[vpp-dev] Delete IPv6 VXLAN fails

[Michal Cmarada via Lists.Fd.Io]

Hi,

I was trying to setup and delete IPV6 vxlan tunel using vpp_api_test:
vat# vxlan_tunnel_dump
sw_if_index   instance src_address dst_address  
encap_vrf_id  decap_next_index  vni  mcast_sw_if_index
vat# vxlan_add_del_tunnel src 10::10 dst 10::11 vni 88 encap-vrf-id 0 
decap-next l2
vat# vxlan_tunnel_dump
sw_if_index   instance src_address dst_address  
encap_vrf_id  decap_next_index  vni  mcast_sw_if_index
  2  0  10::11  10::10  
   0 1   88 -1
vat# vxlan_add_del_tunnel src 10::10 dst 10::11 vni 88 encap-vrf-id 0 
decap-next l2 del
vxlan_add_del_tunnel error: Misc
vat# vxlan_tunnel_dump
sw_if_index   instance src_address dst_address  
encap_vrf_id  decap_next_index  vni  mcast_sw_if_index
vxlan_tunnel_dump error: Misc

The reason is that we have simmilar test in CSIT for honeycomb and it is 
failing, so I wanted to test it manually using vat. I tried to add tunnel which 
was successful, but the delete afterwards failed.
I tested this using 19.01-rc0~8-g642829d~b5417 build, but the tests are failing 
for few days so it is probably present in 18.10 too.
I have HC dumps from vppAPI call too which behaves the same. Add is succesfull 
and Delete fails.
ADD:
VxlanAddDelTunnel{isAdd=1, isIpv6=1, instance=0, srcAddress=[0, 16, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 16], dstAddress=[0, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 17], mcastSwIfIndex=0, encapVrfId=0, decapNextIndex=1, vni=88}
DELETE:
VxlanAddDelTunnel{isAdd=0, isIpv6=1, instance=0, srcAddress=[0, 16, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 16], dstAddress=[0, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 17], mcastSwIfIndex=0, encapVrfId=0, decapNextIndex=1, vni=88}

Seems there might be a bug in VPP. Can somebody check it?

Thanks

Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#10796): https://lists.fd.io/g/vpp-dev/message/10796
Mute This Topic: https://lists.fd.io/mt/27241657/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] bug in api for tapv2

[Michal Cmarada via Lists.Fd.Io]

Hi VPP devs,

I found a bug in API for tapv2. I prepared a fix but I need someone to check it.
gerrit patch: https://gerrit.fd.io/r/#/c/15341/

also it will probably need a cherry-pick to stable 18.10

Thanks

Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#10854): https://lists.fd.io/g/vpp-dev/message/10854
Mute This Topic: https://lists.fd.io/mt/27379050/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] bug in api for tapv2

[Michal Cmarada via Lists.Fd.Io]

Hi Marco,

Done: https://gerrit.fd.io/r/#/c/15342/

Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com<mailto:mcmar...@cisco.com>
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here<http://www.cisco.com/c/en/us/about/legal/terms-sale-software-license-agreement/company-registration-information.html>
 for Company Registration Information.







From: Marco Varlese 
Sent: Wednesday, October 17, 2018 2:16 PM
To: Michal Cmarada -X (mcmarada - PANTHEON TECHNOLOGIES at Cisco) 
; vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] bug in api for tapv2

Hi Michal,

On Wed, 2018-10-17 at 11:49 +0000, Michal Cmarada via Lists.Fd.Io wrote:
Hi VPP devs,

I found a bug in API for tapv2. I prepared a fix but I need someone to check it.
gerrit patch: https://gerrit.fd.io/r/#/c/15341/
Yes, your fix makes sense to me.
I am waiting for the Verified+1 to merge it.


also it will probably need a cherry-pick to stable 18.10
Yes, it will have to be cherry-picked to stable 18.10. Can you push a patch 
there too?



Thanks

Michal
Cheers,
Marco



[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com<mailto:mcmar...@cisco.com>
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here<http://www.cisco.com/c/en/us/about/legal/terms-sale-software-license-agreement/company-registration-information.html>
 for Company Registration Information.








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

Links: You receive all messages sent to this group.



View/Reply Online (#10854): https://lists.fd.io/g/vpp-dev/message/10854

Mute This Topic: https://lists.fd.io/mt/27379050/675056

Group Owner: vpp-dev+ow...@lists.fd.io<mailto:vpp-dev+ow...@lists.fd.io>

Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  
[mvarl...@suse.de<mailto:mvarl...@suse.de>]

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



--
Marco V

SUSE LINUX GmbH | GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg) Maxfeldstr. 5, D-90409, Nürnberg
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#10856): https://lists.fd.io/g/vpp-dev/message/10856
Mute This Topic: https://lists.fd.io/mt/27379050/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] 18.10 release artifacts

[Michal Cmarada via Lists.Fd.Io]

Hi vpp-devs,

I was preparing honeycomb release and while building it I found out that there 
are some artifacts missing in nexus repository for 18.10. 
https://nexus.fd.io/content/repositories/fd.io.release/io/fd/vpp/ . most of 
them were deployed ok, but some of them are missing e.g. jvpp-ioampot. I tried 
to build 18.10 release of vpp locally on Ubuntu and it seems that jvpp is built 
correctly. I then exported jvpp dependencies directly out of debian package and 
it contained jvpp-ioampot. I think that something might happen during upload to 
nexus.

Can someone please check it out? I am not familiar with the build process and 
upload to nexus.

Thanks,

Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#11042): https://lists.fd.io/g/vpp-dev/message/11042
Mute This Topic: https://lists.fd.io/mt/27799255/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Interface dump - link speed value

[Michal Cmarada via Lists.Fd.Io]

Hi VPP-devs,

I was recently testing interface dump and I noticed that it returns weird value 
for link_speed. I tested both Java API and Python API for VPP together with 
console output.
When I called sudo vppctl sh hardware in console I get:
  NameIdx   Link  Hardware
GigabitEthernet0/9/0   1down  GigabitEthernet0/9/0
  Link speed: 1 Gbps
  Ethernet address 08:00:27:04:6e:c0
  ...

Which says that link speed is 1Gbps. When I try this using Java API or Python 
API I get a value of 1078071040.
Part of python script which returns the result:
...
for intf in vpp.api.sw_interface_dump():
print(intf.interface_name.decode() + ", link speed:" + str(intf.link_speed))
...
Output:
local0, link speed:0
GigabitEthernet0/9/0, link speed:1078071040

The provided value should be in kbps according to documentation for API:
define sw_interface_details
...
/* link speed in kbps */
  u32 link_speed;


Is this value in bps or kbps? Should it be divided (1Gbps should be 1 000 000 
000bps) and rounded? Or is this a bug?

Thanks Michal




[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#11330): https://lists.fd.io/g/vpp-dev/message/11330
Mute This Topic: https://lists.fd.io/mt/28266097/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Interface dump - link speed value

[Michal Cmarada via Lists.Fd.Io]

Hi,

Yes it helped now it returns the value in kbps.

Thanks


Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel: 

Cisco Systems, Inc.



Slovakia
cisco.com

Think before you print.
This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
http://www.cisco.com/c/en/us/about/legal/terms-sale-software-license-agreement/company-registration-information.html
-Original Message-
From: vpp-dev@lists.fd.io  On Behalf Of Ole Troan
Sent: Tuesday, November 20, 2018 11:05 AM
To: Michal Cmarada -X (mcmarada - PANTHEON TECHNOLOGIES at Cisco) 

Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] Interface dump - link speed value

Michal,

At least one bug.
Can you verify if
https://gerrit.fd.io/r/16046

Fixes your problem?

Best regards,
Ole

> On 20 Nov 2018, at 10:58, Michal Cmarada via Lists.Fd.Io 
>  wrote:
> 
> Hi VPP-devs,
>  
> I was recently testing interface dump and I noticed that it returns weird 
> value for link_speed. I tested both Java API and Python API for VPP together 
> with console output.
> When I called sudo vppctl sh hardware in console I get:
>   NameIdx   Link  Hardware
> GigabitEthernet0/9/0   1down  GigabitEthernet0/9/0
>   Link speed: 1 Gbps
>   Ethernet address 08:00:27:04:6e:c0
>   …
>  
> Which says that link speed is 1Gbps. When I try this using Java API or Python 
> API I get a value of 1078071040.
> Part of python script which returns the result:
> …
> for intf in vpp.api.sw_interface_dump():
> print(intf.interface_name.decode() + ", link speed:" + 
> str(intf.link_speed)) …
> Output:
> local0, link speed:0
> GigabitEthernet0/9/0, link speed:1078071040
>  
> The provided value should be in kbps according to documentation for API:
> define sw_interface_details
> …
> /* link speed in kbps */
>   u32 link_speed;
>  
> 
> Is this value in bps or kbps? Should it be divided (1Gbps should be 1 000 000 
> 000bps) and rounded? Or is this a bug?
>  
> Thanks Michal
>  
>  
>  
>  
> 
> Michal Cmarada
> Engineer - Software
> mcmar...@cisco.com
> Tel:
>  
>  
>  
>  
> Cisco Systems, Inc.
>  
>  
>  
> Slovakia
> cisco.com
> 
> Think before you print.
> This email may contain confidential and privileged material for the sole use 
> of the intended recipient. Any review, use, distribution or disclosure by 
> others is strictly prohibited. If you are not the intended recipient (or 
> authorized to receive for the recipient), please contact the sender by reply 
> email and delete all copies of this message.
> Please click here for Company Registration Information.
>  
>  
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> 
> View/Reply Online (#11330): 
> https://lists.fd.io/g/vpp-dev/message/11330
> Mute This Topic: https://lists.fd.io/mt/28266097/675193
> Group Owner: vpp-dev+ow...@lists.fd.io
> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  
> [otr...@employees.org]
> -=-=-=-=-=-=-=-=-=-=-=-

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

View/Reply Online (#11333): https://lists.fd.io/g/vpp-dev/message/11333
Mute This Topic: https://lists.fd.io/mt/28266097/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] DHCP relay issues

[Michal Cmarada via Lists.Fd.Io]

Hi vpp devs,

After this patch in VPP: https://gerrit.fd.io/r/#/c/15759/
our DHCP relay tests started to fail. The configuration seems to be written 
correctly (same as before, commands for VPP on DUT1 in attachment), but when 
actual packet is received the vpp seems to crash. Sometimes there are 
postmortem api files generated but I haven't  seen a core dump file. VPP is 
restarted for sure. It has different PID afterwards and configuration is lost. 
What I managed to figure out is that it happens after DHCP option 82 packet is 
sent from TG to DUT1 (attached).
Topology:
#+--+ +-+ port1
#|  | port1 port3 | +--+
#| DUT1 +-+ TG  |  |
#|  +-+ +--+
#|  | port2 port4 | | port2
#+-+-+--+ +-+-+-+

The packet is sent from port3 on TG, then it passes to DUT1 where vpp is 
running and has both ports enabled and IP assigned. The other port on TG is 
running a DHCP server. VPP is set as a DHCP relay to pass the requests sent 
from port3 on TG to port4 on TG where DHCP server is running. But as I said vpp 
crashes when the packet is received. I tried to run it with "dut1 vpp 
commands.txt" configuration on clean vpp. After the same packet is sent VPP 
crashes. So this is the only configuration needed to replicate this behavior on 
such setup.
I used the script provided by CSIT to generate the packet 
https://gerrit.fd.io/r/gitweb?p=csit.git;a=blob;f=resources/traffic_scripts/dhcp/send_and_check_proxy_discover.py;h=d8089713fd10643f92d178bbe024cf4b04ed7e94;hb=refs/heads/master
The command looks like this:
sudo -S sh -c "cd ~/openvpp-testing; virtualenv --system-site-packages 
--never-download env && export PYTHONPATH=\${PWD}; . \${PWD}/env/bin/activate; 
resources/traffic_scripts/dhcp/send_and_check_proxy_messages.py --tx_if enp0s17 
--rx_if enp0s16 --server_ip 192.168.0.100 --server_mac 08:00:27:20:b0:38 
--client_ip 172.16.0.2 --client_mac 08:00:27:62:73:6e --proxy_ip 172.16.0.1"

Can someone take a look?

Thanks
Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







Sending packet out of ens7 of len 286
###[ Ethernet ]###
  dst   = ff:ff:ff:ff:ff:ff
  src   = fa:16:3e:2b:11:9e
  type  = 0x800
###[ IP ]###
 version   = 4L
 ihl   = 5L
 tos   = 0x0
 len   = 272
 id= 1
 flags = 
 frag  = 0L
 ttl   = 64
 proto = udp
 chksum= 0x79dd
 src   = 0.0.0.0
 dst   = 255.255.255.255
 \options   \
###[ UDP ]###
sport = bootpc
dport = bootps
len   = 252
chksum= 0x889
###[ BOOTP ]###
   op= BOOTREQUEST
   htype = 1
   hlen  = 6
   hops  = 0
   xid   = 0
   secs  = 0
   flags = 
   ciaddr= 0.0.0.0
   yiaddr= 0.0.0.0
   siaddr= 0.0.0.0
   giaddr= 0.0.0.0
   chaddr= 
'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00'
   sname = 
'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00'
   file  = 
'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00'
   options   = 'c\x82Sc'
###[ DHCP options ]###
  options   = [message-type=discover end]
dut1:~$ sudo vppctl set int ip address GigabitEthernet0/9/0 192.168.0.1/24
dut1:~$ sudo vppctl set int state GigabitEthernet0/9

[vpp-dev] RPM for vpp and vpp plugins

[Michal Cmarada via Lists.Fd.Io]

Hi,

I was trying to generate Java APIs based on VPP DEB and RPM packages. With 
Debian everything works. With RPM packages I have some difficulties. In Debian 
packages I am able to distinguish between Core API files and plugin API files 
because VPP debian package contains only core api.json and vpp-plugins  DEB 
package contains plugin api.json files. However this is not the case for RPM 
where both vpp and vpp-plugin packages contain both core and plugin api.json 
files. I think that RPMs should follow the same file scheme as DEB packages. 
Now I need to have an internal list of api files for plugins and I need to 
remove them from core files. This is needed because Java Api first generates 
core APIs and afterwards plugin APIs. This list can now only be maintained 
manually. Other option is to use VPP source which seems like overhead to me 
because the same can be done with installed vpp vpp-dev and vpp-plugin packages.

Can anyone take a look and verify if this can be fixed?

Either way I think it is not necessary for both RPM packages to contain 
duplicate files

You can verify this on packages uploaded to package cloud (I checked master and 
18.10 stable but probably this is the same for all versions):
https://packagecloud.io/fdio/1810/packages/el/7/vpp-18.10-16~ge351f35~b63.x86_64.rpm?
https://packagecloud.io/fdio/1810/packages/el/7/vpp-plugins-18.10-16~ge351f35~b63.x86_64.rpm?

https://packagecloud.io/fdio/1810/packages/ubuntu/xenial/vpp_18.10-16~ge351f35~b72_amd64.deb
https://packagecloud.io/fdio/1810/packages/ubuntu/xenial/vpp-plugins_18.10-16~ge351f35~b72_amd64.deb

thanks
Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#11552): https://lists.fd.io/g/vpp-dev/message/11552
Mute This Topic: https://lists.fd.io/mt/28717819/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] IpSec API call ikev2_set_profile_auth causes VPP to crash

[Michal Cmarada via Lists.Fd.Io]

Hi,

I was testing IpSec API via honeycomb and when we called ikev2_set_profile_auth 
with "data" and "data_length" set, vpp crashed. When investigating we found out 
that "data_length" is not processed correctly. I prepared a patch that fixes 
this error. It was verified using HC via api call.
Master: https://gerrit.fd.io/r/#/c/16869/
Stable 19.01: https://gerrit.fd.io/r/#/c/16870/

Can someone take a look at this fix. Thanks

Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#11950): https://lists.fd.io/g/vpp-dev/message/11950
Mute This Topic: https://lists.fd.io/mt/29198001/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Classify session weird behavior

[Michal Cmarada via Lists.Fd.Io]

Hi VPP devs, Raj,

We had an issue when configuring classify session  for classify table, where 
the n_skip_vector value is more than zero. The same issue happens when I am 
creating the configuration over API or via vppctl. The match "k: 
" value is always all zeros in both API and 
vppctl responses. I believe it should be "k: c0a80102" 
that is the match ip of 192.168.1.2.

Can someone check whether there is a bug and the value is not configured at all 
or only the data shown to the user are wrongly interpreted?

I added Raj to copy. He noticed this originally and will try to test the 
configuration on his setup.

Raj: If you have more data available after testing, please provide them here. 
Thanks

API CALLS EXAMPLE:
ClassifyAddDelTable{isAdd=1, delChain=0, tableIndex=-1, nbuckets=2, 
memorySize=1048576, skipNVectors=12, matchNVectors=1, nextTableIndex=-1, 
missNextIndex=-1, currentDataFlag=0, currentDataOffset=0, maskLen=16, mask=[0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, 0, 0]}
ClassifyAddDelSession{isAdd=1, tableIndex=0, hitNextIndex=0, opaqueIndex=0, 
advance=0, action=0, metadata=0, matchLen=208, match=[0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, -64, -88, 1, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]}
ClassifySessionDetailsReplyDump{classifySessionDetails=[ClassifySessionDetails{context=99,
 tableId=0, hitNextIndex=0, advance=0, opaqueIndex=1, matchLength=16, match=[0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]}]}

sudo vppctl show classify tables verbose
  TableIdx  Sessions   NextTbl  NextNode
 0 1-1-1
  Heap: total: 1.06M, used: 1.27K, free: 1.06M, trimmable: 1.06M
no traced allocations

  nbuckets 2, skip 12 match 1 flag 0 offset 0
  mask 
  linear-search buckets 0

[0]: heap offset 1136, elts 2, normal
0: [1136]: next_index 0 advance 0 opaque 1 action 0 metadata 0
k: 
hits 0, last_heard 0.00

1 active elements
1 free lists
0 linear-search buckets

VPPCTL EXAMPLE:
sudo vppctl classify table mask l3 ip4 src skip 12
sudo vppctl classify session policer-hit-next policy1 exceed-color 
table-index 0 match l3 ip4 src 192.168.1.2

sudo vppctl show classify tables verbose
  TableIdx  Sessions   NextTbl  NextNode
 0 1-1-1
  Heap: total: 2.06M, used: 1.27K, free: 2.06M, trimmable: 2.06M
no traced allocations

  nbuckets 2, skip 12 match 1 flag 0 offset 0
  mask 
  linear-search buckets 0

[0]: heap offset 1136, elts 2, normal
0: [1136]: next_index 0 advance 0 opaque 1 action 0 metadata 0
k: 
hits 0, last_heard 0.00

1 active elements
1 free lists
0 linear-search buckets


Thanks
Michal

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here
 for Company Registration Information.







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

View/Reply Online (#12657): https://lists.fd.io/g/vpp-dev/message/12657
Mute This Topic: https://lists.fd.io/mt/30810146/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Classify session weird behavior

[Michal Cmarada via Lists.Fd.Io]

Hi,

Not sure if I understand correctly but I have tried to send the Match vector 
zero prepended (n_skip*16 = 12*16 = 192 zeros and then the actual data) and in 
this case the value is displayed using vppctl command, but I am not sure if it 
works correctly. We have some automated tests which failed after this change.
API call
ClassifyAddDelSession{isAdd=1, tableIndex=0, hitNextIndex=0, opaqueIndex=0, 
advance=0, action=0, metadata=0, matchLen=208, match=[0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -64, -88, 1, 2, 0, 0]}

sudo vppctl show classify tables verbose
  TableIdx  Sessions   NextTbl  NextNode
 0 1-1-1
  Heap: total: 1.06M, used: 1.27K, free: 1.06M, trimmable: 1.06M
no traced allocations

  nbuckets 2, skip 12 match 1 flag 0 offset 0
  mask 
  linear-search buckets 0

[1]: heap offset 1136, elts 2, normal
0: [1136]: next_index 0 advance 0 opaque 1 action 0 metadata 0
k: c0a80102
hits 0, last_heard 0.00

But how would I do this using vppctl because when I used
“sudo vppctl classify session policer-hit-next policy1 exceed-color table-index 
0 match l3 ip4 src 192.168.1.2”
The response was also “k: ”

I tried to understand how it should be done using guide provided at 
https://wiki.fd.io/view/VPP/Introduction_To_N-tuple_Classifiers, where it is 
said that:
“Match is the key for the indicated session. It need not be aligned, but it 
must be table->match_n_vectors*16 octets in length.”
So  I thought that match is not aligned but it needs to be of proper length. 
That is the original implementation we had in honeycomb (data are at beginning 
of match vector and the rest is filled with zeros and matches the desired 
length)

So how is it, does match need to be prepended in classify session?
And how should I replicate this using vppctl command because the command I used 
behaves the same way (it shows all 0 for k: value)


M.

[https://www.cisco.com/c/dam/m/en_us/signaturetool/images/banners/standard/06_standard_graphic.png]




Michal Cmarada
Engineer - Software
mcmar...@cisco.com<mailto:mcmar...@cisco.com>
Tel:










Cisco Systems, Inc.



Slovakia
cisco.com



[http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gif]

Think before you print.

This email may contain confidential and privileged material for the sole use of 
the intended recipient. Any review, use, distribution or disclosure by others 
is strictly prohibited. If you are not the intended recipient (or authorized to 
receive for the recipient), please contact the sender by reply email and delete 
all copies of this message.
Please click 
here<http://www.cisco.com/c/en/us/about/legal/terms-sale-software-license-agreement/company-registration-information.html>
 for Company Registration Information.







From: Andrew 👽 Yourtchenko 
Sent: Thursday, March 28, 2019 12:34 PM
To: Michal Cmarada -X (mcmarada - PANTHEON TECHNOLOGIES at Cisco) 

Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] Classify session weird behavior

Where do you point into the session data vector argument when creating it ?

IIRC you need to point to the point *before* the n-skip vectors - i.e. the 
argument needs to include the “skipped” data as well.

That is if I you have skip 3 words and match 2, the session data vector to be 
passed as a parameter needs to be 5 words.

Back in the time when I noticed it I rationalized it that you might want to 
create sessions from the same packet data so recalculating the offer each time 
can be a pain.


--a

On 28 Mar 2019, at 12:08, Michal Cmarada via Lists.Fd.Io 
mailto:mcmarada=cisco@lists.fd.io>> wrote:
Hi VPP devs, Raj,

We had an issue when configuring classify session  for classify table, where 
the n_skip_vector value is more than zero. The same issue happens when I am 
creating the configuration over API or via vppctl. The match “k: 
” value is always all zeros in both API and 
vppctl responses. I believe it should be “k: c0a80102” 
that is the match ip of 192.168.1.2.

Can someone check whether there is a bug and the value is not configured at all 
or only the data shown to the user are wrongly interpreted?

I added Raj to copy. He noticed this originally and will try to test the 
configuration on his setup.

Raj: If you have more