Skip to Content.
Sympa Menu

rare-users - Re: [RARE-users] [freertr] LACP and VLAN usage configuration

Subject: RARE user and assistance email list

List archive

Re: [RARE-users] [freertr] LACP and VLAN usage configuration


Chronological Thread 
    < Chronological >      < Thread >
  • From: mc36 <>
  • To: Xavier Jeannin <>, "" <>
  • Subject: Re: [RARE-users] [freertr] LACP and VLAN usage configuration
  • Date: Mon, 24 Oct 2022 16:42:51 +0200
hi,

then imho the easiest way to test these is the following:

ppn:
bundle 1
bridge 1
int sdn10
bundle 1
int sdn7
bundle 1
int sdn15.11
bridge 1
int bun1.11
bridge 1

and remove everything else, and do a clear p4 p4

the backbone pe can do the bare minimum:
bun 1
int bun1.11
vrf for inet
ipv4 addr..

and remove everything else, and do a clear p4 p4

imho that is what you mentioned below with the "supporess bridge on
dav0101"...

regarding the strange ping issue, imho you got some garbage in the configs
for now...
(seen similar when you mix the bridge and routed config i've seen in you
original today-v3 configs)
what about a write erase / reload and reconfiguring the boxes only with the
above? :)

br,
cs



On 10/24/22 16:30, Xavier Jeannin wrote:
I removed the Bund1.11
I put below the show bridge outcome

I am a little bit lost

My PPN (DAV0001) will work at L2 - OK
But I need to establish a peering between my client 11.11.11.44 and the
RENATER backbone PE (DAV0101).
It is what I try to do on DAV0101.

I want to terminate L2 VLAN 11 on the RENATER backbone PE (DAV0101)
So
1/ I suppress the bridge on DAV0101
2/ I put an IP@ on sdn5.11 ( 11.11.11.1)

--> is it correct?

I have a very strange behaviour.
I can ping from the linux (11.11.11.44) the IP 11.11.11.1 on DAV0101
but the opposite does not work ...




thank you
Xavier

Le 24/10/2022 16:02, mc36 a crit :
hi.

On 10/24/22 15:57, Xavier Jeannin wrote:

Le 24/10/2022 15:52, mc36 a crit :
uhhh, please remove the ip address from the bvi1 and bun1.11,
as if it have an ip then it'll be exported as routed port...
also please remove the export-port bvi1 dynamic beforehand...
also a clear p4 p4 may be needed after doing so...
after that, if the bridge tables stays the same, it should work...

thank you
I will do it but If I suppress the address on bvi1 and bun1.11, how can I
ping from DAV0001 to DAV0101?

right now, a bridge if exported, then have no addresses, so you cannot ping
in between the boxes...
it only provides layer2 services and no ip addresses can belong to it... as
soon as you do configure
an ip on the bridge, or export it, it becomes a routed bridge, that is, the
box will just minic bridging...
if you want some connectivity check then you need to configure an additional
vlan on the bundle...

I put here show bundle result

nooo please remove that subinterface, it'll be created implicitly when you
configure the bundle...

br,
cs

br
Xavier

br,
cs


On 10/24/22 15:47, Xavier Jeannin wrote:
Hi

Thank you for your help

I just come back and put the IP address on the bviX interface instead of
Bund1.X

Find below the show bridge1
It seems the mac learning works

No the from Bund1.11 and now bvi1 - ping does not work
Would you like I put a IPv4 @ on sdn4.11?



thank you
Xavier

Le 24/10/2022 15:16, mc36 a crit :
hi,
sad to hear that.. does the bundling alone work? i saw a .11 subinterface in
your configs, does that one ping?
also, does the mac learning at least work? could you please share sho bri 1
from both boxes?
thanks,
cs


On 10/24/22 14:37, Xavier Jeannin wrote:
Hello Csaba,

I apply your example but it does not work.
I am pretty sure, it is a misconfiguration :-(

I put the configuration of the 2 switch attached and few show commands below

Thank you
Xavier

DAV0101#show bundle 1
parameter state
backup 0
selected null
replicate false
sequence null
dejitter 0
reporter 0
dynamic 0

interface state quota report
priority
sdn4 up 2000
0
0
sdn5 up 2000
0
0

DAV0101#show bundle 1
parameter state
backup 0
selected null
replicate false
sequence null
dejitter 0
reporter 0
dynamic 0

interface state quota report
priority
sdn4 up 2000
0
0
sdn5 up 2000
0
0

DAV0101#show lacp neighbor
interface system port
state
sdn4 32768.0000.0000.1234.12345 32768.3 63
sdn5 32768.0000.0000.1234.12345 32768.1 63


DAV0001#show bundle 1
parameter state
backup 0
selected null
replicate false
sequence null
dejitter 0
reporter 0
dynamic 0

interface state quota report
priority
sdn7 up 2000
0
0
sdn10 up 2000
0
0

DAV0001#show lacp neighbor
interface system port
state
sdn10 32768.0000.0000.4321.12345 32768.4 63
sdn7 32768.0000.0000.4321.12345 32768.2 63


Le 24/10/2022 13:07, mc36 a crit :
hi,
in general, bundles are created over physical interfaces and not
subinterfaces...
so you have to bundle the sdns togeher, then configure the subinterface on
top of the bundle interface...
much like this: http://sources.freertr.org/cfg/p4lang-rout018.tst
here you create a vlan 11 subinterface on top of sdn3 and sdn4...
br,
cs



On 10/24/22 12:11, Xavier Jeannin wrote:
Hello Csaba,

Thank you for your email.
I want to collect our user traffic and put their traffic in a dedicated VLAN (then established the BGP peering between the RENATER bacbone PE and the client) - see picture below.

My understanding (I am certainly wrong) is that I have to put the interface (sdnXX) in the bundle (not the sub interface) and it should be enough (see configuration attached).
But it does not work
I try also to put the sub-interface but no success.

Could you advice what type of configuration should I use for this use case?

The example available on http://docs.freertr.org/guides/reference/ are

conn-bundle01 bundle of ethernet port
conn-bundle02 bundle of serial port
conn-bundle03 bundle of ethernet ports
conn-bundle04 bundle of serial ports
conn-bundle05 bundle over bridge
conn-bundle06 bridge over bundle
conn-bundle07 replicating bundle
conn-bundle08 unreplicating bundle
conn-bundle09 dejittering bundle
conn-bundle10 loadbalancing bundle
conn-bundle11 multichassis bundle
conn-bundle12 backup bundle

Thank you
Regards
Xavier

Le 20/10/2022 10:03, mc36 a
crit :
hi,
lacp is there:
you have to configure it per member interface with both the lacp knob and the
bundle-group knob...
the first one is just the protocol sender and you can omit if you do "mode
on" on the remote side...
so that latter is the interesting part in the fastpath, and it's heavily
tested with the dataplanes...
br,
cs

On 10/20/22 09:58, wrote:
Hello

I allow myself to post again this request about LACP supporting trunk as if
this would not be implemented, it will impact strongly our use case
deployment.

Thank you
Regards
Xavier

Le 17/10/2022 18:23, Xavier Jeannin a crit :

Hello

I am still stuck with the creation of LACP handling a trunk.
In the meantime, I review my configuration and see that "bridge-group 2" was
not declared in P4 server after correction same pb.
This really something that block this work.

Any suggestion ?

Regards
Xavier

Le 27/09/2022 10:37, Xavier Jeannin a crit :

Hello

I create a testbed on which (see picture attached):

* the client is connected on a WEDGE
(L2-PPN) at Layer 2 using Ethernet
or VLAN
* The WEDGE (L2PPN) should be connected to
our backbone thanks to a bond of interfaces (LACP). The traffic on the LACP
must be a VLAN trunk.

I see a strange behaviour and it is certainly due to how I configure it.

If I put the 2
vlans (11-12) on the all interfaces taking part on the LACP, it does not
work.
If I put the 1 vlan only on each interface of the taking part on the LACP, it works. But in this case the traffic should not be well shared between the LACP links.

I suspect I implement this use case not appropriately (see configuration
attached + picture). Any suggestion?

Regards
Xavier



--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr
--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr

--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr

_._,_._,_
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Groups.io Links:

You receive all messages sent to this group.

View/Reply Online (#768) <https://groups.io/g/freertr/message/768> | Reply To Group <> | Reply To Sender <> | Mute This Topic <https://groups.io/mt/93945667/6006518> | New Topic <https://groups.io/g/freertr/post>
Your Subscription <https://groups.io/g/freertr/editsub/6006518> | Contact Group Owner <> | Unsubscribe <https://groups.io/g/freertr/unsub> []

_._,_._,_
From - Fri Oct 21 08:43:28 2022
X-Mozilla-Status: 0001
X-Mozilla-Status2: 00000000
Return-Path: <>
Received: from zmtain01.partage.renater.fr (LHLO
zmtain01.partage.renater.fr) (194.254.240.20) by
zstore-b1-018.partage.renater.fr with LMTP; Fri, 21 Oct 2022 06:02:16

--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr

--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr

--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr

--
Xavier JEANNIN
Network Dpt.
GN4-3 Network technology evolution (WP6 T1) task leader
T l. : +33 1 5394 2042
Mob. : +33 6 3017 0963
www.renater.fr


Archive powered by MHonArc 2.6.19.

Top of Page