[Edited Message Follows]
hi,
I want to test 464XLAT feature of VPP19.04 . I tried to build and configure the
NAT and MAP plugin using this guide:
https://wiki.fd.io/view/VPP/NAT#464XLAT
But some CLI command changed , So I using this Configuration :
client (Linux):
sudo ifconfig enp0s8 up
hi,
I want to test 464XLAT feature of VPP19.04 . I tried to build and configure the
NAT and MAP plugin using this guide:
https://wiki.fd.io/view/VPP/NAT#464XLAT
But some CLI command changed , So I using this Configuration :
client (Linux):
sudo ifconfig enp0s8 up sudo ifconfig enp0s8
If I change the mode of TenGE4/0/0 and TenGE4/0/1 to L3 Mode then tx/rx
counters of TenGE7/0/0 and TenGE7/0/1 matches
and the rx counter of TenGE7/0/0 is not zero.
I have checked connectivity and sure that's correct beacuse if i only use
TenGE7/0/0 and TenGE7/0/0 in the L2XC mode
everything is
Hi,
I want to use the L2 xconnect or L2 bridge feature on multiple interface , i.e
suppose I have 4 ports :
x1 and x2 and also y1 and y2.
I want to only do
vpp# set interface l2 xconnect x1 x2
vpp# set interface l2 xconnect x2 x1
vpp# set interface l2 xconnect y1 y2
vpp# set interface l2
*Hi Damjan*
*Thnaks for your answer,*
*I think it is possible (I encountered this problem) to have deadlock even when
I use separate CPU for handoff and NAT processing, Or same CPU with congestion
drop mechanism.*
*I studied the code and separate the CPUs of handoff and NAT processing, in
Hi ...
I know that we need handoff mechanism when running multithread because traffic
for specific inside network user must be processed always on same thread in
both directions and We can not remove handoff node from NAT nodes ,because
handoff is faster than locking mechanism.
So the
Hi
I test VPP NAT plugin performance in NAT44 dynamic translations mode with 40
CPU core .
at the beginning show error command is look like :
vpp#
vpp# show errors
Count Node Reason
15 nat44-out2in-worker-handoff congestion drop
Hi
What is difference between "plugins/flowprobe" and "vnet/flow"?
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#11760): https://lists.fd.io/g/vpp-dev/message/11760
Mute This Topic: https://lists.fd.io/mt/28792277/21656
Group Owner:
hi Matus
Thanks for your answer, can you explain more about this issue?
You mean that I should rewrite "nat_ipfix_logging.c" file to works per thread
or "vnet/ipfix-export/flow_report"
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#11697):
*hi*
*Every time I enable ipfix for NAT, the main thread (vpp_main) is going too
slowly, which is very hard to enter any command and also drop rate increase
that all traffic dropped.*
*but I got worker thread deadlock once after some hours.
* *Is it a normal behavior that Every time I enable
vpp 18.04 NAT plugin has bug or it does not work ???
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#11666): https://lists.fd.io/g/vpp-dev/message/11666
Mute This Topic: https://lists.fd.io/mt/28790237/21656
Group Owner: vpp-dev+ow...@lists.fd.io
I used VPP18.04
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#11658): https://lists.fd.io/g/vpp-dev/message/11658
Mute This Topic: https://lists.fd.io/mt/28790237/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe:
vpp# show interface rx-placement
Thread 1 (vpp_wk_0):
node dpdk-input:
TenGigabitEthernet4/0/0 queue 0 (polling)
TenGigabitEthernet4/0/1 queue 0 (polling)
TenGigabitEthernet7/0/0 queue 0 (polling)
TenGigabitEthernet7/0/1 queue 0 (polling)
TenGigabitEthernet21/0/0 queue 0
# vpp show trace
[...]
--- Start of thread 16 vpp_wk_15 ---
Packet 1
00:00:02:897486: dpdk-input
TenGigabitEthernet4/0/0 rx queue 15
buffer 0x25507a1: current data 14, length 46, free-list 0, clone-count 0,
totlen-nifb 0, trace 0x0
show interface command at the beginning :
vpp# show interface
Name Idx State Counter
Count
TenGigabitEthernet21/0/0 5 up rx packets
6146
rx
if I config vpp without NAT (just routing) it works , but When Config for NAT I
have rx-miss on all interfaces.
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#11647): https://lists.fd.io/g/vpp-dev/message/11647
Mute This Topic:
[Edited Message Follows]
Hi,
I want to test VPP NAT plugin performance in NAT44 dynamic translations mode
while supporting ipfix logging. (with 4 10gb ethernet interface , 40 CPU core
and 40gbps throughput)
based on VPP/NAT wiki page I used following configuration :
unix {
nodaemon
log
Hi,
I want to test VPP NAT plugin performance in NAT44 dynamic translations mode
while supporting ipfix logging. (with 4 10gb ethernet interface , 40 CPU core
and 40gbps throughput)
based on VPP/NAT wiki page I used following configuration :
unix {
nodaemon
log /var/log/vpp/vpp.log
hi
I want to enable an INPUT node just for one worker and disable for others
(thread main and other workers) like dpdk-input node.
when thet state of INPUT node is set to VLIB_NODE_STATE_POLLING , it is enable
for all threads, and when it set to VLIB_NODE_STATE_DISABLED it is disable for
all
[Edited Message Follows]
Do you mean it is impossible to have packet processing and memory operation at
the same time,
for example doing vec_validate or vec_free when NAT plugin is working and
create new session.
I have drop rate when vec_free or vec_vlidate for memory size greater than 3GB.
Do you mean it is impossible to have packet processing and memory operation at
the same time,
for example doing vec_validate or vec_free when NAT plugging is working and
create new session.
I have drop rate when vec_free or vec_vlidate for memory size greater than 3GB.
What are your
hi Matus
I know it will take some time when add new deterministic mapping.
But why "show memory" command has drop rate
Or why adding new deterministic mapping cause drop rate
Thanks !
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#10945):
Hi,
I tested VPP performance for CGNAT in Deterministic mode; While VPP is working
and has sessions, "show memory" command causes huge drop rates or when I want
to add another deterministic mapping it takes long delay (for memory
allocation) and again huge drop rate occurs that relate to
Hi,
Is it possible to use Ncurses library function on VPP CLI Command to have nice
view,
i used a basic sample but does not work, compile with -lncurses and use this
commands :
initscr(); /* Start curses mode */
printw("Hello World !!!"); /* Print Hello
Hi dave
I am using 64 bit image ( -DCLIB_VEC64=1 ) and as I said, I can get vectors
with large size (more than 5GB)
My problem is not the size of vector it is time delay for allocation.
when I using vec_validate to allocate huge memory size it has a very very long
delay to response,
I don't
Hi,
I have a problem when using *vec_validate* to allocate large memory size ( more
than 5G ) , I am sure that i have enough memory on my system,
but vec_validate take so long time to allocate memory how can i decrease this
delay time for allocate memory and how can calculate memory size
26 matches
Mail list logo