Re: [vpp-dev] clearing sessions when application exits #vppcom #vpp-hoststack

Hi Tahir, If a vcl app/worker is not closed cleanly, it won’t detach from vpp. In vpp 20.05, the binary api infra will detect that vcl's binary api client crashed, and cleanup the stale app state, but only after about 30s. In 21.01 (master latest) after this [1] vcl can be configured to use

[vpp-dev] clearing sessions when application exits #vppcom #vpp-hoststack

we are using vcl library to send/receive, if the application aborts or kill signal is sent to the application, vcl sessions are still in listening mode even after the thread containing the vcl sessions is closed. we have a process monitor which monitors our application running/not and

[vpp-dev] Stable/2009 branch is *CLOSED* due to release ceremonies...

Dear VPP committers, I am shortly starting the release process for 20.09. Please don’t merge any new patches to VPP stable/2009 branch until I reply to this mail, reopening the branch or until you see the announcement of the release being available. Thanks! --a /* your friendly 20.09

Re: [vpp-dev] vnet buffer leaking on 20.09RC2?

> On 30.09.2020., at 03:49, Chan Wai wrote: > > Hi, vpp-dev, > I am using vpp 20.09RC2. When I bring up the interface, the vnet buffers will > dramatically decreased. > Is this a memory leaking? > > DBGvpp# show version > vpp v20.09-rc2~0-ga87deb77d built by ubuntu on ubuntu-740-1 at >

Re: [vpp-dev] bihash change

> On 30.09.2020., at 17:29, Jon Loeliger via lists.fd.io > wrote: > > On Wed, Sep 30, 2020 at 8:11 AM Damjan Marion via lists.fd.io > > wrote: > > Just a heads-up on under-the-hood change[1] which is merged few minutes ago > and which

Re: [vpp-dev] Support for localsid behaviors (T.Insert, T.Encaps, H.Insert and H.Encaps)

Hi Chinmaya, Not sure I understand what you want to achieve. The SR policy is set to one of either H.Insert or H.Encaps. Any packet steered into that policy will use that behavior. Same applies for the relevant BindingSID. If you want to get both H.Insert and H.Encaps for the same SR Policy,

Re: [vpp-dev] bihash change

On Wed, Sep 30, 2020 at 8:11 AM Damjan Marion via lists.fd.io wrote: > > Just a heads-up on under-the-hood change[1] which is merged few minutes > ago and which affects all features that use bihash. > Now, bihash is allocating memory from the (main) heap, instead of > mmap()-ing it’s own alloc

[vpp-dev] bihash change

Guys, Just a heads-up on under-the-hood change[1] which is merged few minutes ago and which affects all features that use bihash. Now, bihash is allocating memory from the (main) heap, instead of mmap()-ing it’s own alloc arena. There are several beneifs of that approach: - smaller memory

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Filip Thanks for discussing with me. Ok now i understand where you are trying to get. At this point this kind of matching is not supported. At this point you can either get match combination of external-host external-host-port && external external-port in out2in-only twice-nat scenario. I

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Date, Ok now i understand where you are trying to get. At this point this kind of matching is not supported. At this point you can either get match combination of external-host external-host-port && external external-port in out2in-only twice-nat scenario. I will try to look into it and do

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Continued from previous mail For example this rule vpp# nat44 add static mapping tcp local Z.Z.Z.Z 5566 external Y.Y.Y.Y 8080 external-host X.X.X.X 1234 1. check if match source IP and Port in packet with rule "external-host X.X.X.X 1234" 2. check if match dst IP and Port in packet

[vpp-dev] vnet buffer leaking on 20.09RC2?

Hi, vpp-dev, I am using vpp 20.09RC2. When I bring up the interface, the vnet buffers will dramatically decreased. Is this a memory leaking? DBGvpp# show version vpp v20.09-rc2~0-ga87deb77d built by ubuntu on ubuntu-740-1 at 2020-09-30T01:45:40 DBGvpp# show dpdk version DPDK Version:

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Filip (OUTSIDE HOST) X.X.X.X :* -> (NAT ADDRESS)Y.Y.Y.Y:8080 translated to (OUTSIDE HOST)X.X.X.X:* -> (INSIDE HOST)Z.Z.Z.Z:5566 * outside host address should be uniquelly matched to the NAT rule for the (NAT ADDRESS) to (INSIDE HOST) translation ? Yes Other OUTSIDE HOST will not be

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Date, Just to verify you want something like this ? (OUTSIDE HOST) X.X.X.X :* -> (NAT ADDRESS)Y.Y.Y.Y:8080 translated to (OUTSIDE HOST)X.X.X.X:* -> (INSIDE HOST)Z.Z.Z.Z:5566 * outside host address should be uniquelly matched to the NAT rule for the (NAT ADDRESS) to (INSIDE HOST)

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Filip We already used ED mode to re-use the 8080 port But we still cannot limit this rule only for specific source IP. Do you have any idea or advice for this? Thanks and appreciate Regards, Date 寄件者: Filip Varga -X (fivarga - PANTHEON TECH SRO at Cisco)

回覆: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Filip We already used ED mode to re-use the 8080 port But we still cannot limit this rule only for specific source IP. Do you have any idea or advice for this? Thanks and appreciate Regards, Date 寄件者: Filip Varga -X (fivarga - PANTHEON TECH SRO at Cisco)

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Date, I would suggest looking into ED NAT out2in only translations. Just to point out (OUTSIDE HOST) X.X.X.X :* -> (NAT ADDRESS)Y.Y.Y.Y:8080 translated to (OUTSIDE HOST)X.X.X.X:* -> (INSIDE HOST)Z.Z.Z.Z:5566 Let’s go with NAT in VPP set interface nat44 in LAN out WAN nat44 add address (NAT

Re: [vpp-dev] VPP Deterministic NAT Same in/out Interface Not Matching Session

Hi Joshua, Try to run your setup on master branch with recent changes. Few weeks ago i have moved deterministic feature out of snat plugin. Now deterministic feature is running in it’s separate plugin. Please check det44 sub plugin. If you have any issues feel free to write me back. Best

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Nathan [sorry send this mail again, I found that I didn't cc vpp-dev mailing list] so glad to see your reply. My English is poor, so if there is some confusing term, please tell me. I will try my best to describe it with some graph. Just to be sure to understand your use case, you want to

Re: [vpp-dev] vnet buffer leaking on 20.09RC2?

Hi, Yes VPP default to 1024 descriptors per RX and per TX queue for DPDK interface. By default, the number of txq is determined by the number of threads (main thread + worker threads). So if you have 1 interface, 16 rxq and 16 workers, you will allocate 1024 * (16 + (1 + 16)) = 33792 buffers

Re: [vpp-dev] Static NAT rule only match specific source ip or port

Hi Date, Sorry for the late reply, I'm not sure this will be supported by the existing NAT plugin but it might be doable with few additions. Just to be sure to understand your use case, you want to have the following translations happening : * X.X.X.X :* -> Y.Y.Y.Y:8080 translated to

Re: [vpp-dev] Endpoint-Independent Mapping on Determinate NAT

Hi Joshua, This is currently not supported by the deterministic NAT in VPP. Not sure if/when this will be added, so in the meantime you might want to consider customizing the plugin for yourself. Start from snat_det_find_ses_by_in() function where the session entry matching is being done and

Re: [vpp-dev] Endpoint-Independent Mapping on Determinate NAT

Just to clarify, the filtering behavior I’m looking for is often known as “full cone” or “pure cone” NAT. > On Sep 29, 2020, at 6:33 PM, Joshua Moore wrote: > >  > Hello, > > I have a need to relax the session lookup criteria on out2in packet > processing with NAT44 determinate mode. The