Looks good to me. Merged.
Best regards,
Ole
> On 24 Jan 2023, at 08:55, Stanislav Zaikin wrote:
>
> Hello folks,
>
> Any help with review is much appreciated. Both patches are waiting for quite
> a long time.
>
> - https://gerrit.fd.io/r/c/vpp/+/36721
> Short description: "autoendian" was
Thanks Pim! I also wondered the other day where the env variable had disappeared to. 37792 merged. O. On 12 Dec 2022, at 00:44, Pim van Pelt via lists.fd.io wrote:Of course, the link is https://gerrit.fd.io/r/c/vpp/+/37791groet,PimOn Mon, Dec 12, 2022 at 12:41 AM Pim van Pelt via lists.fd.io
Hi Amir,
Have you tried configuring the NAT with “output-feature” enabled?
Best regards,
Ole
> On 21 Sep 2022, at 06:43, Amir Hossein <1377amirhossein1...@gmail.com> wrote:
>
>
> Hi,
> I'm working with nat44-ed plugin and I found that we can set a distinct nat
> ip pool for each in if
Vijay,
VPP is a data plane.
There are various ways of integrating with routing protocols. The simplest is
likely to use the linux-cp plugin and use linux as the control plane.
Best regards,
Ole
> On 12 Sep 2022, at 08:38, Vijay Kumar wrote:
>
>
> Hi,
>
> Based on exploring the
>From rfc4861:
Each prefix has an associated:
AdvValidLifetime
The value to be placed in the Valid
Lifetime in the Prefix Information option,
in seconds. The
Guys,
Apologies for the noise on the list, but just wanted to let you know that I'm
out hiking from now until end of October.
Doing the PCT. At least that's the plan.
I expect the VPP project will happily continue without me, but please expect a
little delay in reviews and merges (i.e. about 5
I’d be happy to have formatting consistency across the project.
Cheers
Ole
> On 2 May 2022, at 19:17, Klement Sekera wrote:
>
> +Real Ole.
>
>>> On 2 May 2022, at 19:15, Klement Sekera via lists.fd.io
>>> wrote:
>>>
>>> Hi Dave,
>>>
>>> good catch about requirements.txt. Should be
Dear Mr Longtrb,
> I am a newbie in VPP. Many thanks to you all for making great Vpp.
> I have many applications in user space (written by java, C, python,...) and I
> want to connect those apps to VPP so what is the fastest way that VPP can
> support without changing the application?
>
> I
>
> If the FIN was received, then this should be okey I think. Both sides are
> aware that the session is closing.
>
> If the FIN and all of its re-transmits are lost, then this can become a
> problem indeed. When the other side sends data after an hour, the packets
> will be dropped and it
> On 18 Mar 2022, at 17:29, Miklós Tirpák wrote:
>
>
> Getting back to some of the topics,
>
>>
>> The main concern about RST was to recover from a 3rd party sending RSTs into
>> the session.
> This would require the sender to spoof the src IP, right?
> Are you aware of any scenario when
> On 18 Mar 2022, at 17:40, Klement Sekera wrote:
>
> I like the idea of VPP sending RST for cases where the session doesn’t exist
> or for some reason the state is invalid.
>
> It might also be a good idea to implement VPP sending TCP keepalives for idle
> sessions to discover whether the
Klement,
>> Following up on this thread.
>> The changes in 34877 led to some undesired behaviour in the "real
>> world(tm)".
>> In the close pattern below it left sessions in established state, and
>> with a relatively low cps
>> would consume the whole session table.
Klement,
Following up on this thread.
The changes in 34877 led to some undesired behaviour in the "real
world(tm)".
In the close pattern below it left sessions in established state, and with
a relatively low cps
would consume the whole session table.
Thanks for quick replies Klement!
>> Following up on this thread.
>> The changes in 34877 led to some undesired behaviour in the "real world(tm)".
>> In the close pattern below it left sessions in established state, and with a
>> relatively low cps
>> would consume the whole session table.
>>
Hi,
Following up on this thread.
The changes in 34877 led to some undesired behaviour in the "real world(tm)".
In the close pattern below it left sessions in established state, and with a
relatively low cps
would consume the whole session table.
The change here
Hi, and welcome!
> ip route add /32 via
The “via” should be the address of the next-hop, not the locally assigned
interface address.
If the default router has an address within that /29 you don’t need this route
at all, just the default route via the default router address.
Hope that
Hi,
> I am not getting a reply from vpp.
> Any pointers will be helpful.
> Thanks in advance.
>
> logs below. after " writing punt socket register...:" log it does not
> come out.
>
> NTPSYNC-TEST:: app start
> NTPSYNC-TEST:: VPP Server Socket bind Successful (/var/run/vpp-punt.sock)
>
> I can't find the meaning of the following commands,even if on the office
> website:
>
> set interface tag toSw1 toSw1
commit 7be864ad0d8e3b139a277fb4a0234480f0cc3daa
Author: Dave Barach
Date: Mon Nov 28 11:41:35 2016 -0500
Add a 64-byte interface "tag" for vhost and tap interfaces
> I am using VPP 21.06 ,an lcp that supports Ipv4 traffic properly .
What do you mean by “an lcp…”?
> I am facing an issue with Ipv6 traffic as it is not reaching the tap
> interface and is handled by the VPP node. It was mentioned in docs to disable
> ND, how can I disabled ND ?
You need to
Wei,
If you use the high level C API called vapi instead of the low level C
interface you may be in for an easier ride.
Cheers
Ole
> On 7 Mar 2022, at 18:14, Florin Coras wrote:
>
> Hi Wei,
>
> I suspect you wanted to register CREATE_LOOPBACK_REPLY instead of
> CREATE_LOOPBACK in
Artyom,
> Could someone please explain to me why VxLAN L3-mode was inverted here?
> https://gerrit.fd.io/r/c/vpp/+/34721
>
> The 'isL3' argument in api or 'l3' in CLI means "use this in L3 mode",
> doesn't it?
>
> This looks confusing, because we set L3 but the tunnel has MAC-address:
>
Rajaram,
> Classification: Confidential
If so, perhaps you shouldn't send to a public archived mailing list? ;-)
> Hi VPP team,
>
>While compiling using “make pkg-rpm”, I am getting following in the file
> mentioned above.
>
> sr/lib/python3.6 using /usr/bin/python3.6
> Compiling
>
Rajith,
Have you tried:
make test-debug TEST=ip4
Cheers,
Ole
> On 4 Feb 2022, at 13:18, Rajith PR via lists.fd.io
> wrote:
>
> Hi All,
>
> We are trying to understand the VPP test framework. To get started we ran an
> example suite (ip4 test) but it seems that the dependent
Hi Subramanya,
In general a router implements the weak host model, see
https://datatracker.ietf.org/doc/html/rfc1122#section-3.3.4
In the weak host model the source address selected is independent of the
outgoing interface.
> I'm trying to ping, and I observe that there is a mismatch between
Guys, can we do a little effort here. Those of you in the To: field, you know
what to do.
https://github.com/vpp-dev/vpp-coverity-report
Best regards,
Ole
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#20744):
> Him Pim,
This was supposed to read "Hi Pim". Although I did to find a nice play on
"Hoi". ;-)
Ole
>
> I'm also doing a round now updating the Geneve implementation:
> - add direct {IPv4, IPv6} encapsulation support
> - validate protocol type (now all payload is blindly treated as Ethernet)
Him Pim,
I'm also doing a round now updating the Geneve implementation:
- add direct {IPv4, IPv6} encapsulation support
- validate protocol type (now all payload is blindly treated as Ethernet)
- drop packets with C bit set
- make well known port configurable
cheers,
Ole
> On 14 Jan 2022,
Florin,
Actually no. Geneve isn’t like a typical UDP application where receiver
reverses the ports.
The IANA allocated port is used as the destination port in both directions.
Making the source port purely used for entropy.
Of course for v6 the whole UDP header would largely be superfluous.
Don't forget December's highlight; the VPP community meeting.
This afternoon: 1600-1700UTC
https://wiki.fd.io/view/VPP/Meeting#Meeting_Details
Best regards,
Ole
signature.asc
Description: Message signed with OpenPGP
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
Akash,
> I am trying to see API flow from packet rx to packet tx, but I am not able to
> see. Please help on this.
I am not sure I can parse your question.
There is tracing of the binary VPP API calls between data plane and control
plane, but that's independent of the packet flow.
There is
Hi Matt,
> On 27 Nov 2021, at 18:00, Matt Kloppmann wrote:
>
> I was looking at the L2TP plugin and was curious if there was any
> particular reason IPv6 was implemented before IPv4.
Just that the first use case for l2tpv3 was over an ipv6 only network.
The API is IP version agnostic and
Akash,
> Identification Field is not set by VPP and there is no parameter for the same
> handled in ip4_packet.h as well. Should that be added as per requirements or
> is that present elsewhere? Also is there any handle on VPP to set those
> values during gtpu encapsulation ?
VPP sets it when
Akash,
> Ultimate aim is to classify drop counters and its reason. I need a VAPI
> handler to fetch all these details but I couldn't find any such VAPI to get
> these.
> Yes "show hard" will get me the interface stats but I need a respective VAPI
> to handle using APK and never use vppctl.
Hi Ben,
Good to see you are looking at MAP. Always interested in deployment details if
you want to share. Feel free to ping me offline.
The code is written this way, considering the IPv6 tunnel endpoint as a
functional address. I.e. not being used to address an interface, but to address
the
Ameen,
> As part of my Ph.D. research, I am building DS-Lite topology with the help of
> VPP hopefully.
What's your research about?
> My DS-Lite topology in a nutshell (as every normal ds-lite) consists of 4
> machines: -
> · Sender: IPv4 only machine, sends traffic to the receiver, the
Ameen,
I see you are unhappy with the level of "support" you get on the mailing list.
Note that this is a community. Support is done on a volunteer basis.
You get out what you put in. Contribute back. Give background and show that you
have done your homework, and people might be more willing to
Pankaj,
> In file vnet/ip/ip6_forward.c(VPP 21.01), function ip6_local_inline(),
> the node's packet processing has the following check:
>
> if (PREDICT_FALSE (need_csum))
> {
> flags = ip6_tcp_udp_icmp_validate_checksum (vm, b[0]);
>
Vijay,
> Got the output of "show errors" as:
> 1 gtpu4-input good packets decapsulated error
> Where severity of above log is error.
> Which looks like there is error while packet decapsulation. But this gtpu
> packet is decapsulated successfully and transmitted to next node.
> Is this legacy
Chinmaya,
> We want to get ip route information from VPP using VAPI. We were referencing
> ip.api file, where we see two versions of ip route dump API i.e.
> ip_route_dump and ip_route_v2_dump. So, which one should we use for our
> implementation?
You should use the latest version. I.e. v2.
Akash,
> I have an important query on "show dpdk buffers" allocation. I have only 1
> NUMA node on my PC and buffers-per-numa is increased from 16800 (default) to
> 128000. I am sending 1G Traffic and after some minutes there is "show dpdk
> buffer" output where free becomes 0 and allocated
Hi Andy,
> Is there an equivalent example of snat
> (https://wiki.fd.io/view/VPP/Progressive_VPP_Tutorial#Source_NAT) for
> nat44-ei in version 21.x?
There is a work in progress patch that adds better NAT documentation here:
https://gerrit.fd.io/r/c/vpp/+/32091
If you have the opportunity to
Hi,
> Any vpp documents I can find to compare the 3 different NAT vpp provided as
> plugins (ED, EI and DET), or you can help to tell here? Can any of them be
> combined to use?
RFC4787 gives a good description about the differences between ED and EI.
And RFC7422 for deterministic NAT.
Best
I don't see this message in the open source VPP.
It's likely that the plugin side handler is doing something funky or the API
definition is wrong.
You can always put some prints in the vpp_serializer to see if you get more
insight.
If the code isn't open soure there isn't much more we can do to
Did you include the whole backtrace?
You also want to use python3.
Cheers,
Ole
> On 30 Sep 2021, at 19:52, satish amara wrote:
>
> Hi,
>I am trying to get IKEv2 SA using Python API. I see following issue.
> File "./client.py", line 88, in
> dump_ikev2_sa()
> File "./client.py",
Venu,
Doesn't seem like you are using the higher level C VAPI API nor the low-level
API in vpp-api/client.c.
Are you sure this is what you want?
O.
> On 30 Sep 2021, at 16:09, Venumadhav Josyula wrote:
>
> Hi Akash,
>
> Thansks, now after devtoolset,
>
> @Benoit Ganne
Hi Venu,
> Any examples for VAPI shared to use binary APIs ?
Try this for a start:
https://git.fd.io/vpp/tree/src/vpp-api/vapi/vapi_c_test.c
Cheers,
Ole
signature.asc
Description: Message signed with OpenPGP
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
Hi,
If you look at f.ex
https://git.fd.io/vpp/tree/src/plugins/nat/nat44-ed/nat44_ed_in2out.c#n1199
You will see that the counter dimensions are thread and interface.
Best regards,
Ole
> On 28 Sep 2021, at 11:11, Юрий Иванов wrote:
>
> Hi,
> Many metrics are self explanatory others have
Filip and I have worked on cleaning up the API trace infrastructure.
https://gerrit.fd.io/r/c/vpp/+/32652
api: API trace improvements
Type: improvement
* add support for JSON format in API trace
* add ability to replay JSON API trace in both VPP and VAT2
* use CRC for backward compatibility
Ravi,
> Thanks for your response, I have a question on other language,
> So we don't have any SOCKET API support for C/C++?
> The socket API support is possible with go/python right?
Yes, contributions to VAPI for socket transport is of course welcome!
Best regards,
Ole
signature.asc
Ravi,
The VPP binary API supports two transports. Shared memory and unix domain
sockets.
VAPI is the C language binding for the binary API. I think it only supports
shared memory now.
Other language bindings support both or only sockets.
Cheers,
Ole
> On 22 Sep 2021, at 20:31, RaviKiran
Artem,
That seems to be a bug in the source address selection routine used by ND.
An NS should not be sent out an interface with an address from another
interface.
In this case it should have sent the ND with the link-local address as source.
In IPv6, there is really no such thing as
Prashant,
> Thanks for the feedback.
> Out of curiosity, what is the motivation of this contract about
> minimal length of chained buffer data -- surely, my case being in
> point, the chaining framework should not make any assumptions about
> how the user would use it.
There are lots of features
Ravi,
> We want to use API calls to create the ACL and ABF policies. We need a
> blocking calls because we need to decide next action on the return values.
> When I checked using the VAPI, I see only Asynchronous calls and we need to
> register callback for return values but I need a blocking
out of ports in the deterministic
pool, fall back
to using shared ports.
Cheers,
Ole
> -Mensagem original-
> De: mar...@mgiga.com.br
> Enviada em: quinta-feira, 12 de agosto de 2021 09:40
> Para: 'Ole Troan'
> Cc: 'vpp-dev'
> Assunto: RES: [vpp-dev] CGNAT port assignment
Marcos,
> I’m aware that VPP NAT Plugin has a limitation of 1000 ports per inside
> users, but eventually that amount of connections is not enough.
>
> I would like to get some guidance on how to change that VPP logic When
> assigning ports to users when working with deterministic nat, so
Hyong,
> Thanks for the info, and using 'use_socket=True' did solve the issue
> ('VPPApiClient' was already in use as it was imported as 'VPP' in my code).
> Out of curiosity, why is the python shared memory transport deprecated?
1) there was no performance gain using shared memory over UDS
> I have a python script that uses 'vpp_papi' to try to connect to the 'vpp'
> running on the local host. It's based on my reading of the source code in
> 'src/vpp-api/python/vpp_papi'. One problem I'm running into is that the
> client fails to connect to the vpp with the error message:
>
>
> I think that is telling you not to use /32 address on interface for which you
> expect there to be connected peers.
> From an IP networking perspective, if you want two peers to be connected on
> an interface, they need to be within the same subnet, so use a /31.
For p2p links there are
X,
> I have a query related to the stat_client. Its query is related to
> client/stat_client.c, part of shared library vppapiclient.
>
> In src/vpp-api/client/stat_client.c (master branch):
>
> stat_segment_data_t *res = 0;
> ==>
> I am using vlib_simple_counter_main_t which gives a single count for my
> subscriber database in vpp. Does vpp have a counter that gives one separate
> rx and tx count?
No, you need to use two separate counters.
Ole
signature.asc
Description: Message signed with OpenPGP
Jérôme,
Did you remember to enqueue the new head buffer’s buffer index to the next node
instead of the original?
Cheers
Ole
> On 29 May 2021, at 13:31, jerome.bay...@student.uliege.be wrote:
>
>
> Hello all,
>
> As I explained it in a previous thread, I'm trying to do some IPv6 in IPv6
>
> On 21 May 2021, at 17:15, Neale Ranns wrote:
>
>
> Right, there’s only so much space available. You’ll need to recompile VPP to
> get more space.
> Change the PRE_DATA_SIZE value in src/vlib/CMakeLists.txt.
Alternatively use a new buffer for the new IPv6 header and extension header
Hi Matt,
Would you mind seeing if this solves the issue?
https://gerrit.fd.io/r/c/vpp/+/32358
Cheers,
Ole
> On 14 May 2021, at 22:04, Matthew Smith via lists.fd.io
> wrote:
>
> Hi all,
>
> When I create a GRE tunnel using the API, and then try to look at an API
> trace to see exactly how
Hi Matt,
Might be an oversight in code auto-generation. Looking.
Cheers,
Ole
> On 14 May 2021, at 22:04, Matthew Smith via lists.fd.io
> wrote:
>
> Hi all,
>
> When I create a GRE tunnel using the API, and then try to look at an API
> trace to see exactly how it was added, the
Matt,
> Source file src/vnet/interface.c has a function vnet_rename_interface(). It
> only appears to be called by the lisp plugin currently. It would be handy to
> be able to rename a DPDK interface without having to change startup.conf and
> restart VPP. I am wondering if I could do that by
X,
> Can I get VPP uptime from stats seg? I am using vpp-prometheus to fetch data
> from stats seg and found some promising entries like:
>
> _sys_last_update
> _sys_heartbeat
>
> but not sure of its precise meaning.
You could add a /sys/last_initialized as a timestamp of when VPP "booted".
Hi Paul,
> I'd like to propose three more changes/cleanup to the api files/vppapigen.
>
> 1. I'd like to remove the logic in vpppigen that imperatively declares the
> service definitions. Instead of calculating the values, we would output the
> necessary values to be added to the api file.
All,
As discussed during yesterdays community meeting we plan to merge this patch a
week from now. April 21st.
Object now, or forever hold your peace. ;-)
Best regards,
Ole
API users,
This summer we found an error in the crc generation and to avoid forcing API
users to update their manifests
> i see it , but our Network architecture may be not that easy to fit with
> this plan.
>
> we create many vrf on vpp, and connect to different device per vrf. some
> device may have the same interconnection ip. So when application send to vpp
> the dst ip should be the device's
If this is a monitoring application, would it not be better to send the pings
from an external application through vpp? Eg connected to linux vua tap.
Cheers,
Ole
> On 9 Apr 2021, at 05:18, yichanglui wrote:
>
> On Fri, Apr 9, 2021 at 12:22 AM, Chris Luke wrote:
> yes , my topic should be
+1
Ole
> On 31 Mar 2021, at 19:58, Damjan Marion via lists.fd.io
> wrote:
>
>
> Dear VPP Committers,
>
> I would like to propose Roy Fan Zhang from Intel as a new VPP committer.
> Fan made significant contributions to the VPP including the async crypto
> infrastructure and crypto
> I need to add some special logic to firewall (it is simple eBPF programs for
> DDoS mitigation).
> Currently we use bare DPDK for launching it.
>
> But it will be much better to use it with VPP.
> Are there any plans about adding such feature (something like XDP in Kernel)?
>
> If there are
> I found that having both global and thread data in vlib_main_t is confusing
> to many people and also bug prone.
>
> I wonder if there is sense in. moving global data to new vlib_global_main_t?
>
> I submitted RFC patch and would like to hear what people think about that….
>
>
Paul,
> Thank you for this, but it is not enough. I don't mean that to come across
> ungratefully, but it doesn't fix the problem.
Nothing will ever be enough. Which is good, then we would be out of jobs. :-)
The general problem to API stability does not have a solution.
It's a continual
API users,
This summer we found an error in the crc generation and to avoid forcing API
users to update their manifests on short notice I put in 9f84e70c6d.
Which "cheats" and returns the old crc value instead of the correct crc value
for a set of messages.
> I am in need to print the VPP statistics from an external application. Is
> that possible? I have gone through some links with telemetry plugins and some
> other but couldn't find out the exact thing which I can do to print this from
> my external application. Kindly help me out with this.
Hi,
I'm trying to configure VPP as a public VPN instance.
Doesn't the VPP wireguard implementation support dynamically updating peer's
endpoint address yet?
Would have thought this should work:
wireguard peer add wg0 public-key allowed-ip 192.168.50.0/24
Which it does on Linux but not on VPP.
> Would anybody have seen this problem before on 20.04 ? plugins enabled NAT
> working with default resources
>
> Trying to add the famous
>
> nat {
>endpoint-dependent
>translation hash buckets 1048576
>translation hash memory 268435456
>user hash buckets 1024
>max
Denis,
> Yep, im understand that IPFIX already implemented. But IPFIX doesnt support
> by LI in our country. Only NetFlow v9 with NSEL/NEL is supported
The solution space would include but not limited to:
1) fix country
2) contribute a patch with NSRL/v9
3) write a v9/ IPFIX proxy
Cheers,
Hi Denis,
The NAT implementation already supports ipfix.
https://wiki.fd.io/view/VPP/NAT#NAT_IPFIX_logging
Cheers,
Ole
> On 24 Feb 2021, at 05:56, Denis Lotarev via lists.fd.io
> wrote:
>
> Hi there!
> In Russian Federation we should use NetFlow v9 format (and if we are using
> NAT, then
artup multi-arch variant configuration fix for interfaces
cause: missing review
30998: tests: switch to GPL license
cause: missing review
30929: vppinfra: fix memory issue in mhash
cause: missing review
30930: vppinfra: fix memory issue in mhash
cause: missing review
papi: Ole Troan ,Paul V
And just to be clear, the failing tests are in the QUIC, bihash and IKEv2
features.
> [4]
> https://logs.fd.io/sandbox/vex-yul-rot-jenkins-2/daw_30773_31019-vpp-verify-master-ubuntu2004-aarch64/4/
> vpp-unittest-QUICEchoIntMStreamTestCase-f5ywwo0l-FAILED
>
> 1. Keep the status quo.
> That means using VAT for some operations (e.g. adding multiple routes [5]),
> and creating "exec scripts" [6] for operations without VAT one-liner.
> Pros: No work needed, good speed, old VPP versions are supported.
> Cons: Relying on VAT functionality (outside API
Dave,
> To be perfectly honest, other than Andrew's proposal to tweak the
> auto-abandon parameters, I have not heard another solution that solves the
> problem of cleaning up the current queue and limiting the size of the queue
> in the future. Is anyone going to volunteer to manually
Vratko,
> > Why do you say that vpp_papi need to be dual licensed?
>
> I think e-mail reply would be long,
> and I will need to address comments anyway,
> so I respond via a Gerrit change [1].
I didn't understand the argument made in that patch. Two points for originality
regarding choice of
> On 28 Jan 2021, at 16:59, Dave Wallace wrote:
>
> Lastly, I have not heard a counter-proposal from either you or Ole on how to
> clean up the current state of the queue.
Write a tool that assigns reviews to maintainers. Generate reports and send
nag-o-grams.
Then if that doesn’t work.
My impression is that there is a disconnect between someone putting something
on gerrit and a vpp maintainer reviewing and contributor merging.
I was thinking of having a script processing the review queue and generating
reports for each maintainer. Then give each author a chance to get their
Hi Radu,
Looks like your patch:
https://gerrit.fd.io/r/c/vpp/+/30228
Introduces an unfortunate coupling between VLIB and VNET.
Can you see if you can find a way to avoid calling into VNET from VLIB?
Best regards,
Ole
static_always_inline void
-vlib_update_nr_variant_default
> Looking at the code, it looks like when RA is tuned off in VPP it stops to
> send periodic RA but continues to respond to RS. And this is coherent with
> the logs you shared.
> I think 'suppress-ra' should turn off both periodic RA and RS reponses if I
> read RFC 4861 correctly [1].
> Ole any
Hemant,
https://wiki.fd.io/view/VPP/Pulling,_Building,_Running,_Hacking_and_Pushing_VPP_Code
Best regards,
Ole
> On 19 Jan 2021, at 19:02, hemant via lists.fd.io
> wrote:
>
> I cloned the gerrit VPP repo, created a new classify-val branch using git,
> made code changes, and used commit.
> Thanks for the trace. I was able to create a UT case for this. It is an issue
> in VPP when parsing the destination option header. I will work on it.
The tunnel encapsulation limit option has never been specified for GRE. It's
arguable if this is a protocol compliant packet.
It is also
Hi Andreas,
> I'm trying to run multiple CGNAT setup with the same VPP instance. Each CGNAT
> is supposed to be bound to a unique combination of in/out interface (each
> interface has a different IP table).
>
> What I have found so far is that this does not work. VPP currently can have
> only
Hi Elias,
>>> Would you mind pushing it to gerrit?
>>
>> Here: https://gerrit.fd.io/r/c/vpp/+/30284
>
> I see you added "code review +1" there, thanks!
>
> What more is needed to get it merged? Do we need to add another
> reviewer?
Merged.
Cheers,
Ole
signature.asc
Description: Message
Thanks Elias, merged.
Would you mind fixing that in master too?
Cheers,
Ole
> On 30 Nov 2020, at 21:41, Elias Rudberg wrote:
>
> Hi Ole, thanks for your answer.
>
>> /w/workspace/vpp-verify-2009-ubuntu1804-x86_64/build-root/install-
>>
> For VPP 20.05 the following works to extract /sys/vector_rate
> statistics:
>
> #!/usr/bin/python3
> from vpp_papi.vpp_stats import VPPStats
> stat = VPPStats("/run/vpp/stats.sock")
> dir = stat.ls(['^/sys/vector_rate'])
> counters = stat.dump(dir)
> vector_rate=counters.get('/sys/vector_rate')
Hi Artem,
> I don't fully understand,
> If I want to support configurable UDP port per tunnel, I need to extend
> "vxlan_tunnel_t" or "vxlan_gpe_tunnel_t" or "vxlan_gbp_tunnel_t" or all of
> them. So this is what I meant in point 2. Why 2.5? :)
> Or can I only use the "vxlan_gbp_tunnel_t" for
Artem,
2.5.
Make existing code support configurable UDP port.
Either per system (if so put it in the _main_t structure. Or per tunnel
(vxlan_gbp_tunnel_t) or per makes sense.
Best regards,
Ole
> On 25 Nov 2020, at 20:45, Artem Glazychev wrote:
>
> Hi everyone,
>
> I want to discuss the
> Can we clarify the .api (syntax) so that enums are separate from flags?
> Today, everything is rendered (at least in Python) as an (enum|aenum).IntFlag.
>
> Here is a sample of how that an enum looks like in python:
> class rx_mode(enum.IntFlag): # u32:
> UNKNOWN = 0
> POLLING = 1
>
Dave,
> Before we deprecate vat - and the vat plugins - it would be nice to solve
> the CLI problem. In particular, to make the "binary-api" debug CLI work in
> some reasonable way with vat2.
Agree, I think a proper answer to that question is required gating for vat2
ever becoming a
Hi,
Problems:
- VAT requires hand-written code for CLI and print routines
- VAT only supports parts of the API (people have forgotten to add VAT code
when adding new APIs)
- API trace supports generating "SCRIPT" code for VAT, but that is also partly
supported, and the trace handlers overlap
1 - 100 of 469 matches
Mail list logo
