vpp main thread "vpp_main" hang up, if i assign to much cpu to
'cpu>corelist-workers' at startup.conf file.
It ends with dead loop in internal_mallinfo(), is there any way to aviod this
situation?
vpp.log
Description: Binary data
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to
my code is on master commit f5667c3055dbd6755277f085c6778c2b1104aa6e
in stat_segment.c: stat_validate_counter_vector(), create two dimension:
*counter[thread_num][max]
*
--
static void
I used ab with ldp, for nginx bench test.
Below is my start command:
>
> sudo env \
> VCL_CONFIG=/tmp/vcl-test-3af8e.conf \
> VCL_DEBUG=1 \
> LDP_DEBUG=1 \
> LDP_SID_BIT=9 \
> gdb -x /tmp/gdb-3af8e
>
I got the vppcom_session_write_inline assert failds. Obviously apache ab's
write function
I found there's no way to add a custom tls engine, and *tls_register_engine*
only support maximum of 4 tls engine.
So I think it's need to add at least one enum tag, like *CRYPTO_ENGINE_CUSTOM*
to enum *crypto_engine_type_t* for vpp user adding their custom tls engine.
And VCL should also
In the Hoststack app, I use write the flowing function to set local addr to
connect remote socket server
>
> clib_error_t * vnet_connect_uri_with_local_addr(vlib_main_t *vm,
> vnet_connect_args_t
> *a, ip46_address_t *local_addr) { session_endpoint_cfg_t sep =
> SESSION_ENDPOINT_CFG_NULL; int
I test nginx with ldp, then nginx work well, but the cpu to hight. ldp thread
cpu usage is 100% and the cpu mostly wasted at epoll function.
The test apache jmeter with ldp is more worse. Jmeter will start thousand
threads, When simulating multiple users for http perf test.
And each java thread
code: g...@github.com:FDio/vpp.git master :
1a41a35b27da6921d6d86a9f1ad5f1b46e1185f7
if i close tls session which is not in thread 0, VPP will assert error.
Below is more Info:
>
>
>
> DBGvpp# sh session verbose
>
>
>
> Connection State Rx-f
vpp will crash when add more then 256 ip to an interface, and then delete the
interface ip.
Below is the crash stack. The reason is that the ip enabled counter
*ip_enabled_by_sw_if_index* is type u8, and oly support less then 255 ips.
But in my vpn scenario, I simulate more then 256 vpn tunnel
VPP show classify filter crash
g...@github.com:FDio/vpp.git master
>
> DBGvpp# sh classify filter
> Filter Used By Table(s)
> --
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x76e88575 in show_classify_filter_command_fn
The cli *show classify filter* indeed crash, and cli: *show classify filter
verbose* work well.
Code git head is 4a06846dd668d7f687e6770215c38e8feb5f1740
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#15385):
I use *demo.c* create a interface *demo0* with ip 10.0.0.1/29, and then use
vcl+iperf3 connect server ip 10.0.0.11, and then the ASSERT crash happend.
VPP start command: *vpp -c as-startup-de192.conf
*
iperf3 start command:
sudo env \
LD_LIBRARY_PATH=/home/dev/code/net-base/dist/script/../lib:
vlib_buffer_t has no ethernet_header_t, *vnet_gso_header_offset_parser* ** will
failed.
I think if vlib_buffer_t valid l2 header, vnet_gso_header_offset_parser should
skip *ethernet_header_t* parse.
Below is the VPP crash message:
>
> 0: /home/dev/code/net-base/build/vpp/src/vnet/ip/ip.h:205
>
VPP crash when start nginx start with ldp. vpp code is master
78565f38e8436dae9cd3a891b5e5d929209c87f9,
The crash stack is below: Anyone has any solution?
>
> DBGvpp# 0: vl_api_memclnt_delete_t_handler:277: Stale clnt delete index
> 16777215 old epoch 255 cur epoch 0 0:
>
[Edited Message Follows]
Iperf3 throughtput with vcl on vpp-20.05-rc0 is only half of vpp-20.01.
Below is the test topology:
+-+ +-+
|iperf3/Server| |iperf3/client|
+-+ +-+
|
Iperf3 throughtput with vcl on vpp-20.05-rc0 is only half of vpp-20.01.
Below is the test topology:
+-+ +-+
|iperf3/Server| |iperf3/client|
+-+ +-+
| |
+-+
Now vpp nat not support nat alg (
https://support.huawei.com/enterprise/en/doc/EDOC1000174068/9130981d/nat-alg )
, any plan to support it?
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#15824): https://lists.fd.io/g/vpp-dev/message/15824
Mute
Any detail reason for that?
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#15829): https://lists.fd.io/g/vpp-dev/message/15829
Mute This Topic: https://lists.fd.io/mt/72095001/21656
Mute #nat: https://lists.fd.io/mk?hashtag=nat=1480452
Group Owner:
Hi:
Is there any ways to set a interface non-arp?
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#16469): https://lists.fd.io/g/vpp-dev/message/16469
Mute This Topic: https://lists.fd.io/mt/74365885/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Hi, Dave Barach:
I want to a realloc funtion like as bollow, the function args doesn't have the
original mem size, and clib_mem_size seems can't return the right memory size.
Is there any way to get the memory size ?
void *my _realloc ( void * p , size_t s)
{
return p ? clib_mem_realloc ( p ,
Hi, experts:
I find clib_mem_size(...) cann't return the mem size alloc by
clib_mem_alloc(...). I alloc 32 bytes memory, but clib_mem_size return 44
length as following snippet:
>
> u32 len, real_len;
> void* p;
>
> len = 32;
> p = clib_mem_alloc(len);
> real_len = clib_mem_size(p);
>
>
Hi,
Currently vpp's hoststack app's callback ' session_cb_vft_t->
fifo_tuning_callback' always return 0 if act == SESSION_FT_ACTION_ENQUEUED.
If act == SESSION_FT_ACTION_DEQUEUED, the bytes is the lenght of data the
session have send.
So I think the value should be the length of the session recv
hi:
When sending 1500 bytes data with session app_send_dgram api, I find
session-queue node will split the data to two udp packets, then vpp will
sending to two separate IP packets out with no fragmentation. Even I set
interface MTU with 9000, It's still sending two separate IP packets. I think
Hi experts:
There is a svm ASAN check error. My vpp version is 21.01-rc0~394-g798267aaa
DBGvpp#
==445==AddressSanitizer CHECK failed:
../../../../libsanitizer/asan/asan_mapping.h:377 "((AddrIsInMem(p))) != (0)"
(0x0, 0x0)
#0 0x774c158a (/root/net-base/script/../lib/libasan.so.5+0x11658a)
Hi ben:
You can try the following steps:
* make VPP_EXTRA_CMAKE_ARGS=-DVPP_ENABLE_SANITIZE_ADDR=ON build
* make run
* session enable
>
> DBGvpp# session enable
> ==91579==AddressSanitizer CHECK failed:
> ../../../../libsanitizer/asan/asan_mapping.h:377 "((AddrIsInMem(p))) !=
> (0)" (0x0, 0x0)
>
Hi ben:
The issue still exist with patch https://gerrit.fd.io/r/c/vpp/+/29969 (
https://gerrit.fd.io/r/c/vpp/+/29969 )
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#18073): https://lists.fd.io/g/vpp-dev/message/18073
Mute This Topic:
Hi guys:
I create two vpp instance and create two ipsec tunnel, when ping 10.0.0.1 on
vpp1, I get a ping error for p2p type interface on vpp master
(21.01-rc0~548-g5e6d54779).
vpp0 start with vpp0.cli and vpp1 with vpp1.cli. The following is the test
netowrk topology and vpp1 console output:
>
Hi experts,
Here is an ASAN error when run vpp with ASAN enabled.
>
> [root@localhost vpp]# make
> VPP_EXTRA_CMAKE_ARGS=-DVPP_ENABLE_SANITIZE_ADDR=ON build& debug
> make[1]: Entering directory `/home/dev/code/vpp/build-root'
> Arch for platform 'vpp' is native
> Finding source for
Hi Florin,
I tried dtls proto with my custom crypto engine, and got error with
dtls_connect. Vpp bracn is 21.06-rc2, h ere is the debug log:
>
> Breakpoint 1, dtls_connect (tep=0x7fffb8c13b80) at
> /home/dev/code/net-base/.vpp-21.06-rc2/src/vnet/tls/tls.c:1224
> 1224 return ctx_handle;
>
Is there any way setting asan only avoiding error in this function ?
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#19596): https://lists.fd.io/g/vpp-dev/message/19596
Mute This Topic: https://lists.fd.io/mt/83600497/21656
Group Owner:
Hi Damjan Marion,
vector function: clib_mask_compare_u16_x64 has ASAN Issue,
>
> clib_mask_compare_u16_x64 (u16 v, u16 *a, u32 n_elts)
> {
> ...
> u16x32u *av = (u16x32u *) a;
> ...
>
> x = i8x32_pack (v16 == av[0], v16 == av[1]); <- *av[0] will read 64
> bytes, but a[0] only have 2 bytes
Hi Dave Barach:
Subinterface's ip did not deleted after subinterface deleted. Hrere is the test
command
>
> set interface state eth0 up
> create sub-interface eth0 1
> set interface ip addr eth0.1 192.168.1.1/24
> show int addr
> delete sub-interface eth0.1
> set interface ip addr eth0
Hi Florin:
Currently unformat_vnet_uri not implemented following RFC 3986. The syntax
`tcp://10.0.0.1/500` should be `tcp://10.0.0.1:500` in rfc 3986.
I noticed in there is a comment for `unformat_vent_uri` in
`src/vnet/session/application_interface.c`,
```
/**
* unformat a vnet URI
*
*
Hi guys,
In my project, I set tcp mtu 9000 in vpp startup file: tcp { mtu 9000 }, and
set interfaces mtu 1500. When sending tcp packet, vlib_buffer_push_ip4 will add
'dont frag' flag for packet, if tcp data more then 1500 bytes, i will "ip4 MTU
exceeded and DF set" error.
always_inline void *
Hi Vratko Polak,
currently sw_interface_dump will dump all interfaces if sw_if_index == 0. In my
case, i need get interface information by id without by dumping all, but there
is no way for this.
I notice there is a TODO comment for sw_interface_dump in *interface.api* file.
>
> /** \brief
Hi guys:
I have a compiling error on vpp master, is thre anyone have the save problem?
>
>
>
> VPP library version : 21.06
>
>
>
> GIT toplevel dir : /home/dev/code/vpp
>
>
>
> Build type : debug
>
>
>
> C flags : -Wno-address-of-packed-member -g -fPIC -Werror
Hi all,
Vpp master branch build failed, anyone have the save issue
>
>
> @@@ Configuring vpp in
> /home/dev/code/vpp/build-root/build-vpp_debug-native/vpp
> -- The C compiler identification is GNU 4.8.5
> -- Check for working C compiler: /usr/lib64/ccache/cc
> -- Check for working C
I build vpp on centos 7 with devtoolset-9. i find the using of devtoolset-9 was
removeed in the commit, is there any purpose for it?
>
> commit a5167edc66c639e139ffb5de4336c54bb3d8a871
> Author: Damjan Marion
> Date: Fri Jul 2 16:04:26 2021 +0200
>
> build: remove unused files and sections
it's not a good news for me
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#20028): https://lists.fd.io/g/vpp-dev/message/20028
Mute This Topic: https://lists.fd.io/mt/85151908/21656
Mute #vpp-dev:https://lists.fd.io/g/vpp-dev/mutehashtag/vpp-dev
Hi guys,
Currently vapi message length is determined by sizeof(vapi_msg_) when
alloced by vapi_msg_alloc. But if the vapi message has a variable length string
type *string xxx[]* , the message real length can be changed, this will make
vapi message length check error.
For example,
[Edited Message Follows]
Hi Damjan,
Currently command: define also has some problem. In cli
file have the following content:
cat /tmp/a.cli:
define IP1 192.168.200.11
show macro
ping $IP1
in vpp cli will fail with the output:
IP1 192.168.200.11
ping: expecting IP4/IP6 address `$IP1'.
Hi Damjan,
Currently command: define also has some problem. In cli
file have the following content:
cat /tmp/a.cli:
```
define IP1 192.168.200.11 show macro ping $IP1
```
in vpp cli will fail with the output:
```
IP1 192.168.200.11 ping: expecting IP4/IP6 address `$IP1'. Usage: ping
{ | ipv4
[Edited Message Follows]
Hi Matt,
I run two vpp instance (different prefix) in same hyper-v vm, the first vpp
instance prefix is vs, second is vc. I have eth1 and eth2, the config is below:
I start the vs and vc instance one by one, after vc instance started, the first
instance vs stuck in dpdk
Hi Matt,
I run two vpp instance (different prefix) in same hyper-v vm, the first vpp
instance prefix is vs, second is vc. I have eth1 and eth2, the config is below:
I start the vs and vc instance one by one, after vc instance started, the first
instance vs stuck in dpdk wait function. Does one
Hi Guangming,
You are right, you can push a patch to https://gerrit.fd.io/r , ipsec
maintainers will review it
Xiaoming
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#22195): https://lists.fd.io/g/vpp-dev/message/22195
Mute This Topic:
Hi neale,
The function ipsec_sa_get call in esp_encrypt_inline/ esp_decrypt_inline has
mem issue.
In ipsec4-input-feature node, vnet_buffer (b[0])->ipsec.sad_index was setted by
sa,
but the sa maybe freed by main thread just after the node dispatch finish, the
next dispatch loop the sa
Hi neale,
The function ipsec_sa_get call in esp_encrypt_inline/esp_decrypt_inline has mem
issue.
In ipsec4-input-feature node, vnet_buffer (b[0])->ipsec.sad_index was setted by
sa,
but the sa maybe freed by main thread just after the node dispatch finish, the
next dispatch loop the sa
Hi neale,
Currently there is only one way detecting whether the sa is in using by
checking sa counter. But if the ike is timeout and is rekeying the sa which
first used by ipsec4-input-feature, the sa may has been deleted in esp_encrypt
node.
I rewrite the ike plugin, in my test case, there
Neale's suggestion is very usefull, the sa deleting crash was solved by
deleting policy and sa in two barrier calls.
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#22230): https://lists.fd.io/g/vpp-dev/message/22230
Mute This Topic:
Ikev2 not use spd policy, maybe has no the problem.
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#22232): https://lists.fd.io/g/vpp-dev/message/22232
Mute This Topic: https://lists.fd.io/mt/95086868/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Hi Sandeep,
Our project build and run on Centos7 with vpp master very well, you can try the
follow command: ` make VPP_SKIP_EXTERNAL= xdp-tools VPP_EXTRA_CMAKE_ARGS="
-DLDP_HAS_GNU_SOURCE=OFF
-DCMAKE_PROGRAM_PATH:PATH='/opt/rh/devtoolset-9/root/bin'" build`, there may be
compiler error with
50 matches
Mail list logo
