nsh_sfc-dev@lists.fd.io | [vpp-dev] VPP Service chaining with NSH

Home Messages Hashtags Subgroups

Date Date 1 - 20 of 27

[vpp-dev] VPP Service chaining with NSH

Keith Burns <alagalah@...>

(moved vpp-dev to Bcc so folks know this has been responded to, and added appropriate alias nsh_sfc-dev@)

Suman, Jim Guichard (on the nsh_sfc-dev list) has a doc that describes a demo env.

I did some work recently on the classifier that I need to redo before merge but for your example what Jim has will suffice… To paraphrase:

SF1 —> VPP(sff1) —> SF2 —> VPP(sff2) —> SF3

I *believe* this should work as I have shown some python agents emulating SFF/SF (from ODL SFC) around a single VPP instance as an SFF.

The issue is I know I have a bug with the classifier+SFF VPP POC, and I am 90% sure I know where it is, just haven’t had time to run it down until end of VPP training next week.

Please note that I did the NSH SFF work purely as POC, and the newly minted NSH_SFC project (underscore… sigh) will be defining its preferred architecture etc and more than likely doing away with the approach I took (which is perfectly valid, but there were some interesting lessons I learned in doing it, so there’s that….)

Please reply all and it will be great to get some traffic/interest on the nsh_sfc list. (have I mentioned “sigh underscore” ??)

From: <vpp-dev-bounces@...> on behalf of Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 5:32 AM
To: <vpp-dev@...>
Cc: <pratik.dave@...>
Subject: [vpp-dev] VPP Service chaining with NSH

Show quoted text

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help
tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>
[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]
[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]
[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,
Suman

=====-----=====-----=====
Notice: The information contained in this e-mail
message and/or attachments to it may contain
confidential or privileged information. If you are
not the intended recipient, any dissemination, use,
review, distribution, printing or copying of the
information contained in this e-mail message
and/or attachments to it are strictly prohibited. If
you have received this communication in error,
please notify us by reply e-mail or telephone and
immediately and permanently delete the message
and any attachments. Thank you
_______________________________________________ vpp-dev mailing list vpp-dev@... https://lists.fd.io/mailman/listinfo/vpp-dev

James Neil Guichard

[Removing vpp-dev alias and adding nsh_sfc-dev alias ..]

Hi Danny,

Yes its hard-coded for now but we will address as part of the NSH SFC project ..

Jim

From: "Zhou, Danny" <danny.zhou@...>
Date: Friday, April 1, 2016 at 10:28 AM
To: Jim Guichard <jguichar@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>, "vpp-dev@..." <vpp-dev@...>
Cc: "pratik.dave@..." <pratik.dave@...>
Subject: RE: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

It looks the VNI and all the fields in NSH (e.g. SP, SI as well as context fields) are all hard-coded in “nsh vxlan” command line and not be able to modify via any method, right?

The NSH SFC project should address this problem for realistic Service Chianing.

-Danny

From: vpp-dev-bounces@... [mailto:vpp-dev-bounces@...] On Behalf Of Jim Guichard (jguichar)

Show quoted text

Sent: Friday, April 1, 2016 9:32 PM
To: Suman Kpatra/KOL/TCS <suman.kpatra@...>; vpp-dev@...
Cc: pratik.dave@...
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

From: <vpp-dev-bounces@...> on behalf of Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 8:32 AM
To: "vpp-dev@..." <vpp-dev@...>
Cc: "pratik.dave@..." <pratik.dave@...>
Subject: [vpp-dev] VPP Service chaining with NSH

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

=====-----=====-----=====
Notice: The information contained in this e-mail
message and/or attachments to it may contain
confidential or privileged information. If you are
not the intended recipient, any dissemination, use,
review, distribution, printing or copying of the
information contained in this e-mail message
and/or attachments to it are strictly prohibited. If
you have received this communication in error,
please notify us by reply e-mail or telephone and
immediately and permanently delete the message
and any attachments. Thank you

James Neil Guichard

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 10:29 AM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>, "vpp-dev@..." <vpp-dev@...>, Keith Burns <alagalah@...>
Subject: Fwd: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4 represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Jim> segment routing is a completely different technology and should not be confused with NSH. NSH provides a service plane; if you want to use segment routing as the outer transport encapsulation to carry NSH then theoretically you can do that although we have as yet not implemented that in VPP.

Thanks,

Suman

Show quoted text

From: "Suman Kpatra/KOL/TCS" <986788@...>
To: "Jim Guichard (jguichar)" <jguichar@...>
Cc: "pratik dave" <pratik.dave@...>, "Suman Kpatra/KOL/TCS" <suman.kpatra@...>, vpp-dev@..., "Keith Burns" <alagalah@...>, "Abhijeet karve" <abhijeet.karve@...>
Sent: Friday, April 1, 2016 7:24:35 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>, vpp-dev@...
Cc: "pratik dave" <pratik.dave@...>
Sent: Friday, April 1, 2016 7:02:03 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

From: <vpp-dev-bounces@...> on behalf of Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 8:32 AM
To: "vpp-dev@..." <vpp-dev@...>
Cc: "pratik.dave@..." <pratik.dave@...>
Subject: [vpp-dev] VPP Service chaining with NSH

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

--
Thanks & Regards
Suman Kalyan Patra
TCS-Gandhinagar,Garima Park
Tata Consultancy Services Limited
Cell Number: +918981065666
Office Nuber: +917961706230
Extension#6230
Mail to: suman.kpatra@...
Web site: www.tcs.com

James Neil Guichard

#15

[Removing vpp-dev and adding correct alias …]

The correct command is:

sudo python vxlan_tool.py -i 'eth1' -d 'forward' -v ‘on'

However, your configuration is wrong somewhere as looks like you are sending packets straight back to .70 rather than to .71 as per the document:

00:10:18:817818: nsh-vxlan-gpe-encap
NSH-VXLAN-ENCAP: tunnel 3
00:10:18:817821: ip4-rewrite-transit
fib 0 adj-idx 6 : GigabitEthernet0/a/0
IP4: 52:54:00:e6:42:b1 -> 52:54:00:39:34:9e flow hash: 0x00000000
IP4: 52:54:00:e6:42:b1 -> 52:54:00:39:34:9e
UDP: 192.168.50.91 -> 192.168.50.70

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 3:16 PM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "vpp-dev@..." <vpp-dev@...>
Subject: Fwd: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

We tried implementing the NSH service chaining as mentioned in your document. But we are facing some challenge in doing the same

1) we are not getting any output for sudo python vxlan_tool.py i 'eth1' d 'forward' v 'on' and no packets shown in tcpdump. But we are getting tcpdump on gbpsfc1.

2) trace output as below

00:10:18:817738: dpdk-input
GigabitEthernet0/a/0 rx queue 0
buffer 0xd4ae: current data 0, length 106, free-list 0, totlen-nifb 0, trace 0x12
PKT MBUF: port 2, nb_segs 1, pkt_len 106
buf_len 2176, data_len 106, ol_flags 0x0,
packet_type 0x0
IP4: 52:54:00:39:34:9e -> 52:54:00:e6:42:b1
UDP: 192.168.50.70 -> 192.168.50.91
tos 0x00, ttl 64, length 92, checksum 0x2e4f
fragment id 0x2650, flags DONT_FRAGMENT
UDP: 4790 -> 4790
length 72, checksum 0x615c
00:10:18:817746: ethernet-input
IP4: 52:54:00:39:34:9e -> 52:54:00:e6:42:b1
00:10:18:817751: ip4-input
UDP: 192.168.50.70 -> 192.168.50.91
tos 0x00, ttl 64, length 92, checksum 0x2e4f
fragment id 0x2650, flags DONT_FRAGMENT
UDP: 4790 -> 4790
length 72, checksum 0x615c
00:10:18:817782: ip4-local
fib 0 adj-idx 4 : local 192.168.50.91/24 flow hash: 0x00000000
00:10:18:817788: ip4-udp-lookup
UDP: src-port 4790 dst-port 4790
00:10:18:817804: nsh-vxlan-gpe-input
NSH-VXLAN: tunnel 0 next 4 error 0
ver 0 len 6 (24 bytes) md_type 1 next_protocol 1
spi 1 si 255 c1 9901 c2 9902 c3 9903 c4 9904
00:10:18:817818: nsh-vxlan-gpe-encap
NSH-VXLAN-ENCAP: tunnel 3
00:10:18:817821: ip4-rewrite-transit
fib 0 adj-idx 6 : GigabitEthernet0/a/0
IP4: 52:54:00:e6:42:b1 -> 52:54:00:39:34:9e flow hash: 0x00000000
IP4: 52:54:00:e6:42:b1 -> 52:54:00:39:34:9e
UDP: 192.168.50.91 -> 192.168.50.70
tos 0x00, ttl 253, length 92, checksum 0xd79e
fragment id 0x0000
UDP: 4790 -> 4790
length 72, checksum 0x0000
00:10:18:817822: GigabitEthernet0/a/0-output
GigabitEthernet0/a/0
IP4: 52:54:00:e6:42:b1 -> 52:54:00:39:34:9e
UDP: 192.168.50.91 -> 192.168.50.70
tos 0x00, ttl 253, length 92, checksum 0xd79e
fragment id 0x0000
UDP: 4790 -> 4790
length 72, checksum 0x0000
00:10:18:817828: GigabitEthernet0/a/0-tx
GigabitEthernet0/a/0 tx queue 0
buffer 0xd4ae: current data 0, length 106, free-list 0, totlen-nifb 0, trace 0x12
IP4: 52:54:00:e6:42:b1 -> 52:54:00:39:34:9e
UDP: 192.168.50.91 -> 192.168.50.70
tos 0x00, ttl 253, length 92, checksum 0xd79e
fragment id 0x0000
UDP: 4790 -> 4790
length 72, checksum 0x0000

Can you please help us to resolve this issue.?

Thanks,

Suman

Show quoted text

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>
Cc: "pratik dave" <pratik.dave@...>, "Abhijeet karve" <abhijeet.karve@...>, "Keith Burns" <alagalah@...>, "nsh sfc-dev" <nsh_sfc-dev@...>
Sent: Friday, April 1, 2016 11:38:46 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4 represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

James Neil Guichard

#16

Please send output from:

show nsh vxlan tunnel

Dear Jim,

We tried implementing the NSH service chaining as mentioned in your document. But we are facing some challenge in doing the same

1) we are not getting any output for sudo python vxlan_tool.py i 'eth1' d 'forward' v 'on' and no packets shown in tcpdump. But we are getting tcpdump on gbpsfc1.

2) trace output as below

Can you please help us to resolve this issue.?

Thanks,

Suman

Show quoted text

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4 represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

Suman Kpatra/KOL/TCS <suman.kpatra@...>

#17

Dear Jim,

Thnaks for the hint we are correcting configurations for NSH Vxlan Tunnel. Will keep you posted on this.

Regarding segment routing my concern was that can we achieve service chaining with segment routing ipv6 for the scenario

service1--> vpp--> service2--> vpp--> service3

If yes, can you share some configuration document similar to this above usecase.

Thanks
Suman,

Show quoted text

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 10:29 AM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>,
"vpp-dev@..." <vpp-dev@...>, Keith Burns <alagalah@...>
Subject: Fwd: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4
represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Suman Kpatra/KOL/TCS" <986788@...>
To: "Jim Guichard (jguichar)" <jguichar@...>
Cc: "pratik dave" <pratik.dave@...>, "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@..., "Keith Burns" <alagalah@...>, "Abhijeet karve" <abhijeet.karve@...>
Sent: Friday, April 1, 2016 7:24:35 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@...
Cc: "pratik dave" <pratik.dave@...>
Sent: Friday, April 1, 2016 7:02:03 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

=====-----=====-----=====

Notice: The information contained in this e-mail

message and/or attachments to it may contain

confidential or privileged information. If you are

not the intended recipient, any dissemination, use,

review, distribution, printing or copying of the

information contained in this e-mail message

and/or attachments to it are strictly prohibited. If

you have received this communication in error,

please notify us by reply e-mail or telephone and

immediately and permanently delete the message

and any attachments. Thank you

--
Thanks & Regards

Suman Kalyan Patra

TCS-Gandhinagar,Garima Park

Tata Consultancy Services Limited

Cell Number: +918981065666

Office Nuber: +917961706230

Extension#6230

Mail to: suman.kpatra@...

Web site: www.tcs.com

James Neil Guichard

Segment routing is a completely different technology unrelated to NSH. Theoretically you could use segment routing as the outer transport for carrying NSH but as I said before we have not implemented that at this time in VPP.

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 3:37 PM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, nsh sfc-dev <nsh_sfc-dev@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thnaks for the hint we are correcting configurations for NSH Vxlan Tunnel. Will keep you posted on this.

Regarding segment routing my concern was that can we achieve service chaining with segment routing ipv6 for the scenario

service1--> vpp--> service2--> vpp--> service3

If yes, can you share some configuration document similar to this above usecase.

Thanks
Suman,

Show quoted text

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 10:29 AM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>,
"vpp-dev@..." <vpp-dev@...>, Keith Burns <alagalah@...>
Subject: Fwd: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4
represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Suman Kpatra/KOL/TCS" <986788@...>
To: "Jim Guichard (jguichar)" <jguichar@...>
Cc: "pratik dave" <pratik.dave@...>, "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@..., "Keith Burns" <alagalah@...>, "Abhijeet karve" <abhijeet.karve@...>
Sent: Friday, April 1, 2016 7:24:35 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@...
Cc: "pratik dave" <pratik.dave@...>
Sent: Friday, April 1, 2016 7:02:03 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

=====-----=====-----=====

Notice: The information contained in this e-mail

message and/or attachments to it may contain

confidential or privileged information. If you are

not the intended recipient, any dissemination, use,

review, distribution, printing or copying of the

information contained in this e-mail message

and/or attachments to it are strictly prohibited. If

you have received this communication in error,

please notify us by reply e-mail or telephone and

immediately and permanently delete the message

and any attachments. Thank you

--
Thanks & Regards

Suman Kalyan Patra

TCS-Gandhinagar,Garima Park

Tata Consultancy Services Limited

Cell Number: +918981065666

Office Nuber: +917961706230

Extension#6230

Mail to: suman.kpatra@...

Web site: www.tcs.com

Suman Kpatra/KOL/TCS <suman.kpatra@...>

#18

Hi Jim,

Got your point.

Now the obervation in existing NSH setup is like when we are running "python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904" in "gbpsfc1", Our NSH vxlan tunnels are getting vanished.

Sometimes we are getting response from VPP1 sometimes not and getting below error,

_________________________________________________ root@ubuntu:~/sfc/sfc-py/sfc# ./genNsh.sh
INFO:__main__:Sending VXLAN-GPE/NSH/IPv4 packet to SFF: ('192.168.50.91', 4790)
ERROR:__main__:No response from 192.168.50.91:4790

_________________________________________________

Thanks,
Abhijeet

Show quoted text

----- Original Message -----
From: Jim Guichard (jguichar) <jguichar@...>
To: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, nsh sfc-dev <nsh_sfc-dev@...>
Sent: Sat, 02 Apr 2016 01:09:01 +0530 (IST)
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 3:37 PM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>,
nsh sfc-dev <nsh_sfc-dev@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thnaks for the hint we are correcting configurations for NSH Vxlan Tunnel. Will keep you posted on this.

Regarding segment routing my concern was that can we achieve service chaining with segment routing ipv6 for the scenario

service1--> vpp--> service2--> vpp--> service3

If yes, can you share some configuration document similar to this above usecase.

Thanks

Suman,

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>
Cc: "pratik dave" <pratik.dave@...>, "Abhijeet karve" <abhijeet.karve@...>, "Keith Burns" <alagalah@...>,
"nsh sfc-dev" <nsh_sfc-dev@...>
Sent: Friday, April 1, 2016 11:38:46 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 10:29 AM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Suman Kpatra/KOL/TCS
<suman.kpatra@...>,

"vpp-dev@..." <vpp-dev@...>, Keith Burns <alagalah@...>
Subject: Fwd: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4

represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Suman Kpatra/KOL/TCS" <986788@...>
To: "Jim Guichard (jguichar)" <jguichar@...>
Cc: "pratik dave" <pratik.dave@...>, "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@..., "Keith Burns" <alagalah@...>, "Abhijeet karve" <abhijeet.karve@...>
Sent: Friday, April 1, 2016 7:24:35 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@...
Cc: "pratik dave" <pratik.dave@...>
Sent: Friday, April 1, 2016 7:02:03 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

=====-----=====-----=====

Notice: The information contained in this e-mail

message and/or attachments to it may contain

confidential or privileged information. If you are

not the intended recipient, any dissemination, use,

review, distribution, printing or copying of the

information contained in this e-mail message

and/or attachments to it are strictly prohibited. If

you have received this communication in error,

please notify us by reply e-mail or telephone and

immediately and permanently delete the message

and any attachments. Thank you

--
Thanks & Regards

Suman Kalyan Patra

TCS-Gandhinagar,Garima Park

Tata Consultancy Services Limited

Cell Number: +918981065666

Office Nuber: +917961706230

Extension#6230

Mail to: suman.kpatra@...

Web site: www.tcs.com

James Neil Guichard

#19

Please send output from show nsh vxlan tunnel as your tunnels are probably not setup correctly.

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 3:54 PM
To: Jim Guichard <jguichar@...>
Cc: Suman Kpatra/KOL/TCS <suman.kpatra@...>, pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, nsh sfc-dev <nsh_sfc-dev@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

_________________________________________________

Thanks,
Abhijeet

Show quoted text

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 3:37 PM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>,
nsh sfc-dev <nsh_sfc-dev@...>, Suman Kpatra/KOL/TCS <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thnaks for the hint we are correcting configurations for NSH Vxlan Tunnel. Will keep you posted on this.

Regarding segment routing my concern was that can we achieve service chaining with segment routing ipv6 for the scenario

service1--> vpp--> service2--> vpp--> service3

If yes, can you share some configuration document similar to this above usecase.

Thanks

Suman,

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>
Cc: "pratik dave" <pratik.dave@...>, "Abhijeet karve" <abhijeet.karve@...>, "Keith Burns" <alagalah@...>,
"nsh sfc-dev" <nsh_sfc-dev@...>
Sent: Friday, April 1, 2016 11:38:46 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

[Moving to nsh_sfc-dev alias …]

Hi Suman,

Please see inline ..

Jim

From: Suman Kpatra/KOL/TCS <suman.kpatra@...>
Date: Friday, April 1, 2016 at 10:29 AM
To: Jim Guichard <jguichar@...>
Cc: pratik dave <pratik.dave@...>, Abhijeet karve <abhijeet.karve@...>, Suman Kpatra/KOL/TCS
<suman.kpatra@...>,

"vpp-dev@..." <vpp-dev@...>, Keith Burns <alagalah@...>
Subject: Fwd: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progress.

If you can explain the following

#!/bin/bash

python3 sff_client.py remotesffip=192.168.50.91 remotesffport=4790 sfpid=1 sfpindex=255 encapsulate=gpenshipv4 innersrcip=10.1.1.1 innerdestip=10.1.1.2 ctx1=9901 ctx2=9902 ctx3=9903 ctx4=9904

1) what does this innersrcip=10.1.1.1 innerdestip=10.1.1.2 represent?

Jim> this is just constructing the correct packet format at the python agent..

2) what does these ctx1,ctx2,ctx3,ctx4

represent?

Jim> these represent the 4 context headers within a type-1 NSH packet.

3) how can we achieve this nsh service chaining with ipv6?

Jim> as of now you cannot.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Suman Kpatra/KOL/TCS" <986788@...>
To: "Jim Guichard (jguichar)" <jguichar@...>
Cc: "pratik dave" <pratik.dave@...>, "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@..., "Keith Burns" <alagalah@...>, "Abhijeet karve" <abhijeet.karve@...>
Sent: Friday, April 1, 2016 7:24:35 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Dear Jim,

Thanks for the document that you shared. Will definitely follow the same and try out with kvm vms , would keep you posted about the progess.

Another input required, can we achieve the same service chaining scenario as you mentioned in the document with segment routing ipv6 in VPP.

Thanks,

Suman

From: "Jim Guichard (jguichar)" <jguichar@...>
To: "Suman Kpatra/KOL/TCS" <suman.kpatra@...>,
vpp-dev@...
Cc: "pratik dave" <pratik.dave@...>
Sent: Friday, April 1, 2016 7:02:03 PM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Suman,

Ive attached a small document that shows a basic demo setup; you should be able to grab what you need from there.

Jim

Dear All,

Could you please share the some configuration details regarding service chaining for the following scenario via NSH vxlan tunnel/gre tunnel:

service1--> VPP-->service2-->VPP-->service3

or if you could explain some usecase with configuration using the below nsh command

vppctl nsh vxlan help

tunnel nsh vxlan tunnel src <ip4-addr> dst <ip4-addr> c1 <nn> c2 <nn> c3 <nn> c4 <nn> spi <nn> si <nn> vni <nn>

[encap-fib-id <nn>] [decap-fib-id <nn>] [o-bit <1|0>] [c-bit <1|0>]

[md-type <nn>][next-ip4][next-ip6][next-ethernet][next-nsh]

[tlv <xx>][decap-next [ip4|ip6|ethernet|nsh-encap]][del]

Thanks,

Suman

=====-----=====-----=====

Notice: The information contained in this e-mail

message and/or attachments to it may contain

confidential or privileged information. If you are

not the intended recipient, any dissemination, use,

review, distribution, printing or copying of the

information contained in this e-mail message

and/or attachments to it are strictly prohibited. If

you have received this communication in error,

please notify us by reply e-mail or telephone and

immediately and permanently delete the message

and any attachments. Thank you

--
Thanks & Regards

Suman Kalyan Patra

TCS-Gandhinagar,Garima Park

Tata Consultancy Services Limited

Cell Number: +918981065666

Office Nuber: +917961706230

Extension#6230

Mail to: suman.kpatra@...

Web site: www.tcs.com

Pratik

#20

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

nsh_vxlan_gpe_tunnel0 11 down nsh_vxlan_gpe_tunnel0
NSH_VXLAN_GPE
nsh_vxlan_gpe_tunnel1 12 down nsh_vxlan_gpe_tunnel1
NSH_VXLAN_GPE
nsh_vxlan_gpe_tunnel2 13 down nsh_vxlan_gpe_tunnel2
NSH_VXLAN_GPE
nsh_vxlan_gpe_tunnel3 14 down nsh_vxlan_gpe_tunnel3
NSH_VXLAN_GPE

root@vpp:~# vppctl sh nsh vxlan tunnel

[0] 192.168.50.70 (src) 192.168.50.91 (dst) fibs: encap 0, decap 0 decap next nsh-vxlan-gpe
vxlan VNI 16777215 nsh ver 0 len 6 (24 bytes) md_type 1 next_protocol 1
service path 1 service index 255
c1 9901 c2 9902 c3 9903 c4 9904
[1] 192.168.50.91 (src) 192.168.50.71 (dst) fibs: encap 0, decap 0 decap next nsh-vxlan-gpe
vxlan VNI 16777215 nsh ver 0 len 6 (24 bytes) md_type 1 next_protocol 1
service path 1 service index 255
c1 9901 c2 9902 c3 9903 c4 9904
[2] 192.168.50.71 (src) 192.168.50.91 (dst) fibs: encap 0, decap 0 decap next nsh-vxlan-gpe
vxlan VNI 16777215 nsh ver 0 len 6 (24 bytes) md_type 1 next_protocol 1
service path 1 service index 254
c1 9901 c2 9902 c3 9903 c4 9904
[3] 192.168.50.91 (src) 192.168.50.70 (dst) fibs: encap 0, decap 0 decap next nsh-vxlan-gpe
vxlan VNI 16777215 nsh ver 0 len 6 (24 bytes) md_type 1 next_protocol 1
service path 1 service index 254
c1 9901 c2 9902 c3 9903 c4 9904

Thanks,
Pratik
On behalf of suman

James Neil Guichard

Your gigabitethernet interface is not up and your tunnels are down.

Did you do:

cd ... /vpp/build-root/vagrant

export VPP_VAGRANT_NICS=1

edit Vagrantfile

change:

config.vm.network "private_network", type: "dhcp"

to:

config.vm.network "private_network", ip: "192.168.50.91”

Exit

And then:

vagrant up; vagrant ssh.
sudo su.
stop vpp.
ifconfig eth1 down.
ip addr flush dev eth1.
start vpp.

And then:

set int state GigabitEthernet0/8/0 up

set int ip address GigabitEthernet0/8/0 192.168.50.91/24

From: Pratik Dave <pratik.dave@...> on behalf of Pratik Dave <pratik.dave@...>
Date: Friday, April 1, 2016 at 4:00 PM
To: Jim Guichard <jguichar@...>, nsh sfc-dev <nsh_sfc-dev@...>
Cc: Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

James Neil Guichard

Hi Pratik,

Whats the output from vppctl show interface

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

Abhijeet karve

Hi Jim,

We are seting up the environment with KVM hypervisor. Please find complete set up commands we run to configure the at VPP instance setup:

_______________________________________________________________________________________________________________________
vppctl set int state GigabitEthernet0/a/0 up
vppctl set int ip address GigabitEthernet0/a/0 192.168.50.91/24

vppctl nsh vxlan tunnel src 192.168.50.70 dst 192.168.50.91 c1 9901 c2 9902 c3 9903 c4 9904 spi 1 si 255 vni 16777215 mdtype 1 nextip4 decapnext nshvxlangpe

vppctl nsh vxlan tunnel src 192.168.50.91 dst 192.168.50.71 c1 9901 c2 9902 c3 9903 c4 9904 spi 1 si 255 vni 16777215 mdtype 1 nextip4 decapnext nshvxlangpe

vppctl nsh vxlan tunnel src 192.168.50.71 dst 192.168.50.91 c1 9901 c2 9902 c3 9903 c4 9904 spi 1 si 254 vni 16777215 mdtype 1 nextip4 decapnext nshvxlangpe

vppctl nsh vxlan tunnel src 192.168.50.91 dst 192.168.50.70 c1 9901 c2 9902 c3 9903 c4 9904 spi 1 si 254 vni 16777215 mdtype 1 nextip4 decapnext nshvx
_______________________________________________________________________________________________________________________

Interface output on top of which tunnel is configured,
---------

GigabitEthernet0/a/0 7 up GigabitEthernet0/a/0
Ethernet address 52:54:00:e6:42:b1
Red Hat Virtio
carrier up full duplex speed 10000 mtu 9216
tx frames ok 4
tx bytes ok 232
rx frames ok 432
rx bytes ok 22964
extended stats:
rx good packets 432
tx good packets 4
rx good bytes 22964
tx good bytes 232
rx q0 packets 432
rx q0 bytes 22964
tx q0 packets 4
tx q0 bytes 232
rx q0 good packets 432
rx q0 good bytes 22964
rx q0 multicast packets 426
rx q0 undersize packets 419
rx q0 size 65 127 packets 12
rx q0 size 128 255 packets 1
tx q0 good packets 4
tx q0 good bytes 232
tx q0 multicast packets 1
tx q0 broadcast packets 1
tx q0 undersize packets 3
tx q0 size 65 127 packets 1

---------

Thanks,
Abhijeet Karve

From: "Jim Guichard (jguichar)" <jguichar@...>
To: Pratik Dave <pratik.dave@...>, nsh sfc-dev <nsh_sfc-dev@...>
Cc: Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "suman.kpatra@..." <suman.kpatra@...>
Date: 04/02/2016 01:37 AM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Your gigabitethernet interface is not up and your tunnels are down.

Did you do:

cd ... /vpp/build-root/vagrant
export VPP_VAGRANT_NICS=1

edit Vagrantfile
change:
config.vm.network "private_network", type: "dhcp"
to:
config.vm.network "private_network", ip: "192.168.50.91”

Exit

And then:

1. vagrant up; vagrant ssh.
2. sudo su.
3. stop vpp.
4. ifconfig eth1 down.
5. ip addr flush dev eth1.
6. start vpp.

And then:

set int state GigabitEthernet0/8/0 up

set int ip address GigabitEthernet0/8/0 192.168.50.91/24

From: Pratik Dave <pratik.dave@...> on behalf of Pratik Dave <pratik.dave@...>
Date: Friday, April 1, 2016 at 4:00 PM
To: Jim Guichard <jguichar@...>, nsh sfc-dev <nsh_sfc-dev@...>
Cc: Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

Abhijeet karve

Hi Jim,

The output is as below.

oot@vpp:~# vppctl sh int
Name Idx State Counter Count
GigabitEthernet0/8/0 5 down
GigabitEthernet0/9/0 6 down
GigabitEthernet0/a/0 7 up rx packets 69
rx bytes 3656
drops 69
ip6 2
GigabitEthernet0/b/0 8 down
GigabitEthernet0/c/0 9 down
GigabitEthernet0/d/0 10 down
local0 0 down
nsh_vxlan_gpe_tunnel0 11 up
nsh_vxlan_gpe_tunnel1 12 up
nsh_vxlan_gpe_tunnel2 13 up
nsh_vxlan_gpe_tunnel3 14 up
pg/stream-0 1 down
pg/stream-1 2 down
pg/stream-2 3 down
pg/stream-3 4 down

Thanks,
Abhijeet

From: "Jim Guichard (jguichar)" <jguichar@...>
To: Pratik Dave <pratik.dave@...>, nsh sfc-dev <nsh_sfc-dev@...>
Cc: Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "suman.kpatra@..." <suman.kpatra@...>
Date: 04/02/2016 01:43 AM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Pratik,

Whats the output from vppctl show interface

From: Pratik Dave <pratik.dave@...> on behalf of Pratik Dave <pratik.dave@...>
Date: Friday, April 1, 2016 at 4:00 PM
To: Jim Guichard <jguichar@...>, nsh sfc-dev <nsh_sfc-dev@...>
Cc: Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

James Neil Guichard

Hi Abhijeet,

Please clear the trace in VPP and then send full output from show trace so we can see what is going on here ..

Jim

From: Abhijeet Karve <abhijeet.karve@...>
Date: Friday, April 1, 2016 at 4:17 PM
To: Jim Guichard <jguichar@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

The output is as below.

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

James Neil Guichard

#10

Can you also provide output from show error

Hi Jim,

The output is as below.

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

Abhijeet karve

#11

Hi Jim,

Tried but it seems no output in sh trace as well as in sh error. When we issue "genNsh.sh" all our vpp interfaces are getting down immediately and tunnel configs are vanished.

Thanks,
Abhijeet

From: "Jim Guichard (jguichar)" <jguichar@...>
To: Abhijeet Karve <abhijeet.karve@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Date: 04/02/2016 01:53 AM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Abhijeet,

Please clear the trace in VPP and then send full output from show trace so we can see what is going on here ..

Jim

From: Abhijeet Karve <abhijeet.karve@...>
Date: Friday, April 1, 2016 at 4:17 PM
To: Jim Guichard <jguichar@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

The output is as below.

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

James Neil Guichard

#12

Hmmm .. Okay please send your genNSH.sh and Vagrantfile. Also to set the trace do trace add dpdk-input 10

From: Abhijeet Karve <abhijeet.karve@...>
Date: Friday, April 1, 2016 at 4:28 PM
To: Jim Guichard <jguichar@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

Abhijeet karve

#13

Hi Jim,

We are not using vagrant or virtual box environment for VPP executions. We are building a Ubuntu based guest VM on top of KVM hypervisor and installing VPP in it. My two clients(gbpsfc1 & gbpsfc2) are also on KVM hypervisor.

Also we tried setting up trace add dpdk-input 10 multiple times, but it seems no change in the output.

Thanks,
Abhijeet

From: "Jim Guichard (jguichar)" <jguichar@...>
To: Abhijeet Karve <abhijeet.karve@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Date: 04/02/2016 02:04 AM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hmmm .. Okay please send your genNSH.sh and Vagrantfile. Also to set the trace do trace add dpdk-input 10

From: Abhijeet Karve <abhijeet.karve@...>
Date: Friday, April 1, 2016 at 4:28 PM
To: Jim Guichard <jguichar@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Tried but it seems no output in sh trace as well as in sh error. When we issue "genNsh.sh" all our vpp interfaces are getting down immediately and tunnel configs are vanished.

Thanks,
Abhijeet

From: "Jim Guichard (jguichar)" <jguichar@...>
To: Abhijeet Karve <abhijeet.karve@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Date: 04/02/2016 01:53 AM
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Pratik,

Whats the output from vppctl show interface

From: Pratik Dave <pratik.dave@...> on behalf of Pratik Dave <pratik.dave@...>
Date: Friday, April 1, 2016 at 4:00 PM
To: Jim Guichard <jguichar@...>, nsh sfc-dev <nsh_sfc-dev@...>
Cc: Abhijeet karve <abhijeet.karve@...>, Keith Burns <alagalah@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

James Neil Guichard

#21

To be honest I think you need to remove the NSH stuff and start with getting a basic topology working where you can verify that you get the correct interfaces up with necessary IP connectivity through VPP between the VMs (ping etc) .. Once that is working fine we can add the NSH stuff.

Jim

From: Abhijeet Karve <abhijeet.karve@...>
Date: Friday, April 1, 2016 at 4:39 PM
To: Jim Guichard <jguichar@...>
Cc: Keith Burns <alagalah@...>, "nsh_sfc-dev@..." <nsh_sfc-dev@...>, Pratik Dave <pratik.dave@...>, "suman.kpatra@..." <suman.kpatra@...>
Subject: Re: [vpp-dev] VPP Service chaining with NSH

Hi Jim,

Requested output as below:

vppctl sh hardware showing this:

1 - 20 of 27

Previous Topic Next Topic

Home Hashtags Subgroups Terms