Re: [vpp-dev] VPP and TCP half-open connection

If your question concerns the host stack, the answer is no. Florin > On Nov 19, 2018, at 5:08 AM, Rubina Bianchi wrote: > > Hi Dear VPP > > Is there any mechanism in VPP to restrict TCP half-open connection count? If > answer is yes how it can be configured? > > Thanks, > Sincerely > -=-=-=-

Re: [vpp-dev] VPP and TCP half-open connection

Hi Rubina, That’s a tcp protocol startup configuration parameter that can be used to preallocate half-open connections and therefore prevent the underlying pool from expanding. It’s especially useful if you expect applications attached to vpp’s host stack to open a large number of tcp connecti

Re: [vpp-dev] How to actively close the client connections in http server? #vnet

. So, apart from calling vnet_disconnect_session () on an rx event, have you done anything else to the code? Regards, Florin > On Nov 21, 2018, at 1:36 AM, Andreas Schultz > wrote: > > Florin Coras mailto:fcoras.li...@gmail.com>> schrieb > am Fr., 18. Mai 2018 um 08:12 Uh

Re: [vpp-dev] How to actively close the client connections in http server? #vnet

> On Nov 21, 2018, at 8:55 AM, Andreas Schultz > wrote: > > Florin Coras mailto:fcoras.li...@gmail.com>> schrieb > am Mi., 21. Nov. 2018 um 17:18 Uhr: > Hi Andreas, > > The trace lower suggests your tcp connection was freed but you still got data > fo

Re: [vpp-dev] vppcom: why __vcl_worker_index is thread local? #vpp

It’s done on purpose in order to avoid constant locking. Applications should accept new session on the right thread. If you plan to distribute the load across multiple workers, then just have them all listen on the same ip:port tuple and vpp will load balance. Florin > On Nov 22, 2018, at 6:5

Re: [vpp-dev] How to actively close the client connections in http server? #vnet

Ow, so the issue is the reset. Could you try to call stream_session_disconnect instead of stream_session_cleanup in the handle? Florin > On Nov 22, 2018, at 6:51 AM, Andreas Schultz > wrote: > > Hi, > > Florin Coras mailto:fcoras.li...@gmail.com>> schrieb > am

Re: [vpp-dev] About in-band telnet/ssh support of VPP

We’re working on adding support via LDP/VCL but anything not finished is pretty much unusable. We’re trying to finish it as soon as possible but no idea if it’ll make it into 19.01 Florin > On Nov 27, 2018, at 6:08 AM, tianye wrote: > > That is a in-band solution. > No fun :-) > > > 在 2018年

Re: [vpp-dev] vppcom: why __vcl_worker_index is thread local? #vpp

I am working on enabling forking with LDP but keep in mind that we mainly support it for testing purposes. That is, it can’t work with statically linked applications and we don’t plan on supporting all possible socket/setsockopts/getsockopts/fcntl options. Florin > On Nov 28, 2018, at 3:15 AM

Re: [vpp-dev] Verify issues (GRE)

Hi Juraj, Those tests exercise the stack in vpp, so they don’t use up linux stack ports. Moreover, both cut-through and through-the-stack tests use self.shm_prefix when connecting to vpp’s binary api. So, as long as that variable is properly updated, VCL and implicitly LDP will attach and use

Re: [vpp-dev] Verify issues (GRE)

> worker process is not handled in this case, though I wonder why > run_in_venv_with_cleanup.sh doesn't clean it up. > > Juraj > > From: Florin Coras [mailto:fcoras.li...@gmail.com] > Sent: Thursday, November 29, 2018 5:04 PM > To: Juraj Linkeš > Cc: Ole Troa

[vpp-dev] sshd with vpp's host stack and LD_PRELOAD

Folks, I know there has been quite some interest in running sshd with the host stack, so here’s a quick update on where that stands. Dave and I have been working on supporting applications that fork children with vcl and, to actually test the code, we’ve decided to run sshd on top of it. Bec

Re: [vpp-dev] VPP/Envoy integration status

Hi Dai, No. The release plan is just a statement of intent. The actual release notes are here [1]. I know there are ongoing efforts in this area but I’m not sure about their current status. Florin [1] https://git.fd.io/vpp/tree/RELEASE.md > On Dec 6, 2018, at 2:34 PM, via Lists.Fd.Io > w

Re: [vpp-dev] regarding cross compiling

Let me reiterate what Dave said: the committer community saw real value in moving to cmake. Those extra build minutes are subtracted from the lives of those who build vpp on a daily basis. Florin > On Dec 19, 2018, at 8:46 AM, Andrew Pinski wrote: > > On Wed, Dec 19, 2018 at 11:29 AM Dave Ba

Re: [vpp-dev] VPP Failing (all?) Verify Jobs

Hi Jon, Yeah, it’s an osleap problem. As far as I can tell, the issue is not with make install-dep step but with this one [1]. Not entirely sure about the solution but maybe a “-f” would force the installation of the newer package? Marco, Vanessa, Ed, any other suggestions? Florin [1] https:

Re: [vpp-dev] VPP Failing (all?) Verify Jobs

Paul came up with a better fix [1]. Rebasing of the patches should solve the verify problems now. Florin [1] https://gerrit.fd.io/r/#/c/16631/ > On Dec 27, 2018, at 9:54 AM, Florin Coras via Lists.Fd.Io > wrote: > > Hi Jon, > > Yeah, it’s an osleap problem. As far as I c

Re: [vpp-dev] Sequence Number Checking in TCP Protocol

Hi Brayan, I’m not entirely sure I understand your question. Obviously, we have sequence validation in tcp as per rfc 793. For details, see tcp_segment_validate in tcp_input.c. As part of that function, we also check for paws as per rfc 1323/7323. Hope this helps, Florin > On Dec 29, 2018,

Re: [vpp-dev] Sequence Number Checking in TCP Protocol

he wants RFC 6528. > > Jim > >> On Dec 29, 2018, at 10:59 AM, Florin Coras wrote: >> >> Hi Brayan, >> >> I’m not entirely sure I understand your question. Obviously, we have >> sequence validation in tcp as per rfc 793. For details, see >> t

Re: [vpp-dev] Sequence Number Checking in TCP Protocol

> On Dec 29, 2018, at 8:26 PM, Jim Thompson wrote: > > > >> On Dec 29, 2018, at 6:42 PM, Florin Coras wrote: >> >> Hi Jim, >> >> That has to do with the initial sequence number generation. > > Understood. Thus the title of "Defe

Re: [vpp-dev] VPP 19.01 + SPDK

Hi Ram, Tomasz cc'ed, amongst others, have been working on integrating SPDK with VPP's host stack. Maybe the patch here [1] can provide some answers. Florin [1] https://review.gerrithub.io/c/spdk/spdk/+/417056 > On Jan 2, 2019, at 2:05 PM, Ramaraj Pandian wrote: > > Hello, > > I am new

Re: [vpp-dev] Sequence Number Checking in TCP Protocol

Hi Jim, Here’s the patch [1]. Regards, Florin [1] https://gerrit.fd.io/r/#/c/16675/ > On Dec 29, 2018, at 10:59 PM, Florin Coras via Lists.Fd.Io > wrote: > > > >> On Dec 29, 2018, at 8:26 PM, Jim Thompson wrote: >> >> >> >>>

Re: [vpp-dev] vpp + spdk - spdk crash on getting work item

Ramaraj, You’ll have to check with the SPDK guys, but I think they’re still based on ubuntu 18.07. If you’re trying to run SPDK with vcl, so not using the patch that I previously pointed you to, probably it won’t work. VCL needs explicit configuration of workers and, from the crash lower, I

Re: [vpp-dev] wireshark vpp dispatch trace dissector merged

Awesome! Florin > On Jan 17, 2019, at 5:11 AM, Dave Barach via Lists.Fd.Io > wrote: > > I’m pleased to announce that our dispatch trace wireshark dissector has been > merged. At some point in the indefinite future, everyone’s favorite distro > will include a copy of wireshark which knows how

Re: [vpp-dev] RFC: buffer manager rework

Awesome! Looking forward to using it in the host stack ;-) Florin > On Jan 25, 2019, at 9:08 AM, Damjan Marion via Lists.Fd.Io > wrote: > > > I am very close to the finish line with buffer management rework patch, and > would like to > ask people to take a look before it is merged. > > htt

Re: [vpp-dev] Packet copy in plugins/nsim

VPP typically works best if configured with a relatively small number of buffers. On the other hand, nsim can induce a large amount of delay, so the buffers cannot be used to temporarily store the data. Florin > On Jan 29, 2019, at 3:19 AM, Raj wrote: > > Hello all, > > While I was browsing

Re: [vpp-dev] VPP send map-register with R=0 on LISP

Does this [1] solve the issue? Florin [1] https://gerrit.fd.io/r/#/c/17151/ > On Jan 29, 2019, at 4:34 PM, Yosvany wrote: > > VPP with LISP send the reachable bit to 0. When use one iid to vni different > that 0. > > How can change the reachable bit to 1 on map-register. > > I am testin

Re: [vpp-dev] Packet copy in plugins/nsim

Hi Raj, > On Jan 30, 2019, at 6:49 AM, Raj wrote: > > Hello Florin, > > Thanks a lot for your reply. > > On Tue, Jan 29, 2019 at 9:15 PM Florin Coras wrote: >> >> VPP typically works best if configured with a relatively small number of >> buffers.

Re: [vpp-dev] VPP delete map-reply from cisco router

Yosvany, The trace you provided shows that the packet was processed okay and then dropped. Do you see any lisp errors if you do “show error” or in syslog? It may turn out to be that we don’t have a parser for some record in the map-reply. Note that testing has been done exclusively with ODL Li

Re: [vpp-dev] VPP delete map-reply from cisco router

e new > wireshark extensions? > > On Wed, Jan 30, 2019 at 1:23 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Yosvany, > > The trace you provided shows that the packet was processed okay and then > dropped. Do you see any lisp errors if you do “show error”

Re: [vpp-dev] VPP send map-register with R=0 on LISP

de enero de 2019 8:37:45 PM GMT-05:00, Florin Coras > mailto:fcoras.li...@gmail.com>> escribió: > Does this [1] solve the issue? > > Florin > > [1] https://gerrit.fd.io/r/#/c/17151/ <https://gerrit.fd.io/r/#/c/17151/> > > On Jan 29, 2019, at 4:34 PM, Yosvany

Re: [vpp-dev] VPP send map-register with R=0 on LISP

remote 192.168.76.60 >14401This is other cisco router sonly working as > XTR > vpp# > vpp# > vpp# > vpp# > vpp# show one st > statistics status > vpp# show one statistics details > [src-EID, dst-EID] [loc-rloc, rmt

Re: [vpp-dev] Packet copy in plugins/nsim

Hi Raj, Partly. Check “show buffers” Florin > On Jan 31, 2019, at 1:04 AM, Raj wrote: > > Hi Florin, > > On Wed, Jan 30, 2019 at 10:01 PM Florin Coras wrote: > >> Because if it can work with all of its buffers cached, without flushing them >> to main memor

Re: [vpp-dev] VPP 19.01 release is out!

Awesome! Thanks, Andrew! Florin > On Jan 31, 2019, at 1:05 PM, Andrew Yourtchenko wrote: > > Hi all, > > I am happy to announce that 19.01 release is now available. > > Release artificats can be found on both Nexus and Packagecloud. > > Thanks to all the contributors and committers, and also

[vpp-dev] vpp-verify-master-centos failures

It seems centos verify jobs are failing with errors of the type: 00:27:16 FAILED: vnet/CMakeFiles/vnet.dir/span/node.c.o 00:27:16 ccache /opt/rh/devtoolset-7/root/bin/cc -DWITH_LIBSSL=1 -Dvnet_EXPORTS -I/w/workspace/vpp-verify-master-centos7/src -I. -Iinclude -Wno-address-of-packed-member -m

Re: [vpp-dev] VPP send map-register with R=0 on LISP

e any Information > Show one eid-table remote > > Conclusion: Cisco XTR receive a bad Map-Reply from VPP. > > · Another problem is that VPP is making a Loop sendingmap-request > to MR/MS. In this map-request can se a map-reply. > > > De: Florin Coras

Re: [vpp-dev] L2 with LISP

end you more details. > > Can send me one example how implement LISP with L2 and bd, (What comand need > use and what make MAC ADDRES TO RLOC???) > > > > > De: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-de

Re: [vpp-dev] vpp-verify-master-centos failures

super fast turning around these changes, so hopefully in a >>> couple hours. >>> >>> Apologies for the trouble. We have seen a 4-6x increase (depending on OS) >>> in the last 5 months >>> and so it finally started pinching my memory reserva

Re: [vpp-dev] iperf3 with vcl

Hi Ram, What os and what vpp version? We’re constantly running iperf3 in our CI in ubuntu 18.04. Thanks, Florin > On Feb 20, 2019, at 6:07 PM, Ramaraj Pandian wrote: > > I am trying to run iperf server and client to measure performance between two > vpp instances as per “https://wiki.fd.io/v

Re: [vpp-dev] iperf3 with vcl

Hi Florin, Its vpp 19.01 on centos, seems like there is a patch available, I > will try that. Thanks. > > From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Florin > Coras > Sent: Thursday, February 21, 2019 8:40 AM > To: Ramaraj Pandian > Cc: vpp-dev@lists

Re: [vpp-dev] New vpp project committer nomination: Paul Vinciguerra

+1 Florin > On Feb 27, 2019, at 5:20 AM, Dave Barach via Lists.Fd.Io > wrote: > > +1... Dave > > From: vpp-dev@lists.fd.io > On Behalf Of Dave Barach via Lists.Fd.Io > Sent: Wednesday, February 27, 2019 7:38 AM > To: vpp-dev@lists.fd.

Re: [vpp-dev] New vpp project committer: Paul Vinciguerra

Congrats, Paul!! Florin > On Feb 28, 2019, at 2:37 PM, Dave Barach via Lists.Fd.Io > wrote: > > Please join me in congratulating Paul on his election as a vpp project > committer! > > Thanks... Dave > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > > Vie

Re: [vpp-dev] multi-threaded application, "epoll_wait" and "epoll_ctl" have "received signal SIGABRT, Aborted".

Interesting. What application are you running and does this [1] fix the issue for you? In short, many of vls’ apis check if the call is coming in on a new pthread and program vcl accordingly if yes. The patch makes sure vls_attr does that as well. Thanks, Florin [1] https://gerrit.fd.io/r/#/c

Re: [vpp-dev] multi-threaded application, "epoll_wait" and "epoll_ctl" have "received signal SIGABRT, Aborted".

eturn rv; > } > @@ -773,8 +776,10 @@ vls_epoll_ctl (vls_handle_t ep_vlsh, int op, > vls_handle_t vlsh, >vls_table_rlock (); >ep_vls = vls_get_and_lock (ep_vlsh); >vls = vls_get_and_lock (vlsh); > + vls_mt_guard (0, VLS_MT_OP_XPOLL); >ep_sh = vls_to_sh (ep_vls); >

Re: [vpp-dev] multi-threaded application, "epoll_wait" and "epoll_ctl" have "received signal SIGABRT, Aborted".

se, you’d probably have to change your app to directly work with vls or vcl. Hope this helps, Florin > > Thanks, > Sharath. > > > On Sat 30 Mar, 2019, 1:27 AM Florin Coras, <mailto:fcoras.li...@gmail.com>> wrote: > Just so I understand, does the patch not fi

Re: [vpp-dev] Packet with custom ip-proto value drops in ip4-local node

Hi Satish, From the error counters lower, it seems that ip local is not the one to drop the packets. Also, if the packet is not tcp or udp, checksum should not computed, as per ip4_local_check_l4_csum [1]. Also, packets that fail the source ip check should be dropped with IP4_ERROR_SPOOFED_LOC

Re: [vpp-dev] VPP 19.04 centos7 RPM package dependency errors

Are we building rpms on a host that has mbedtls installed? Just uninstalling it should disable the tlsmbedtls plugin, which is what we pretty much want. Florin > On Apr 11, 2019, at 1:00 PM, Dave Wallace wrote: > > Tom/Billy, > > Do you know what the current status is with the VPP rpm's on

Re: [vpp-dev] VPP 19.04 centos7 RPM package dependency errors

run in the container? It seems that even python3 has some dependency errors. I guess that once we align the vagrant version with it, things should just work. Florin > On Apr 11, 2019, at 1:55 PM, Ed Kern (ejk) wrote: > > > >> On Apr 11, 2019, at 2:22 PM, Florin Coras &

Re: [vpp-dev] VCL, ld-preload and VPP shared memory better understanding

Hi Ido, You can find some high level documentation about vcl here [1]. In particular, check the presentations at the bottom of the page and [2]. For more details, you’ll have to read the code. If you need to exercise vcl, you may want to try using iperf [3] or the the vcl test server/client ap

Re: [vpp-dev] [csit-dev] VPP 19.04 Release is complete!

Awesome! Thanks for managing this one, Dave! Cheers, Florin > On Apr 24, 2019, at 3:06 PM, Dave Wallace wrote: > > Folks, > > I am pleased to announce that the VPP 19.04 Release is complete and the > release artifacts are available on > PackageCloud:https://packagecloud.io/fdio/release > <

Re: [vpp-dev] how to trace tcp4-output?

Hi Andreas, The tracing should work but because we’re reusing the syn packet to build the syn-ack, it might not show correctly. Could you breakpoint the tcp4-output and try to see if the trace code is hit? As for the checksum, tcp only tags the packet with a request to have it offloaded. If t

Re: [vpp-dev] how to trace tcp4-output?

Unfortunately, as things stand, the tunnel should compute the checksum before encapsulating the traffic. Otherwise the output node won’t be able to find the right headers. Will look into simplifying this. Florin > On Apr 29, 2019, at 8:18 AM, Andreas Schultz > wrote: > > Thanks, that help

Re: [vpp-dev] how to trace tcp4-output?

re-assembly/segmentation. Florin > On Apr 29, 2019, at 9:38 AM, Andreas Schultz > wrote: > > Am Mo., 29. Apr. 2019 um 18:21 Uhr schrieb Florin Coras > mailto:fcoras.li...@gmail.com>>: > Unfortunately, as things stand, the tunnel should compute the checksum before >

Re: [vpp-dev] finding a virtual memory leak in VPP

Hi Andreas, Maybe you allocate a lot of shared memory segments for UPF? I seem to remember you were using the host stack, are you using it in this test? Florin > On May 9, 2019, at 9:17 AM, Andreas Schultz > wrote: > > Hi, > > Something in VPP (most likely my UPF/GTP plugin) is causing an

Re: [vpp-dev] failed to init shmem: tcp_echo server not working

Hi, It seems that clib_mem_create_fd is not working in your vm. What kernel are you running? Regards, Florin > On Jul 1, 2020, at 8:33 AM, sadhanakesa...@gmail.com wrote: > > thank you very much, looks like a version issue - what version are you using? > i built using the master branch > i

Re: [vpp-dev] failed to init shmem: tcp_echo server not working

As replied on the private thread, vpp-echo needs to know about the api prefix you’ve configured (vpp1). Try adding “chroot prefix vpp1” to the list of options. Regards, Florin > On Jul 1, 2020, at 10:46 AM, sadhanakesa...@gmail.com wrote: > > also i tried to get use-svm-api in existing kernel

Re: [vpp-dev] failed to init shmem: tcp_echo server not working

You’re binding and connecting to different IPs: bind: vpp_echo socket-name /var/run/vpp.sock server uri tcp://172.16.1.1/1234 use-svm-api chroot prefix vpp1 connect: test echo client uri tcp://122.16.1.1/24 Moreover, you’re also mixing external echo app (bind) with the builtin one (connect).

Re: [vpp-dev] VPP20.01: vpp-20.01/src/vnet/session/application.c:960 (vnet_listen) assertion `vlib_thread_is_main_w_barrier ()' fails

Hi, vnet_bind_uri can only be called from main thread. If a worker wants to start listening, you should rpc the request to main thread with vlib_rpc_call_main_thread. Regards, Florin > On Jul 6, 2020, at 3:46 AM, ais...@gmail.com wrote: > > [Edited Message Follows] > > HI, > I am trying to

Re: [vpp-dev] using vpp with iperf3

Hi, If you want to run iperf3 through VPP’s host stack, please use [1]. Two vpp instances are enough for the test. Regards, Florin [1] https://wiki.fd.io/view/VPP/HostStack/LDP/iperf > On Jul 7, 2020, at 7:54 AM, sadhanakesa...@gmail.com wrote: > > Hi Team, > i am trying to follow this >

Re: [vpp-dev] executing vpp hoststack setup

Hi, Yes, it can. That is, you can run iperf + vpp on one side and iperf + linux on the other. Regards, Florin > On Jul 8, 2020, at 2:17 PM, sadhanakesa...@gmail.com wrote: > > Hi , > I am trying to execute this example here for hoststack : > https://wiki.fd.io/view/VPP/HostStack/LDP/iperf

Re: [vpp-dev] TCP timer race and another possible TCP issue

Hi Ivan, Thanks for the detailed report! I assume this is a situation where most of the connections time out and the rate limiting we apply on the pending timer queue delays handling for long enough to be in a situation like the one you described. Here’s a draft patch that starts tracking pen

Re: [vpp-dev] hoststack app fifo_tuning_callback enqueue bytes always 0

Hi, First of all, note that fifo tunning is not fully supported yet. If you enable it by setting the SESSION_F_CUSTOM_FIFO_TUNING flag, you will get notifications whenever new data is enqueued with the aim of allowing the app to update the size of the fifo based on needs. Currently the expecta

Re: [vpp-dev] app_send_dgram should do ip fragmentation if data more than snd_mss

Hi, By default udp computes its mss starting from a 1500 mtu. You can avoid this by either changing the default, i.e., in startup.conf under the udp stanza add “mtu ”, or alternatively you can force mss on each connect, by explicitly setting it in the remote endpoint’s transport_endpoint_cfg_t

Re: [vpp-dev] VPP vs Kernel proxy performance

Hi Sebastiano, The test proxy application is just an example, so it’s far from optimized. Nonetheless, last time I tested it was capable of saturating a 10Gbps nic. So some things to consider while debugging: - fifo size configuration. The wiki page does not set a fifo size and as a result a s

Re: [vpp-dev] VPP vs Kernel proxy performance

Hi, It looks like your linux vm iperf3 client could be saturated, so not sure if there’s more that could be done. You can control fifo size from vcl.conf (rx-fifo-size and tx-fifo-size). Regards, Florin > On Jul 21, 2020, at 10:44 AM, sadhanakesa...@gmail.com wrote: > > hi Team, > i tried t

Re: [vpp-dev] VPP vs Kernel proxy performance

"vpp/src/svm/svm_fifo.c:410 (svm_fifo_init_ooo_lookup) assertion > `!rb_tree_is_init (&f->ooo_deq_lookup)' fails" > > What is causing this error? > > Thanks again, > Sebastiano > > [1] https://gist.github.com/sebymiano/8bc582bc6491cc88f5a608d4

Re: [vpp-dev] TCP timer race and another possible TCP issue

e problems with VPP host stack and some of the > VPP plugins/examples, maybe we should consider adding them to the VPP CI, too? > > On Thu, Jul 16, 2020 at 8:33 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Ivan, > > Thanks for the detailed report! &g

Re: [vpp-dev] TCP timer race and another possible TCP issue

h/vpp-tcp-test > <https://github.com/ivan4th/vpp-tcp-test> > > P.S. As the tests do expose some problems with VPP host stack and some of the > VPP plugins/examples, maybe we should consider adding them to the VPP CI, too? > > On Thu, Jul 16, 2020 at 8:33 PM Florin Coras

Re: [vpp-dev] VPP vs Kernel proxy performance

s. > > Thanks again for your support. > > Regards, > Sebastiano > > Il giorno mer 22 lug 2020 alle ore 18:27 Florin Coras <mailto:fcoras.li...@gmail.com>> ha scritto: > Hi Sebastiano, > > You’re running a debug image, so that is expected. Try to run a r

Re: [vpp-dev] TCP timer race and another possible TCP issue

can see there's something like that in srv6-mobile > plugin [1]. > > [1] > https://github.com/travelping/vpp/blob/feature/2005/upf/src/plugins/srv6-mobile/extra/runner.py#L125 > > <https://github.com/travelping/vpp/blob/feature/2005/upf/src/plugins/srv6-mobile/extra/runner.p

Re: [vpp-dev] TCP timer race and another possible TCP issue

ommit/04512323f311ceebfda351672372033b567d37ca > > <https://github.com/travelping/vpp/commit/04512323f311ceebfda351672372033b567d37ca> > [3] > https://github.com/ivan4th/vpp-tcp-test/blob/master/logs/crash-debug-proxy-tcp_lookup_is_valid.log#L71 > > <https://github.com/i

Re: [vpp-dev] TCP timer race and another possible TCP issue

https://github.com/ivan4th/vpp-tcp-test/blob/a3b02ec/start.sh#L34-L35 > <https://github.com/ivan4th/vpp-tcp-test/blob/a3b02ec/start.sh#L34-L35> > On Thu, Jul 23, 2020 at 5:10 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Ivan, > > Updated [1] but I’m not se

Re: [vpp-dev] TCP timer race and another possible TCP issue

van...@gmail.com> is not registered in your account, and you lack > 'forge committer' permission. > > Because of these issues I'm attaching both patches here for now. > > On Thu, Jul 23, 2020 at 6:47 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: >

Re: [vpp-dev] TCP timer race and another possible TCP issue

afe to merge it. > Thanks! > > I will try to see what is the remaining problem with http_static > > On Fri, Jul 24, 2020 at 8:15 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Ivan, > > Adding Vanessa to see if she can help with the account issues. >

Re: [vpp-dev] TCP timer race and another possible TCP issue

b/src/plugins/http_static/static_server.c > index 45d8731af..7772a9ec0 100644 > --- a/src/plugins/http_static/static_server.c > +++ b/src/plugins/http_static/static_server.c > @@ -124,6 +124,7 @@ http_static_server_session_timer_stop (http_session_t * > hs) > return; >clib_sp

Re: [vpp-dev] TCP timer race and another possible TCP issue

cleanup. Thanks for looking into it! Regards, Florin [1] https://gerrit.fd.io/r/c/vpp/+/28096 <https://gerrit.fd.io/r/c/vpp/+/28096> > > > On Mon, Jul 27, 2020 at 10:59 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Ivan, > > Took a look a

Re: [vpp-dev] TCP timer race and another possible TCP issue

e.g., git commit -a --amend) and “git review” them again. That will update the existing patch set in gerrit. Regards, Florin > > > On Tue, Jul 28, 2020 at 9:27 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Ivan, > > Inline. > >> On

Re: [vpp-dev] VPP 2005 is crashing on stopping the VCL applications #vpp-hoststack

Hi Raj, Does stopping include a call to vppcom_app_exit or killing the applications? If the latter, the apps might be killed with some mutexes/spinlocks held. For now, we only support the former. Regards, Florin > On Jul 29, 2020, at 1:49 PM, Raj Kumar wrote: > > Hi, > In my UDP applicatio

Re: [vpp-dev] VPP 2005 is crashing on stopping the VCL applications #vpp-hoststack

ou please suggest some > cleanup functions ( vppcom functions) that I should call before exiting a > thread and the main application for a proper cleanup. > I also tried vppcom_app_destroy() before exiting the main application but > still I am seeing the same issue. > > thanks,

Re: [vpp-dev] VPP 2005 is crashing on stopping the VCL applications #vpp-hoststack

ons ( vppcom functions) that I should call before exiting a >> thread and the main application for a proper cleanup. >> I also tried vppcom_app_destroy() before exiting the main application but >> still I am seeing the same issue. >> >> thanks, >> -Raj >>

Re: [vpp-dev] VPP 2005 is crashing on stopping the VCL applications #vpp-hoststack

; rx-fifo-size 400 > tx-fifo-size 400 > app-scope-local > app-scope-global > use-mq-eventfd > api-socket-name /tmp/vpp-api.sock > } > > thanks, > -Raj > > On Tue, Aug 4, 2020 at 12:08 AM Florin Coras <mailto:fcoras.li...@gmail.com>>

Re: [vpp-dev] VPP 2005 is crashing on stopping the VCL applications #vpp-hoststack

-Raj > > > On Tue, Aug 4, 2020 at 10:24 AM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Raj, > > Interesting. For some reason, the message queue’s underlying > pthread_cond_timedwait does not work in your setup. Not sure what to make of > that, u

Re: [vpp-dev] 100G with iperf3 server using VCL library

Hi Vijay, Cool experiment. More inline. > On Sep 11, 2020, at 9:42 AM, Vijay Sampath wrote: > > Hi, > > I am using iperf3 as a client on an Ubuntu 18.04 Linux machine connected to > another server running VPP using 100G NICs. Both servers are Intel Xeon with > 24 cores. May I ask the freq

Re: [vpp-dev] 100G with iperf3 server using VCL library

Hi Vijay, Inline. > On Sep 11, 2020, at 1:08 PM, Vijay Sampath wrote: > > Hi Florin, > > Thanks for the response. Please see inline: > > On Fri, Sep 11, 2020 at 10:42 AM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Vijay, > > Cool exp

Re: [vpp-dev] 100G with iperf3 server using VCL library

Hi Vijay, Quick replies inline. > On Sep 11, 2020, at 7:27 PM, Vijay Sampath wrote: > > Hi Florin, > > Thanks once again for looking at this issue. Please see inline: > > On Fri, Sep 11, 2020 at 2:06 PM Florin Coras <mailto:fcoras.li...@gmail.com>> w

Re: [vpp-dev] 100G with iperf3 server using VCL library

Hi Vijay, > On Sep 12, 2020, at 10:06 AM, Vijay Sampath wrote: > > Hi Florin, > > On Fri, Sep 11, 2020 at 11:23 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Vijay, > > Quick replies inline. > >> On Sep 11, 2020, at 7:27 PM,

Re: [vpp-dev] 100G with iperf3 server using VCL library

Hi VIjay, > On Sep 12, 2020, at 12:06 PM, Vijay Sampath wrote: > > Hi Florin, > > On Sat, Sep 12, 2020 at 11:44 AM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Vijay, > > >> On Sep 12, 2020, at 10:06 AM, Vijay Sampath > <mailto

Re: [vpp-dev] 100G with iperf3 server using VCL library

3G > for the same numa core vs different. > > Thanks, > > Vijay > > On Sat, Sep 12, 2020 at 2:40 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi VIjay, > > >> On Sep 12, 2020, at 12:06 PM, Vijay Sampath > <mailto:vsamp.

Re: [vpp-dev] 100G with iperf3 server using VCL library

py. VPP delivers the data to iperf3 in a shared memory fifo, but iperf3 then wants it copied into its own buffer. Unfortunately, we can’t optimize this without changing the apps. Regards, Florin > > Thanks, > > Vijay > > On Mon, Sep 14, 2020 at 8:30 PM Florin Coras <mailto:fc

Re: [vpp-dev] 100G with iperf3 server using VCL library

ry from test to test. But I was never able to get beyond > 50G with 10connections with iperf on the other numa node. As I mentioned in > the previous email, when I repeat this test with Linux TCP as the server, I > am able to get 100G no matter which cores I start iperf on. > > Thanks

Re: [vpp-dev] 100G with iperf3 server using VCL library

ions. Thanks for all the help. I have a much better idea now. > > Thanks, > > Vijay > > On Tue, Sep 15, 2020 at 11:25 AM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Vijay, > > Yes, that is the case for this iperf3 test. The data is already

Re: [vpp-dev] 100G with iperf3 server using VCL library

ided completely, but the applications have to still reside > as a separate process. > > Thanks, > > Vijay > > On Tue, Sep 15, 2020 at 11:44 AM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Vijay, > > Currently, builtin applications can

Re: [vpp-dev] 100G with iperf3 server using VCL library

not yet present, > but can be added with little effort. Is that correct? > > Thanks, > > Vijay > > On Tue, Sep 15, 2020 at 1:31 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Vijay, > > Oh, by no means. Builtin applications, i.e., applicati

Re: [vpp-dev] In the host stack, TCP cannot receive data less than 6 bytes？

Hi, TCP coalesces both when sending and when receiving. If the sender really sends only every second a few bytes, those should be the only ones delivered, assuming the receiver reads the data as soon as it was received. To check that everything behaves as expected, try sending only 2 bytes and

Re: [vpp-dev] Distributing VPPCOM TX sessions to different worker thread #vnet #vppcom #TxBuff

Hi Ravi, I assume you’re using vppcom_session_connect to open those udp sessions? If the listeners receive incoming data, make sure you also set the VPPCOM_ATTR_SET_CONNECTED on the listener with vppcom_session_attr() before calling listen. By default udp sessions are opened on thread 1 and s

Re: [vpp-dev] Distributing VPPCOM TX sessions to different worker thread #vnet #vppcom #TxBuff

Hi Ravi, At this time, there’s no option to pin connected udp sessions to workers, instead we default to worker 1. You could try changing the logic here [1] to select a random thread for each connection. We’re not doing that because whenever we receive return traffic we pin the session to the

Re: [vpp-dev] VPP committers: VPP PTL vote

+1 Florin > On Sep 25, 2020, at 12:13 PM, Dave Barach via lists.fd.io > wrote: > > Folks, > > The self-nomination period closed yesterday. We had one self-nomination, from > Damjan Marion. At this point, we can proceed with a vote. > > I’m sure that Damjan will do a great job, so let me start

Re: [vpp-committers] [vpp-dev] VPP committers: VPP PTL vote

Congrats, Damjan!! Those are some huge “shoes" you’ll have to fill but I’m sure they’ll fit ;-) Dave, a few sentences won’t do all your contributions as PTL any justice! Nonetheless, thank you for starting the project, for diligently working towards growing it and for entrusting us, as a commun

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 th

Re: [vpp-dev] fifo tuning callback issue

Hi Bill, Welcome to the project and thanks for the report! Agreed, that is a bug in non-default scenarios like yours where custom fifo tuning is required. Would you be willing to push a patch or should I do it? Regards, Florin > On Nov 11, 2020, at 10:33 AM, Bill Vaughan wrote: > > Hi all,

Re: [vpp-dev] fifo tuning callback issue

ed, Nov 11, 2020 at 2:09 PM Florin Coras <mailto:fcoras.li...@gmail.com>> wrote: > Hi Bill, > > Welcome to the project and thanks for the report! Agreed, that is a bug in > non-default scenarios like yours where custom fifo tuning is required. Would > you be will

Re: [vpp-dev] #vpp #vpp-dev

Hi Nikhil, If your client and server are attached to the same vpp instance, data/segments never make it into vpp. Instead, during the connect, after vpp detects that the two endpoints are both attached locally it sets up the session buffers (fifos) in a piece of shared memory that’s mapped by

<    1   2   3   4   5   6   7   8   >