28 Jan
2020
28 Jan
'20
2 p.m.
lreproxy: add new lreproxy module
<!-- Kamailio Pull Request Template -->
<!--
IMPORTANT:
- for detailed contributing guidelines, read:
https://github.com/kamailio/kamailio/blob/master/.github/CONTRIBUTING.md
- pull requests must be done to master branch, unless they are backports
of fixes from master branch to a stable branch
- backports to stable branches must be done with 'git cherry-pick -x
...'
- code is contributed under BSD for core and main components (tm, sl, auth, tls)
- code is contributed GPLv2 or a compatible license for the other components
- GPL code is contributed with OpenSSL licensing exception
-->
#### Pre-Submission Checklist
<!-- Go over all points below, and after creating the PR, tick all the checkboxes
that apply -->
<!-- All points should be verified, otherwise, read the CONTRIBUTING guidelines
from above-->
<!-- If you're unsure about any of these, don't hesitate to ask on
sr-dev mailing list -->
- [ ] Commit message has the format required by CONTRIBUTING guide
- [ ] Commits are split per component (core, individual modules, libs, utils, ...)
- [ ] Each component has a single commit (if not, squash them into one commit)
- [ ] No commits to README files for modules (changes must be done to docbook files
in `doc/` subfolder, the README file is autogenerated)
#### Type Of Change
- [ ] Small bug fix (non-breaking change which fixes an issue)
- [ ] New feature (non-breaking change which adds new functionality)
- [ ] Breaking change (fix or feature that would change existing functionality)
#### Checklist:
<!-- Go over all points below, and after creating the PR, tick the checkboxes that
apply -->
- [ ] PR should be backported to stable branches
- [ ] Tested changes locally
- [ ] Related to issue #XXXX (replace XXXX with an open issue number)
#### Description
<!-- Describe your changes in detail -->
You can view, comment on, or merge this pull request online at:
https://github.com/kamailio/kamailio/pull/2208
-- Commit Summary --
* lreproxy module
-- File Changes --
A src/modules/lreproxy/README.md (1)
A src/modules/lreproxy/lreproxy.c (1976)
A src/modules/lreproxy/lreproxy.h (120)
A src/modules/lreproxy/lreproxy_funcs.c (460)
A src/modules/lreproxy/lreproxy_funcs.h (41)
A src/modules/lreproxy/lreproxy_hash.c (521)
A src/modules/lreproxy/lreproxy_hash.h (70)
-- Patch Links --
https://github.com/kamailio/kamailio/pull/2208.patch
https://github.com/kamailio/kamailio/pull/2208.diff
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208
28 Jan
28 Jan
2:02 p.m.
The lreproxy is a Light-RTP-Engine module for Kamailio project. The lreproxy engine have
developed based LKM for linux. It uses off-loading technique for relaying huge number of
RTP. It could be used in ingress nodes in your voip network. This module use a customised
rtp packet classification that could lead to process faster and less consumption resource
than other rtp engines.
Basically all packet classifications will done based five-tuple fields in Linux kernel
while in lreproxy it could be done with customised 2 or 3 tuple of packet fields.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-579192200
2:07 p.m.
there's no doc at all. Please take a look over other modules and provide the
documentation.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-579194500
2:13 p.m.
Sure, It is under developing.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-579196515
2:24 p.m.
sergey-safarov commented on this pull request.
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ *
+ */
+
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <sys/time.h>
+#include <netinet/in.h>
+#include <netinet/in_systm.h>
+#ifndef __USE_BSD
+#define __USE_BSD
Is required force define `__USE_BSD` on all systems? Say on Linux.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#pullrequestreview-349301250
2:31 p.m.
Where I can get source (description) of `lre proxy server`?
Is ` lre proxy server` fork of [`RTP proxy`](https://www.rtpproxy.org/)?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-579202672
29 Jan
29 Jan
12:30 a.m.
I don't publish the lreproxy engine because of quite bit of updating.
No, Actually is a linux kernel module that would be insert to linux module for working
with this module.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-579465211
9 Mar
9 Mar
10:39 p.m.
The kernel module does not have to be published here, but is it open source? If yes, is it
available somewhere (another github repository or a public portal)?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-596741812
10 Mar
10 Mar
6:25 a.m.
sobomax commented on this pull request.
@@ -0,0 +1,120 @@
+/*
+ * Copyright (C) 2019-2020 Mojtaba Esfandiari.S, Nasim-Telecom
From brief look to it, this is pretty much carboon-copy
of the rtpproxy module, as such original (c) line of the code should be reproduced.
Thanks!
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#pullrequestreview-371660263
6:27 a.m.
sobomax commented on this pull request.
@@ -0,0 +1,1976 @@
+/*
+ * Copyright (C) 2019-2020 Mojtaba Esfandiari.S, Nasim-Telecom
From brief look to it, this is pretty much carboon-copy
of the rtpproxy module, as such original (c) line of the code should be reproduced.
Thanks!
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#pullrequestreview-371660670
12:33 p.m.
mojtabaesfandiari commented on this pull request.
@@ -0,0 +1,1976 @@
+/*
+ * Copyright (C) 2019-2020 Mojtaba Esfandiari.S, Nasim-Telecom
For cause of compatibilities with other codes in Kamailio, I decided to use rtpproxy
module,such as naming of functions, parsing URL, SDP, get media descriptions and etc. But
the process of relaying media is completely different. The LREProxy uses PREROUTING in
netfilter hooks in Linux networking stack.
The LREProxy engine is running pass the last steps of developing, Because it needs to use
netlink library for interacting between user-pace and kernel-space, it took more time.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#discussion_r390188919
1:44 p.m.
@mojtabaesfandiari are you looked [`DPDK`](https://www.dpdk.org/) initiative.
This initiative driven by many hardware and software
[vendors](https://www.dpdk.org/ecosystem/#members).
`DPDK` move packet processing from kernel to network interface card.
`DPDK` allow get network packet processing at 40Gb/s and more speed.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-597020191
1:58 p.m.
@sobomax - thanks for looking into it.
@mojtabaesfandiari - if there are continuous parts of code copied over from an existing
module such as rtpproxy, you have to keep original copyright line and add yours for the
additions. You can split the file in two, one with copied code and one with your code, if
you want to have them clearly separate.
I haven't compared any files to have an opinion on it, but if there is a copyright
dispute, it should be settled before merging.
@mojtabaesfandiari - as I looked a bit over, I noticed large parts of commented code.
Remove them if they are not useful anymore, makes the files big and hard to follow.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-597025761
3:12 p.m.
@sergey-safarov Just a quite little, I should look it more and compare its results with
my results.
Thanks for sharing it
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-597053485
3:21 p.m.
@miconda Sure,
I just pasted these lines "copy rights "at top of my code because i wanted to
obey the roles of adding new module in Kamailio.
I obey these roles in other files, like lreproxy_funcs.h. Is it right?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-597056837
30 Jun
30 Jun
11:33 a.m.
Short note to say that probably branching of 5.4 will happen soon, if you want this merged
for 5.4.x release series, then it has to be made ready during the next days.
For the code you wrote, then you put your copyright. For the code you took from other
places, you need to copy over the copyright.
Anyone else to review this PR?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-651643780
9:46 p.m.
@miconda I had a conversation with the author of the pull request, he wants to rework the
code substancially. I don't think this should be merged in this form.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-651976347
1 Jul
1 Jul
1:21 p.m.
Yes, As @henningw is mentioned, some rework substancially is doing right now, I don't
think it's over for this branching.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-652332858
25 Nov
25 Nov
8:48 p.m.
Considering the existing modules related to RTP
```
rtp_media_server.html
rtpengine.html
rtpproxy.html
```
I would recommend adding RTP at the beginning of the name, to make sure anyone can
immediately assume this is about RTP forwarding/relaying/proxying
rtp_lr or rtp_light or `rtp_lw_relay` lightweight based on your previous name and
presentation
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-733856384
7 Dec
7 Dec
10:46 p.m.
Few questions out of curiosity ...
Did you benchmark RTPengine with the kernel module or without ?
I have the impression that this solution is using libnetfilter, queue, etc ?
Doe it mean that everything is self contain or the code to manage libnetfiler is somewhere
else ?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-740139347
8 Dec
8 Dec
2:02 a.m.
@jchavanton Thanks for asking some important question,
Q1: Yes, Because RTPengine use local hooking in network stack while this module works in
prerouting hooking in network stack.
Q2: We have completely different designed hash search index based of SDP parameter in
session. It no have relation to libnetfilter, Connection tracking or etc.
Q3: It composed of two different layer as i mentioned before. LREProxy engine is developed
with Python Service, Kernel part and Unix sokect section. Also we are planing to implement
the python section with C again, but later:)
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-740233175
4:52 a.m.
About Q1, I am not sure how to interpret your answer, RTP engine is using IPtables with
the kernel module, but you need to build it properly and load it, did you do it explicitly
?
The step to load it :
```
# RUN echo ">>>> building IP iptables-extension" \
# && cd /git/rtpengine/iptables-extension \
# && cp libxt_RTPENGINE.so $(pkg-config xtables --variable=xtlibdir)
# RUN echo ">>>> building Kernel module" \
# && cd /git/rtpengine/kernel-module \
# && make && cp xt_RTPENGINE.ko /lib/modules/$(uname -r)/updates \
# && depmod -a
```
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-740307487
4:54 a.m.
About Q3, can you add a link to the source code how can anyone use it ?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-740307942
9 Dec
9 Dec
10:55 a.m.
Sure, We have to develop a kernel module for lreproxy for forwarding RTP packets based on
SDP properties.
The LREProxy engine will share in git site
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-741600564
5:52 p.m.
Hi Motjaba, I found your answers very confusing, here is what I understand/deduct.
1- We can not conclude that lreproxy is faster than rtpengine since rtpengine was not
tested with its kernel module.
2- No one can use the lreproxy module since there python/C code required is not developed
yet.
In would recommend to close the PR and consider opening later when this module will become
usable by kamailio users.
If you want to conclude that this media proxy is faster than RTPengine, I would also
recommend that you run the tests again making sure that RTPengine is using its Kernel
module.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-741821950
10:59 p.m.
The lreproxy engine with Python had done before and everyone could use the lreproxy after
i push the service in github.
In regard of comparison with rtpengine, i suggest check it in practical scenario.
While the lreproxy could also works with multiple rtpengine services in cloud sulotions.
Just let me share it after preparing README file.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-742013492
11:05 p.m.
Also i could have online demo in realtime scenario to show it's performance. Actually
we used offloading technique in lreproxy. Our resaults are published in our paper. :))
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-742017006
17 Dec
17 Dec
10:18 a.m.
The Lreproxy service is published in my git, The insulation help is in Readme file. It
might be confusing,
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-747260908
20 Dec
20 Dec
2:33 p.m.
@mojtabaesfandiari pushed 1 commit.
c5efad09dd821e1a3f3c2c037a6461c7060775bc lreproxy
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
https://github.com/kamailio/kamailio/pull/2208/files/15fd09bb5e4d49bb9d61ee…
2:37 p.m.
the lreproxy module is updated.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-748595933
21 Dec
21 Dec
1:31 p.m.
@mojtabaesfandiari pushed 1 commit.
beda3fef42dfbbe96d76cf62e6b1a9f56bc957af lreproxy
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
https://github.com/kamailio/kamailio/pull/2208/files/c5efad09dd821e1a3f3c2c…
3 Jan
3 Jan
9:53 a.m.
https://ieeexplore.ieee.org/document/9303608
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-753577000
10:44 p.m.
In Kamailio we accept modules that offer the same/similar functionality that is provided
by other modules. So from this point of view, it is not a problem that we already have
rtpproxy, rtpengine or mediaproxy as alternatives for relaying RTP packets.
Personally, as I understood after a brief check, I can't see this module being more
performant than rtpengine in any way, because rtpengine does already kernel/iptables-based
forwarding. But performance is also not a criterion to accept a contribution.
First criterion is to be sure is no conflict in terms of copyright. Second is not to be
confusing in relation with the other existing modules. And, the third, very important as
well, is that the author of the module is willing to maintain and support the module for
at least a few years.
From my point of view, the second is quite debatable,
because, as I understand, the LREProxy name expands to Light-RTP-Engine-Proxy. In my
opinion, that suggests it is a proxy for the existing RTPEngine, which obviously is not
the case. From what we have now, it is also a little bit confusing considering we have
both RTP+Proxy and RTP+Engine, this one is somehow combining the both.
Also, from a personal point of view, `Proxy` is a generic term (used in things like
RTPProxy or MediaProxy) referring mainly to `get-in/send-out`, while `Engine` suggests
more, like `processing`, `transformation`, ... which clearly the existing RTPEngine does
with encryption/decryption or transcoding.
If LREProxy aims to focus on performance, I would propose to change the `E` in the name
suggest what this module does, like `LRKProxy` (`Light-RTP-Kernel-Proxy`) or `LRIProxy`
(`Light-RTP-IPTable-Proxy`), or meaning of `E` in something that relates, like `Express`
(`Light-RTP-Express-Proxy`). This will greatly avoid any confusion with existing modules,
making this one distinct in name and purpose.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-753666893
7 Jan
7 Jan
3:18 p.m.
@miconda Thanks for your suggestion about its name. I agree with you. The name could be
change LRKProxy or something else.
I dont want to say this module has more performant than rtpengine or compare with other
media engine. But just as you mentioned, the rtpengine does already kernel/iptables-based
forwarding. It means it works in LOCAL INPUT in Netfilter Linux
networking stack to perform packet handling while this module works in PREROUTING in
Netfilter before the routing decision. "All the packets, with no exceptions, hit this
hook, which is reached before the routing decision and after all the IP header sanity
checks are fulfilled.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-756082259
8 Feb
8 Feb
8:32 p.m.
@mojtabaesfandiari did you end-up sharing your code, opensource it ?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-775314438
9 Feb
9 Feb
12:46 p.m.
@jchavanton Sure, feel free for using it right now.
It could be used right now, But because this module is new, it might be updated,
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-775806425
12:49 p.m.
For example, some upcoming updates are:
Multiple NIC relaying.
Video supporting.
Some quite bit changes for relaying RTP to unix socket.
:)
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-775808351
12:57 p.m.
@mojtabaesfandiari is this the source code of the kernel module you mentioned (which I
think was part of Q3 in this thread)?
https://github.com/mojtabaesfandiari/py-lreproxy/tree/main/lib/c
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-775813426
1:07 p.m.
yes.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-775820414
12 Feb
12 Feb
1:28 a.m.
@mojtabaesfandiari
Just made a few tests, documentation is good, well done.
However I got a crash quickly, probably due to miss config, but crash in the kernel module
makes it much more problematic.
```
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195416] RIP:
0010:main_hook_post+0x70/0xb0 [lreproxy_module]
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195422] Code: c1 e5 02 41 83 e5 3c 4d
01 e5 74 e0 41 0f b7 45 02 49 89 f6 69 c0 47 86 c8 61 c1 e8 16 48 8b 1c c5 c0 b3 0c c1 48
85 db 74 c2 <8b> 43 1c 4c 89 e2 4c 89 ee 4c 89 f7 41 89 44 24 0c 0f b7 43 24 66
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195426] RSP: 0018:ffffc39844dd7ad8
EFLAGS: 00010286
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195431] RAX: 000000000000fd02 RBX:
c9ec991dd1f95aa5 RCX: 0000000000000000
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195434] RDX: 00000000d43c0000 RSI:
0000000000000000 RDI: ffff9fa6df1948d8
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195437] RBP: ffffc39844dd7af8 R08:
ffff9fa6df1948d8 R09: 0000000000000006
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195440] R10: 0000000000001550 R11:
0000000000000010 R12: ffff9fa6df194810
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195443] R13: ffff9fa6df194824 R14:
ffff9fa5dd67f100 R15: 00000000000000b4
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195448] FS: 00007f3dda056700(0000)
GS:ffff9fa6fc440000(0000) knlGS:0000000000000000
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195452] CS: 0010 DS: 0000 ES: 0000
CR0: 0000000080050033
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195456] CR2: 000000000293af68 CR3:
000000050d1c0006 CR4: 00000000003606e0
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195462] DR0: 0000000000000000 DR1:
0000000000000000 DR2: 0000000000000000
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195465] DR3: 0000000000000000 DR6:
00000000fffe0ff0 DR7: 0000000000000400
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195467] Call Trace:
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195478] nf_hook_slow+0x48/0xc0
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195485] ip_output+0xcf/0xe0
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195492] ?
__ip_finish_output+0x1d0/0x1d0
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195498] ip_local_out+0x3b/0x50
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195503] ip_send_skb+0x19/0x40
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195509]
udp_send_skb.isra.48+0x156/0x370
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195515] udp_sendmsg+0x6c3/0xd10
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195521] ? ip_do_fragment+0x530/0x530
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195530] ? __switch_to_asm+0x34/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195535] ? __switch_to_asm+0x40/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195540] ? __switch_to_asm+0x34/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195546] ? _cond_resched+0x19/0x40
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195555] ? aa_sk_perm+0x43/0x190
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195559] ? __switch_to_asm+0x34/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195565] inet_sendmsg+0x5d/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195569] ? inet_sendmsg+0x5d/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195577] sock_sendmsg+0x5c/0x70
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195582] __sys_sendto+0x13f/0x180
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195590] ?
hrtimer_nanosleep+0xc0/0x150
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195596] ?
hrtimer_init_sleeper+0x90/0x90
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195601] __x64_sys_sendto+0x28/0x30
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195608] do_syscall_64+0x57/0x190
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195614]
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195618] RIP: 0033:0x7f3de55b9a82
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195623] Code: 48 83 ec 18 44 89 4c 24
08 e8 aa f5 ff ff 44 8b 4c 24 08 89 c5 4d 89 f8 45 89 f2 4c 89 ea 4c 89 e6 89 df b8 2c 00
00 00 0f 05 <48> 3d 00 f0 ff ff 77 3a 89 ef 48 89 44 24 08 e8 da f5 ff ff 48 8b
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195628] RSP: 002b:00007f3dda055b10
EFLAGS: 00000293 ORIG_RAX: 000000000000002c
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195636] RAX: ffffffffffffffda RBX:
000000000000000a RCX: 00007f3de55b9a82
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195639] RDX: 00000000000000ac RSI:
00007f3dd0018bc8 RDI: 000000000000000a
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195641] RBP: 0000000000000000 R08:
00007f3dd0017c2c R09: 0000000000000010
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195644] R10: 0000000000000000 R11:
0000000000000293 R12: 00007f3dd0018bc8
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195647] R13: 00000000000000ac R14:
0000000000000000 R15: 00007f3dd0017c2c
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195652] Modules linked in:
lreproxy_module(OE) xt_nat xt_tcpudp veth xt_conntrack xt_MASQUERADE nf_conntrack_netlink
xfrm_user xfrm_algo xt_addrtype iptable_filter iptable_nat nf_nat nf_conntrack
nf_defrag_ipv6 nf_defrag_ipv4 libcrc32c bpfilter br_netfilter bridge stp llc
nfnetlink_queue nfnetlink_log nfnetlink rfcomm ccm cmac aufs overlay bnep nls_iso8859_1
mei_hdcp intel_rapl_msr snd_sof_pci snd_sof_intel_hda_common snd_soc_hdac_hda
snd_hda_codec_hdmi snd_sof_intel_hda snd_sof_intel_byt snd_sof_intel_ipc snd_sof
snd_sof_xtensa_dsp snd_hda_ext_core snd_soc_acpi_intel_match x86_pkg_temp_thermal
snd_soc_acpi intel_powerclamp snd_soc_core snd_hda_codec_realtek snd_hda_codec_generic
snd_compress coretemp ac97_bus snd_pcm_dmaengine kvm_intel uvcvideo kvm iwlmvm
snd_hda_intel input_leds snd_intel_dspcfg videobuf2_vmalloc joydev snd_hda_codec
videobuf2_memops rapl videobuf2_v4l2 intel_cstate mac80211 serio_raw snd_hda_core
videobuf2_common wmi_bmof libarc4 intel_wmi_thunderbolt snd_hwdep
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195710] thinkpad_acpi btusb videodev
btrtl nvram btbcm btintel ledtrig_audio snd_seq_midi snd_pcm mc iwlwifi bluetooth
snd_seq_midi_event ecdh_generic ecc snd_rawmidi cfg80211 mei_me intel_lpss_pci mei
intel_lpss snd_seq idma64 virt_dma snd_seq_device snd_timer ucsi_acpi
processor_thermal_device typec_ucsi intel_rapl_common intel_soc_dts_iosf typec
intel_pch_thermal snd soundcore int3403_thermal int340x_thermal_zone int3400_thermal
acpi_thermal_rel mac_hid acpi_pad sch_fq_codel parport_pc ppdev lp parport ip_tables
x_tables autofs4 algif_skcipher af_alg dm_crypt hid_generic usbhid hid nouveau
crct10dif_pclmul crc32_pclmul mxm_wmi ghash_clmulni_intel i915 nvme ttm i2c_algo_bit
aesni_intel drm_kms_helper crypto_simd syscopyarea cryptd sysfillrect glue_helper psmouse
e1000e sysimgblt fb_sys_fops nvme_core drm sdhci_pci thunderbolt cqhci sdhci wmi
pinctrl_cannonlake video pinctrl_intel [last unloaded: lreproxy_module]
Feb 11 14:09:04 anton-ThinkPad-P43s kernel: [225916.195783] ---[ end trace
ff79f804c7960a14 ]---
```
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-777834651
2:07 a.m.
I was running with the default config
```
sudo python3 pylreproxy.py
```
lreproxy was using a fake IP not present in the system, maybe this was part of the problem
generating the crash ...
I got one working test.
```
/etc/py_lre_proxy/py_lre_proxy.ini
[DEFAULT]
start_port : 20000
end_port : 30000
current_port : 20000
internal_ip : 192.168.0.111
external_ip : 192.168.0.111
```
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-777853183
6:58 p.m.
I left it running for several hours, no more crash, there seems to be something odd with
kamailio responsiveness degrading over time.
Could be related to my config, did you ever do long lasting end2end tests ?
One obvious missing feature is hosted NAT traversal/port latching, this will greatly
reduce the use cases where it can be use.
Nice work !
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-778279756
15 Feb
15 Feb
1:25 p.m.
@jchavanton - is the module ready to merge in your opinion (docs, code, or what you had
time to look over)?
@mojtabaesfandiari - is it ok to rename the module to LRKProxy as you made a remark on a
comment above?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-779122253
5:11 p.m.
@miconda Sure, although, lreproxy can be searched and tracked on internet, but i think
LRKProxy is better. What do you think? If yes, i have to change the name of lreproxy in
project. let me do it during next days, Also some updates will publish as soon. Thanks
@jchavanton I am trying to check what you mentioned, Just let me a few days, I will update
you during next days.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-779247638
16 Feb
16 Feb
9:37 a.m.
As before, I strongly suggest that you remove all references to "RTPengine" as
it is another product, not a generic name for an RTP proxy. LRKproxy is better.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-779620679
8:59 p.m.
Few comments on the module being ready to merge :
- I noticed the module interface was outdated it should be tested one more time,
`child_init` and `mod_init` may not be working.
- The documentation is good however it is made in the README instead of the XML files
- Based on my quick tests there was an unexpected performance degradation seen this may
not be a real concern if the module is mainly still an alpha POC
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-780015058
17 Feb
17 Feb
10:34 a.m.
@jchavanton,there was a quite bit issue when the lreproxy kernel engine received duplicate
ports. You could check it after update. Thanks
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-780365977
10:38 a.m.
@jchavanton, Would you please let me know more details related to first comment that you
mentioned?
The second comment is accepted and i will do it.
Let me know how do you test it?What unexpected performance degradation did you see?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-780367930
10:42 a.m.
Actually, I am planning for having a workshop of using lreproxy( LRKProxy ) as soon.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-780369541
11:23 a.m.
@jchavanton, Would you please let me know more details
related to first comment that you mentioned?
The second comment is accepted and i will do it.
Let me know how do you test it? What unexpected performance degradation did you see?
@mojtabaesfandiari
Considering that my test report is very limited, we can ignore the comment of signaling
degradation, it may have been caused by a miss-config, like the one causing the crash. I
was simply curious to know if you did some small sustained load tests.
But now that I think about it, it seems obvious you did such tests since you have shared a
presentation comparing load tests performance, I am guessing these tests where using the
module as well. In this case this seems like a significant load testing effort.
My previous comment :
"I left it running for several hours, no more crash, there seems to be something odd
with Kamailio responsiveness degrading over time. Could be related to my config, did you
ever do long lasting end2end tests ?"
I am also glad that you that may have found an issue related to duplicate ports, not sure
I will find more time to test again shortly.
I have a very good impression on the work done !
About my other comment on the module interface, it is is still a problem, it should
trigger a compilation warning, if you do `make clean && make` in the module
folder.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-780388746
21 Feb
21 Feb
2:40 p.m.
@mojtabaesfandiari pushed 1 commit.
0a8717667a0ccdb5ae1ae9f32077d0d9e053d634 lreproxy
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
https://github.com/kamailio/kamailio/pull/2208/files/beda3fef42dfbbe96d76cf…
2:42 p.m.
Please adjust your commit messages to follow the project's guidelines for commit
messages. Just committing with the module name is not explaining anything.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-782843875
3:05 p.m.
@mojtabaesfandiari pushed 0 commits.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
https://github.com/kamailio/kamailio/pull/2208/files/c077c81cb3e02b207eea68…
3:06 p.m.
lreproxy is updated right now.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-782847249
3:07 p.m.
Please adjust your commit messages to follow the
project's guidelines for commit messages. Just committing with the module name is not
explaining anything.
I fixed my commit message in last update. Thanks
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-782847381
22 Feb
22 Feb
3:26 p.m.
I am planning to change lreproxy to lrkproxy. any idea in this regard?
@miconda, let me know do i have to create new PR after renaming module or some things
else?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-783338108
23 Feb
23 Feb
6:01 p.m.
I think it would be good to create a PR with the new name, so it will be a clean merge of
the module. Thanks!
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-784263655
3 Mar
3 Mar
3:23 p.m.
lreprxoy name is changed to lrkproxy
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-789677108
5:55 p.m.
OK, thanks - closing this one then, discussions can continue on the new PR #2657.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#issuecomment-789771983
5:55 p.m.
Closed #2208.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/kamailio/kamailio/pull/2208#event-4402476534
1360
days inactive
1760
days old
59 comments
9 participants
participants (9)
-
Daniel-Constantin Mierla -
Giacomo Vacca
-
Henning Westerholt
-
Julien Chavanton
-
Maksym Sobolyev
-
mojtabaesfandiari
-
Olle E. Johansson
-
sergey-safarov
-
Victor Seva