
a wish list for future:

1 vxLan tunneled traffic

2 VPP in guest with layer 2 and layer 3 vRouted traffic.

3 Additional Overlay/Underlay: MPLS

--TFH
--
Thomas F Herbert
SDN Group
Office of Technology
Red Hat
___
vpp-dev mailing list
vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>
https://lists.fd.io/mailman/listinfo/vpp-dev


--
Thomas F Herbert
SDN Group
Office of Technology
Red Hat


--
Thomas F Herbert
SDN Group
Office of Technology
Red Hat


--
Thomas F Herbert
SDN Group
Office of Technology
Red Hat





___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] vHost user test scenarios for CSIT

2016-11-22 Thread Pierre Pfister (ppfister) 
Hello Thomas,

Sorry I haven't reached out faster, I was travelling.

Please have a look at vppsb/vhost-test
It includes a standalone script which provides VPP and VM configuration for PVP 
tests.
- Runs testpmd in the VM
- Supports various CPU configuration for VPP
- Can run with or without gdb, debug or release

Not committed yet:
- Supports VM restart
- Support for VPP restart
- Support for multiple additional (dead) vhost interface

I did that outside of the context of CSIT so people can:
- Look at it and see what are the optimisations that are used
- Use it without CSIT

I will keep using and improving it because I use it for my own development and 
testing purposes.

Rest of this inline.

Le 8 nov. 2016 à 22:25, Thomas F Herbert 
> a écrit :


All:

Soliciting opinions from people as to vhost-user testing scenarios and guest 
modes in fd.io CSIT testing of VPP - vhost-user.

I will forward to this mailing list as well as summarize any additional 
feedback.

I asked some people that happen to be here at OVSCON as well as some Red Hat 
and Intel people. I am also including some people that are involved in upstream 
vhost-user work in DPDK.

So far, I have the following feedback with an attempt to condense feedback and 
to keep the list small. If I left out anything, let me know.

In addition to the PVP tests done now with small packets.

Testpmd in guest is OK for now.

1 Add multiple VMs (How many?)

Makes sense to me. 2 is enough (4 would be good number).

2 Both multi-queue and single-queue

Yes. Ideally, 1-2-4 queues.
With different number of workers (0 workers, i.e. single VPP thread, 1 worker, 
queues*2 workers).

3 Tests that cause the equivalent of multiple flows in OVS. Varying variety of 
traffic including layer 2 and layer 3 traffic.

Yes. Should test with L2 and L3.

4 Multiple IF's (Guest or Host or Both?)

Possibly.
But more importantly, I think, we need to have VM restart and interface restart 
(delete - create).
OpenStack integration generates a significant amount of delete-recreate of 
vhost interface.

The following might not be doable by 17.01 and if not consider the following as 
a wish list for future:

1 vxLan tunneled traffic

2 VPP in guest with layer 2 and layer 3 vRouted traffic.

3 Additional Overlay/Underlay: MPLS

--TFH
--
Thomas F Herbert
SDN Group
Office of Technology
Red Hat
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Creating vpp-test-verify job ... - it's time I think !

2016-11-08 Thread Pierre Pfister (ppfister) 
Thanks for the quick response,

Please have a look at your change: 
https://gerrit.fd.io/r/#/c/3345/17/test/test_lb.py

Really the issue is that you uncommented checkCapture in all v6 related 
scenarios.
checkCapture should only be executed for v4 in gre4.

It should work if you just comment these lines again.

- Pierre


> Le 8 nov. 2016 à 17:49, Klement Sekera -X (ksekera - PANTHEON TECHNOLOGIES at 
> Cisco) <ksek...@cisco.com> a écrit :
> 
> Hi Pierre,
> 
> I've enabled the tests and remove the checkes from the ipv6 branch in
> checkCapture as suggested.
> 
> Now I still see failures, even a vpp crash.
> 
> Please see the attached diff and let me know if I need to do something
> else.
> 
> Attached also the output of the test case..
> 
> Thanks,
> Klement
> 
> On Tue, Nov 08, 2016 at 04:16:18PM +, Pierre Pfister (ppfister) wrote:
>> Right.
>> 
>> Klement Sekera was kindly trying to make testing with IPv6 and GRE6 work (It 
>> was partially tested, but there are issues with scapy).
>> Tests are failing because he enabled those tests.
>> I will suggest on the gerrit patch that he rolls-back to the previous test 
>> situation.
>> 
>> - Pierre
>> 
>> 
>>> Le 8 nov. 2016 à 16:37, Maciek Konstantynowicz (mkonstan) 
>>> <mkons...@cisco.com> a écrit :
>>> 
>>> Oups… Agree. I know someone was working on it.. but it’s still failing.. 
>>> And I don’t know how much work to make it stop doing so..
>>> If you could help that would be great.
>>> 
>>> $ git blame test/test_lb.py
>>> 
>>> -Maciek
>>> 
>>>> On 8 Nov 2016, at 15:23, Pierre Pfister (ppfister) <ppfis...@cisco.com> 
>>>> wrote:
>>>> 
>>>> Hello Maciek,
>>>> 
>>>> Do you know what is going wrong with the LB test case ?
>>>> Nobody bothered to tell me the test was failing.
>>>> It would be cool to have a chance to fix it rather than just disabling it.
>>>> 
>>>> Thanks,
>>>> 
>>>> - Pierre
>>>> 
>>>> 
>>>> 
>>>>> Le 8 nov. 2016 à 15:57, Maciek Konstantynowicz (mkonstan) 
>>>>> <mkons...@cisco.com> a écrit :
>>>>> 
>>>>> Hi,
>>>>> 
>>>>> Following from recent discussions on weekly VPP call, IMV the test 
>>>>> framework and 
>>>>> tests in master branch are ready to be used regularly on a per patch 
>>>>> basis.
>>>>> 
>>>>> Hence I propose someone to create a vpp-test-verify* job(s) to do exactly 
>>>>> that.
>>>>> Ed, could you do this or assist some other volunteer to do this?
>>>>> For ubuntu1404, ubuntu1604, centos7, etc.
>>>>> 
>>>>> The only failing test due to unfinished port to somewhat refactored test 
>>>>> framework is test_lb.py,
>>>>> and Klement just pushed a patch that will skip this test: 
>>>>> https://gerrit.fd.io/r/#/c/3726/
>>>>> 
>>>>> Thoughts?
>>>>> 
>>>>> -Maciek
>>>>> 
>>>>> P.S. Current tests in vpp master branch are: 
>>>>> $ ls test
>>>>> doclog.py scapy_handlers   test_ip6.pyc   
>>>>> test_l2xc.py   test_mpls.pyc   vpp_interface.py   vpp_pg_interface.pyc
>>>>> framework.py   log.pyctemplate_bd.py   test_ip.py 
>>>>> test_l2xc.pyc  test_vxlan.py   vpp_interface.pyc  vpp_sub_interface.py
>>>>> framework.pyc  Makefile   template_bd.pyc  test_ip.pyctest_lb.py  
>>>>>test_vxlan.pyc  vpp_papi_provider.py   vpp_sub_interface.pyc
>>>>> hook.pyrun_tests.py   test_infra.mdtest_l2bd.py   test_lb.pyc 
>>>>>util.py vpp_papi_provider.pyc
>>>>> hook.pyc   run_tests.pyc  test_ip6.py  test_l2bd.pyc  
>>>>> test_mpls.py   util.pycvpp_pg_interface.py
>>>>> 
>>>>> $ make test
>>>>> <skipped…>
>>>>> ==
>>>>> IPv6 Test Case
>>>>> ==
>>>>> IPv6 FIB test   OK
>>>>> ==
>>>>> MPLS Test Case
>>

Re: [vpp-dev] Creating vpp-test-verify job ... - it's time I think !

2016-11-08 Thread Pierre Pfister (ppfister) 
Right.

Klement Sekera was kindly trying to make testing with IPv6 and GRE6 work (It 
was partially tested, but there are issues with scapy).
Tests are failing because he enabled those tests.
I will suggest on the gerrit patch that he rolls-back to the previous test 
situation.

- Pierre


> Le 8 nov. 2016 à 16:37, Maciek Konstantynowicz (mkonstan) 
> <mkons...@cisco.com> a écrit :
> 
> Oups… Agree. I know someone was working on it.. but it’s still failing.. 
> And I don’t know how much work to make it stop doing so..
> If you could help that would be great.
> 
> $ git blame test/test_lb.py
> 
> -Maciek
> 
>> On 8 Nov 2016, at 15:23, Pierre Pfister (ppfister) <ppfis...@cisco.com> 
>> wrote:
>> 
>> Hello Maciek,
>> 
>> Do you know what is going wrong with the LB test case ?
>> Nobody bothered to tell me the test was failing.
>> It would be cool to have a chance to fix it rather than just disabling it.
>> 
>> Thanks,
>> 
>> - Pierre
>> 
>> 
>> 
>>> Le 8 nov. 2016 à 15:57, Maciek Konstantynowicz (mkonstan) 
>>> <mkons...@cisco.com> a écrit :
>>> 
>>> Hi,
>>> 
>>> Following from recent discussions on weekly VPP call, IMV the test 
>>> framework and 
>>> tests in master branch are ready to be used regularly on a per patch basis.
>>> 
>>> Hence I propose someone to create a vpp-test-verify* job(s) to do exactly 
>>> that.
>>> Ed, could you do this or assist some other volunteer to do this?
>>> For ubuntu1404, ubuntu1604, centos7, etc.
>>> 
>>> The only failing test due to unfinished port to somewhat refactored test 
>>> framework is test_lb.py,
>>> and Klement just pushed a patch that will skip this test: 
>>> https://gerrit.fd.io/r/#/c/3726/
>>> 
>>> Thoughts?
>>> 
>>> -Maciek
>>> 
>>> P.S. Current tests in vpp master branch are: 
>>> $ ls test
>>> doclog.py scapy_handlers   test_ip6.pyc   test_l2xc.py  
>>>  test_mpls.pyc   vpp_interface.py   vpp_pg_interface.pyc
>>> framework.py   log.pyctemplate_bd.py   test_ip.py test_l2xc.pyc 
>>>  test_vxlan.py   vpp_interface.pyc  vpp_sub_interface.py
>>> framework.pyc  Makefile   template_bd.pyc  test_ip.pyctest_lb.py
>>>  test_vxlan.pyc  vpp_papi_provider.py   vpp_sub_interface.pyc
>>> hook.pyrun_tests.py   test_infra.mdtest_l2bd.py   test_lb.pyc   
>>>  util.py vpp_papi_provider.pyc
>>> hook.pyc   run_tests.pyc  test_ip6.py  test_l2bd.pyc  test_mpls.py  
>>>  util.pycvpp_pg_interface.py
>>> 
>>> $ make test
>>> <skipped…>
>>> ==
>>> IPv6 Test Case
>>> ==
>>> IPv6 FIB test   OK
>>> ==
>>> MPLS Test Case
>>> ==
>>> MPLS V4 Explicit NULL test  OK
>>> MPLS V6 Explicit NULL test  OK
>>> ==
>>> IPv4 Test Case
>>> ==
>>> IPv4 FIB test   OK
>>> ==
>>> L2XC Test Case
>>> ==
>>> L2XC test   OK
>>> ==
>>> VXLAN Test Case
>>> ==
>>> Decapsulation test  OK
>>> Encapsulation test  OK
>>> ==
>>> L2BD Test Case
>>> ==
>>> L2BD MAC learning test  OK
>>> ___
>>> vpp-dev mailing list
>>> vpp-dev@lists.fd.io
>>> https://lists.fd.io/mailman/listinfo/vpp-dev
> 

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Creating vpp-test-verify job ... - it's time I think !

2016-11-08 Thread Pierre Pfister (ppfister) 
Hello Maciek,

Do you know what is going wrong with the LB test case ?
Nobody bothered to tell me the test was failing.
It would be cool to have a chance to fix it rather than just disabling it.

Thanks,

- Pierre



> Le 8 nov. 2016 à 15:57, Maciek Konstantynowicz (mkonstan) 
>  a écrit :
> 
> Hi,
> 
> Following from recent discussions on weekly VPP call, IMV the test framework 
> and 
> tests in master branch are ready to be used regularly on a per patch basis.
> 
> Hence I propose someone to create a vpp-test-verify* job(s) to do exactly 
> that.
> Ed, could you do this or assist some other volunteer to do this?
> For ubuntu1404, ubuntu1604, centos7, etc.
> 
> The only failing test due to unfinished port to somewhat refactored test 
> framework is test_lb.py,
> and Klement just pushed a patch that will skip this test: 
> https://gerrit.fd.io/r/#/c/3726/
> 
> Thoughts?
> 
> -Maciek
> 
> P.S. Current tests in vpp master branch are: 
> $ ls test
> doclog.py scapy_handlers   test_ip6.pyc   test_l2xc.py   
> test_mpls.pyc   vpp_interface.py   vpp_pg_interface.pyc
> framework.py   log.pyctemplate_bd.py   test_ip.py test_l2xc.pyc  
> test_vxlan.py   vpp_interface.pyc  vpp_sub_interface.py
> framework.pyc  Makefile   template_bd.pyc  test_ip.pyctest_lb.py 
> test_vxlan.pyc  vpp_papi_provider.py   vpp_sub_interface.pyc
> hook.pyrun_tests.py   test_infra.mdtest_l2bd.py   test_lb.pyc
> util.py vpp_papi_provider.pyc
> hook.pyc   run_tests.pyc  test_ip6.py  test_l2bd.pyc  test_mpls.py   
> util.pycvpp_pg_interface.py
> 
> $ make test
> 
> ==
> IPv6 Test Case
> ==
> IPv6 FIB test   OK
> ==
> MPLS Test Case
> ==
> MPLS V4 Explicit NULL test  OK
> MPLS V6 Explicit NULL test  OK
> ==
> IPv4 Test Case
> ==
> IPv4 FIB test   OK
> ==
> L2XC Test Case
> ==
> L2XC test   OK
> ==
> VXLAN Test Case
> ==
> Decapsulation test  OK
> Encapsulation test  OK
> ==
> L2BD Test Case
> ==
> L2BD MAC learning test  OK
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] [csit-dev] vhost multi-queue patch - verify job failing

2016-10-28 Thread Pierre Pfister (ppfister) 
So I tested in my setup with qemu from git with tags: v2.5.0-rc0 v2.5.0-rc1 
v2.5.0 v2.5.1
They all work.

Could you please have a look at the patch and give me some more logs ?
VPP with VHOST_USER_DEBUG_SOCKET set to 1 usually helps a lot, as it logs all 
vhost messages received by VPP.

Thanks,

- Pierre


Le 26 oct. 2016 à 23:45, Maciek Konstantynowicz (mkonstan) 
<mkons...@cisco.com<mailto:mkons...@cisco.com>> a écrit :

Excellent, thanks Ed! And let’s see …

-Maciek

On 26 Oct 2016, at 22:40, Edward Warnicke 
<hagb...@gmail.com<mailto:hagb...@gmail.com>> wrote:

Maciek,

Looks like the patch was based at a point in the past that was pointing to the 
old CSIT branch.

I rebased the patch, so hopefully it will now pick up the correct CSIT branch.

Ed

On Wed, Oct 26, 2016 at 2:15 PM, Maciek Konstantynowicz (mkonstan) 
<mkons...@cisco.com<mailto:mkons...@cisco.com>> wrote:
//reduced list, updated subject

Peter is right: https://gerrit.fd.io/r/#/c/2922/16

looking at the last failed job - 
https://jenkins.fd.io/job/vpp-csit-verify-virl-master/1940/consoleFull - it's 
using old csit branch - "CSIT_BRANCH=oper-161017", without ubuntu1604, so 
pointless. you should rebase to pick up this commit 
https://gerrit.fd.io/r/#/c/3553/ that's updating to latest csit branch 
oper-161024.

-Maciek

On 26 Oct 2016, at 10:06, Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at 
Cisco) <pmi...@cisco.com<mailto:pmi...@cisco.com>> wrote:

Looks like it is still testing with oper-161017 not oper-161024.

+ CSIT_BRANCH=oper-161017
+ git clone https://gerrit.fd.io/r/csit --branch oper-161017

Can you please use the latest one (code rebase)?

Thanks.

Peter Mikus
Engineer – Software
Cisco Systems Limited

Planned absence: 28.10., 1.11., 17.11., 9.12., 19.-31.12.

From: Pierre Pfister (ppfister)
Sent: Wednesday, October 26, 2016 10:52 AM
To: Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at Cisco) 
<pmi...@cisco.com<mailto:pmi...@cisco.com>>
Cc: Andrew Theurer <atheu...@redhat.com<mailto:atheu...@redhat.com>>; Douglas 
Shakshober <dsh...@redhat.com<mailto:dsh...@redhat.com>>; Damjan Marion 
(damarion) <damar...@cisco.com<mailto:damar...@cisco.com>>; Bill Michalowski 
<bmich...@redhat.com<mailto:bmich...@redhat.com>>; vpp-dev 
<vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>>; Rashid Khan 
<rk...@redhat.com<mailto:rk...@redhat.com>>; 
kris...@redhat.com<mailto:kris...@redhat.com>; 
csit-...@lists.fd.io<mailto:csit-...@lists.fd.io>; Edward Warnicke 
<hagb...@gmail.com<mailto:hagb...@gmail.com>>; Thomas F Herbert 
<therb...@redhat.com<mailto:therb...@redhat.com>>
Subject: Re: [vpp-dev] updated ovs vs. vpp results for 0.002% and 0% loss

The recheck failed.
https://jenkins.fd.io/job/vpp-csit-verify-virl-master/1937/console

Are you sure you updated qemu ?

Thanks,

- Pierre



___
csit-dev mailing list
csit-...@lists.fd.io<mailto: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

Re: [vpp-dev] [csit-dev] vhost multi-queue patch - verify job failing

2016-10-27 Thread Pierre Pfister (ppfister) 
Oh. Just notices you merged.
I actually merged and pushed too... So a new test is started. Will probably 
fail.

Ideal Qemu would be 2.7.

2.5 is *old*. And the failing qemu was 2.5-rc1. Anything above 2.5-rc1 should 
be good.

- Pierre


Le 27 oct. 2016 à 07:36, Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at 
Cisco) <pmi...@cisco.com<mailto:pmi...@cisco.com>> a écrit :

Hi,

Taking look. Yes it is using correct latest branch and Qemu 2.5.0 which is 
default version of 16.04.1LTS (Debian 1:2.5+dfsg-5ubuntu10.5). As still 
failing, there is question which Qemu version is specifically required (I do 
not see it in the comments of patch) or if the patch itself is working.

Peter Mikus
Engineer – Software
Cisco Systems Limited

Planned absence: 28.10., 1.11., 17.11., 9.12., 19.-31.12.

From: Edward Warnicke [mailto:hagb...@gmail.com]
Sent: Thursday, October 27, 2016 12:35 AM
To: Maciek Konstantynowicz (mkonstan) 
<mkons...@cisco.com<mailto:mkons...@cisco.com>>
Cc: Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at Cisco) 
<pmi...@cisco.com<mailto:pmi...@cisco.com>>; Pierre Pfister (ppfister) 
<ppfis...@cisco.com<mailto:ppfis...@cisco.com>>; 
csit-...@lists.fd.io<mailto:csit-...@lists.fd.io>; Damjan Marion (damarion) 
<damar...@cisco.com<mailto:damar...@cisco.com>>; vpp-dev 
<vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>>
Subject: Re: [csit-dev] vhost multi-queue patch - verify job failing

Still failing, but apparently now using the correct CSIT branch:

https://jenkins.fd.io/job/vpp-csit-verify-virl-master/1949/consoleFull

"14:55:18 + git clone https://gerrit.fd.io/r/csit --branch oper-161024"

On Wed, Oct 26, 2016 at 2:45 PM, Maciek Konstantynowicz (mkonstan) 
<mkons...@cisco.com<mailto:mkons...@cisco.com>> wrote:
Excellent, thanks Ed! And let’s see …

-Maciek

On 26 Oct 2016, at 22:40, Edward Warnicke 
<hagb...@gmail.com<mailto:hagb...@gmail.com>> wrote:

Maciek,

Looks like the patch was based at a point in the past that was pointing to the 
old CSIT branch.

I rebased the patch, so hopefully it will now pick up the correct CSIT branch.

Ed

On Wed, Oct 26, 2016 at 2:15 PM, Maciek Konstantynowicz (mkonstan) 
<mkons...@cisco.com<mailto:mkons...@cisco.com>> wrote:
//reduced list, updated subject

Peter is right: https://gerrit.fd.io/r/#/c/2922/16

looking at the last failed job - 
https://jenkins.fd.io/job/vpp-csit-verify-virl-master/1940/consoleFull - it's 
using old csit branch - "CSIT_BRANCH=oper-161017", without ubuntu1604, so 
pointless. you should rebase to pick up this commit 
https://gerrit.fd.io/r/#/c/3553/ that's updating to latest csit branch 
oper-161024.

-Maciek

On 26 Oct 2016, at 10:06, Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at 
Cisco) <pmi...@cisco.com<mailto:pmi...@cisco.com>> wrote:

Looks like it is still testing with oper-161017 not oper-161024.

+ CSIT_BRANCH=oper-161017
+ git clone https://gerrit.fd.io/r/csit --branch oper-161017

Can you please use the latest one (code rebase)?

Thanks.

Peter Mikus
Engineer – Software
Cisco Systems Limited

Planned absence: 28.10., 1.11., 17.11., 9.12., 19.-31.12.

From: Pierre Pfister (ppfister)
Sent: Wednesday, October 26, 2016 10:52 AM
To: Peter Mikus -X (pmikus - PANTHEON TECHNOLOGIES at Cisco) 
<pmi...@cisco.com<mailto:pmi...@cisco.com>>
Cc: Andrew Theurer <atheu...@redhat.com<mailto:atheu...@redhat.com>>; Douglas 
Shakshober <dsh...@redhat.com<mailto:dsh...@redhat.com>>; Damjan Marion 
(damarion) <damar...@cisco.com<mailto:damar...@cisco.com>>; Bill Michalowski 
<bmich...@redhat.com<mailto:bmich...@redhat.com>>; vpp-dev 
<vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>>; Rashid Khan 
<rk...@redhat.com<mailto:rk...@redhat.com>>; 
kris...@redhat.com<mailto:kris...@redhat.com>; 
csit-...@lists.fd.io<mailto:csit-...@lists.fd.io>; Edward Warnicke 
<hagb...@gmail.com<mailto:hagb...@gmail.com>>; Thomas F Herbert 
<therb...@redhat.com<mailto:therb...@redhat.com>>
Subject: Re: [vpp-dev] updated ovs vs. vpp results for 0.002% and 0% loss

The recheck failed.
https://jenkins.fd.io/job/vpp-csit-verify-virl-master/1937/console

Are you sure you updated qemu ?

Thanks,

- Pierre



___
csit-dev mailing list
csit-...@lists.fd.io<mailto: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