Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions: - simply compressing (gz) - compressing and encapsulation as body in a http request/reply - itv
My proposal would be: - either rename the module or embed in another module (e.g., corex) what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming() - have a new module obfuscate to implement some algorithm as proof of concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla miconda@gmail.com wrote:
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions:
- simply compressing (gz)
- compressing and encapsulation as body in a http request/reply
- itv
My proposal would be:
- either rename the module or embed in another module (e.g., corex) what
was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming()
- have a new module obfuscate to implement some algorithm as proof of
concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
-- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Done all changes as you suggested.
1. The code is embedded in corex module. Two new files added named, corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions:
- simply compressing (gz)
- compressing and encapsulation as body in a http request/reply
- itv
My proposal would be:
- either rename the module or embed in another module (e.g., corex) what
was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming()
- have a new module obfuscate to implement some algorithm as proof of
concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
-- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files - git commit -- to commit changes - git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad <shaheryarkh@gmail.com mailto:shaheryarkh@gmail.com> wrote:
Thank you for your valuable suggestions, i appreciate it. I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it. Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc. There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends. Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone. Thank you. On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Hello, I looked at the code and my conclusion was that the name is a bit confusing. While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out. Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given. Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions: - simply compressing (gz) - compressing and encapsulation as body in a http request/reply - itv My proposal would be: - either rename the module or embed in another module (e.g., corex) what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming() - have a new module obfuscate to implement some algorithm as proof of concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http). Of course, these are my opinions, so the discussion can go on for deciding on how to proceed. Cheers, Daniel On 02/08/14 23:57, Muhammad Shahzad wrote: Hi, As already discussed in detail in following email thread, https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html The new Kamailio module obfuscate is ready for testing and can be downloaded at, http://webrtc.voip-demos.com/obfuscate.tbz2 It contains full code, with documentation and usage scenarios etc. Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any. Thank you. -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> - http://www.linkedin.com/in/miconda
These commands do not seem to work for me. Can you please do the patch?
1. adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
2. making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
3. git format-patch does not work,
git format-patch [no-output]
4. Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla miconda@gmail.com wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files
- git commit -- to commit changes
- git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions:
- simply compressing (gz)
- compressing and encapsulation as body in a http request/reply
- itv
My proposal would be:
- either rename the module or embed in another module (e.g., corex) what
was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming()
- have a new module obfuscate to implement some algorithm as proof of
concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
-- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
You should use:
git format-patch -1 <sha>
The <sha> in your case appears to be f415b0c . If it is last commit, should work with:
git format-patch -1 HEAD
Cheers, Daniel
On 05/08/14 11:25, Muhammad Shahzad wrote:
These commands do not seem to work for me. Can you please do the patch?
- adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
- making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
- git format-patch does not work,
git format-patch [no-output]
- Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla <miconda@gmail.com mailto:miconda@gmail.com> wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use: - git add -- to add new files - git commit -- to commit changes - git format-patch -- to get the commit in a file Cheers, Daniel On 05/08/14 01:14, Muhammad Shahzad wrote:Done all changes as you suggested. 1. The code is embedded in corex module. Two new files added named, corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc. The code is available for review and download at, http://webrtc.voip-demos.com/corex.tbz2 Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on. Thank you. On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad <shaheryarkh@gmail.com <mailto:shaheryarkh@gmail.com>> wrote: Thank you for your valuable suggestions, i appreciate it. I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it. Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc. There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends. Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone. Thank you. On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Hello, I looked at the code and my conclusion was that the name is a bit confusing. While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out. Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given. Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions: - simply compressing (gz) - compressing and encapsulation as body in a http request/reply - itv My proposal would be: - either rename the module or embed in another module (e.g., corex) what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming() - have a new module obfuscate to implement some algorithm as proof of concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http). Of course, these are my opinions, so the discussion can go on for deciding on how to proceed. Cheers, Daniel On 02/08/14 23:57, Muhammad Shahzad wrote: Hi, As already discussed in detail in following email thread, https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html The new Kamailio module obfuscate is ready for testing and can be downloaded at, http://webrtc.voip-demos.com/obfuscate.tbz2 It contains full code, with documentation and usage scenarios etc. Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any. Thank you. -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> - http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda
OK, created the patch finally...
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.
I will just let you know i hit my head on wall just 17 times to make it work. :-(
Thank you.
On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla miconda@gmail.com wrote:
You should use:
git format-patch -1 <sha>
The <sha> in your case appears to be f415b0c . If it is last commit, should work with:
git format-patch -1 HEAD
Cheers, Daniel
On 05/08/14 11:25, Muhammad Shahzad wrote:
These commands do not seem to work for me. Can you please do the patch?
- adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
- making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
- git format-patch does not work,
git format-patch [no-output]
- Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files
- git commit -- to commit changes
- git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions:
- simply compressing (gz)
- compressing and encapsulation as body in a http request/reply
- itv
My proposal would be:
- either rename the module or embed in another module (e.g., corex)
what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming()
- have a new module obfuscate to implement some algorithm as proof of
concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
-- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
On 05/08/14 11:55, Muhammad Shahzad wrote:
OK, created the patch finally...
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
I will look over it and push it to the repository if everything is ok.
I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.
Strange, I never had to do it -- did you clone the git repository from sip-router.org?
I will just let you know i hit my head on wall just 17 times to make it work. :-(
Hopefully it was facebook wall :-)
Daniel
Thank you.
On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla <miconda@gmail.com mailto:miconda@gmail.com> wrote:
You should use: git format-patch -1 <sha> The <sha> in your case appears to be f415b0c . If it is last commit, should work with: git format-patch -1 HEAD Cheers, Daniel On 05/08/14 11:25, Muhammad Shahzad wrote:These commands do not seem to work for me. Can you please do the patch? 1. adding files work, git add modules/corex/corex_nio.c modules/corex/corex_nio.h git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml # 2. making commit seems to work, git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h 3. git format-patch does not work, git format-patch [no-output] 4. Also tried, git format-patch master --stdout [no-output] git format-patch master --stdout > /tmp/nio.patch [no-output] I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy. i miss subversion so badly... Thank you. On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use: - git add -- to add new files - git commit -- to commit changes - git format-patch -- to get the commit in a file Cheers, Daniel On 05/08/14 01:14, Muhammad Shahzad wrote:Done all changes as you suggested. 1. The code is embedded in corex module. Two new files added named, corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc. The code is available for review and download at, http://webrtc.voip-demos.com/corex.tbz2 Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on. Thank you. On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad <shaheryarkh@gmail.com <mailto:shaheryarkh@gmail.com>> wrote: Thank you for your valuable suggestions, i appreciate it. I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it. Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc. There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends. Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone. Thank you. On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Hello, I looked at the code and my conclusion was that the name is a bit confusing. While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out. Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given. Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions: - simply compressing (gz) - compressing and encapsulation as body in a http request/reply - itv My proposal would be: - either rename the module or embed in another module (e.g., corex) what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming() - have a new module obfuscate to implement some algorithm as proof of concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http). Of course, these are my opinions, so the discussion can go on for deciding on how to proceed. Cheers, Daniel On 02/08/14 23:57, Muhammad Shahzad wrote: Hi, As already discussed in detail in following email thread, https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html The new Kamailio module obfuscate is ready for testing and can be downloaded at, http://webrtc.voip-demos.com/obfuscate.tbz2 It contains full code, with documentation and usage scenarios etc. Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any. Thank you. -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> - http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda
yup, i always download kamailio from official repo mentioned here,
http://www.kamailio.org/wiki/install/devel/git
Thank you.
On Tue, Aug 5, 2014 at 3:03 PM, Daniel-Constantin Mierla miconda@gmail.com wrote:
On 05/08/14 11:55, Muhammad Shahzad wrote:
OK, created the patch finally...
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
I will look over it and push it to the repository if everything is ok.
I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.
Strange, I never had to do it -- did you clone the git repository from sip-router.org?
I will just let you know i hit my head on wall just 17 times to make it work. :-(
Hopefully it was facebook wall :-)
Daniel
Thank you.
On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
You should use:
git format-patch -1 <sha>
The <sha> in your case appears to be f415b0c . If it is last commit, should work with:
git format-patch -1 HEAD
Cheers, Daniel
On 05/08/14 11:25, Muhammad Shahzad wrote:
These commands do not seem to work for me. Can you please do the patch?
- adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
- making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
- git format-patch does not work,
git format-patch [no-output]
- Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files
- git commit -- to commit changes
- git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions:
- simply compressing (gz)
- compressing and encapsulation as body in a http request/reply
- itv
My proposal would be:
- either rename the module or embed in another module (e.g., corex)
what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming()
- have a new module obfuscate to implement some algorithm as proof of
concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
Hi,
As already discussed in detail in following email thread,
https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html
The new Kamailio module obfuscate is ready for testing and can be downloaded at,
http://webrtc.voip-demos.com/obfuscate.tbz2
It contains full code, with documentation and usage scenarios etc.
Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any.
Thank you.
-- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
I see that the code introduces a new variable $raw_msg. It looks like being message buffer which is returned by $mb, thus redundant.
Cheers, Daniel
On 05/08/14 12:06, Muhammad Shahzad wrote:
yup, i always download kamailio from official repo mentioned here,
http://www.kamailio.org/wiki/install/devel/git
Thank you.
On Tue, Aug 5, 2014 at 3:03 PM, Daniel-Constantin Mierla <miconda@gmail.com mailto:miconda@gmail.com> wrote:
On 05/08/14 11:55, Muhammad Shahzad wrote:OK, created the patch finally... http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patchI will look over it and push it to the repository if everything is ok.I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.Strange, I never had to do it -- did you clone the git repository from sip-router.org <http://sip-router.org>?I will just let you know i hit my head on wall just 17 times to make it work. :-(Hopefully it was facebook wall :-) DanielThank you. On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: You should use: git format-patch -1 <sha> The <sha> in your case appears to be f415b0c . If it is last commit, should work with: git format-patch -1 HEAD Cheers, Daniel On 05/08/14 11:25, Muhammad Shahzad wrote:These commands do not seem to work for me. Can you please do the patch? 1. adding files work, git add modules/corex/corex_nio.c modules/corex/corex_nio.h git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml # 2. making commit seems to work, git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h 3. git format-patch does not work, git format-patch [no-output] 4. Also tried, git format-patch master --stdout [no-output] git format-patch master --stdout > /tmp/nio.patch [no-output] I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy. i miss subversion so badly... Thank you. On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use: - git add -- to add new files - git commit -- to commit changes - git format-patch -- to get the commit in a file Cheers, Daniel On 05/08/14 01:14, Muhammad Shahzad wrote:Done all changes as you suggested. 1. The code is embedded in corex module. Two new files added named, corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc. The code is available for review and download at, http://webrtc.voip-demos.com/corex.tbz2 Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on. Thank you. On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad <shaheryarkh@gmail.com <mailto:shaheryarkh@gmail.com>> wrote: Thank you for your valuable suggestions, i appreciate it. I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it. Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc. There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends. Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone. Thank you. On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Hello, I looked at the code and my conclusion was that the name is a bit confusing. While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out. Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given. Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions: - simply compressing (gz) - compressing and encapsulation as body in a http request/reply - itv My proposal would be: - either rename the module or embed in another module (e.g., corex) what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming() - have a new module obfuscate to implement some algorithm as proof of concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http). Of course, these are my opinions, so the discussion can go on for deciding on how to proceed. Cheers, Daniel On 02/08/14 23:57, Muhammad Shahzad wrote: Hi, As already discussed in detail in following email thread, https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html The new Kamailio module obfuscate is ready for testing and can be downloaded at, http://webrtc.voip-demos.com/obfuscate.tbz2 It contains full code, with documentation and usage scenarios etc. Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any. Thank you. -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> - http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda
humm, it is suppose to contain the original message that was feed to event_route[network:msg], it would be same as $mb till this route changes the contents of it stores the new contents in $mb, then the $mb and $raw_msg would contain different contents (original vs modified), this may be useful in other routes, such as error_route or local_route e.g. for comparison etc.
Of course, you can remove this variable, since i don't have any concrete use of it in my mind (other then comparing any changes to message were done by event_route[network:msg] in other routes).
Thank you.
On Fri, Aug 8, 2014 at 11:07 AM, Daniel-Constantin Mierla <miconda@gmail.com
wrote:
I see that the code introduces a new variable $raw_msg. It looks like being message buffer which is returned by $mb, thus redundant.
Cheers, Daniel
On 05/08/14 12:06, Muhammad Shahzad wrote:
yup, i always download kamailio from official repo mentioned here,
http://www.kamailio.org/wiki/install/devel/git
Thank you.
On Tue, Aug 5, 2014 at 3:03 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
On 05/08/14 11:55, Muhammad Shahzad wrote:
OK, created the patch finally...
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
I will look over it and push it to the repository if everything is ok.
I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.
Strange, I never had to do it -- did you clone the git repository from sip-router.org?
I will just let you know i hit my head on wall just 17 times to make it work. :-(
Hopefully it was facebook wall :-)
Daniel
Thank you.
On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
You should use:
git format-patch -1 <sha>
The <sha> in your case appears to be f415b0c . If it is last commit, should work with:
git format-patch -1 HEAD
Cheers, Daniel
On 05/08/14 11:25, Muhammad Shahzad wrote:
These commands do not seem to work for me. Can you please do the patch?
- adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
- making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
- git format-patch does not work,
git format-patch [no-output]
- Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files
- git commit -- to commit changes
- git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad <shaheryarkh@gmail.com
wrote:
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Hello,
I looked at the code and my conclusion was that the name is a bit confusing.
While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out.
Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given.
Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions:
- simply compressing (gz)
- compressing and encapsulation as body in a http request/reply
- itv
My proposal would be:
- either rename the module or embed in another module (e.g., corex)
what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming()
- have a new module obfuscate to implement some algorithm as proof of
concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http).
Of course, these are my opinions, so the discussion can go on for deciding on how to proceed.
Cheers, Daniel
On 02/08/14 23:57, Muhammad Shahzad wrote:
> Hi, > > As already discussed in detail in following email thread, > > > https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html > > The new Kamailio module obfuscate is ready for testing and can be > downloaded at, > > http://webrtc.voip-demos.com/obfuscate.tbz2 > > It contains full code, with documentation and usage scenarios etc. > > Adding Kamailio dev team, can you guys review the code and add this > module to Kamailio trunk if possible? I will be responsible for its > maintenance and bug fixes etc. if any. > > Thank you. > >
>
Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
Here is the new function you added:
+int pv_get_raw_msg(sip_msg_t *msg, pv_param_t *param, + pv_value_t *res) +{ + str sb; + if(msg==NULL || res==NULL) + return -1; + sb.s = msg->buf; + sb.len = msg->len; + return pv_get_strval(msg, param, res, &sb); +}
And next is the existing function for $mb:
int pv_get_msg_buf(struct sip_msg *msg, pv_param_t *param, pv_value_t *res) { str s; if(msg==NULL) return -1;
s.s = msg->buf; s.len = msg->len; return pv_get_strval(msg, param, res, &s); }
They are the same apart of variables, so no matter where they will be used (before or after event route processing), they point to the same buffer, therefore they will return the same.
Cheers, Daniel
On 08/08/14 12:23, Muhammad Shahzad wrote:
humm, it is suppose to contain the original message that was feed to event_route[network:msg], it would be same as $mb till this route changes the contents of it stores the new contents in $mb, then the $mb and $raw_msg would contain different contents (original vs modified), this may be useful in other routes, such as error_route or local_route e.g. for comparison etc.
Of course, you can remove this variable, since i don't have any concrete use of it in my mind (other then comparing any changes to message were done by event_route[network:msg] in other routes).
Thank you.
On Fri, Aug 8, 2014 at 11:07 AM, Daniel-Constantin Mierla <miconda@gmail.com mailto:miconda@gmail.com> wrote:
I see that the code introduces a new variable $raw_msg. It looks like being message buffer which is returned by $mb, thus redundant. Cheers, Daniel On 05/08/14 12:06, Muhammad Shahzad wrote:yup, i always download kamailio from official repo mentioned here, http://www.kamailio.org/wiki/install/devel/git Thank you. On Tue, Aug 5, 2014 at 3:03 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: On 05/08/14 11:55, Muhammad Shahzad wrote:OK, created the patch finally... http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patchI will look over it and push it to the repository if everything is ok.I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.Strange, I never had to do it -- did you clone the git repository from sip-router.org <http://sip-router.org>?I will just let you know i hit my head on wall just 17 times to make it work. :-(Hopefully it was facebook wall :-) DanielThank you. On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: You should use: git format-patch -1 <sha> The <sha> in your case appears to be f415b0c . If it is last commit, should work with: git format-patch -1 HEAD Cheers, Daniel On 05/08/14 11:25, Muhammad Shahzad wrote:These commands do not seem to work for me. Can you please do the patch? 1. adding files work, git add modules/corex/corex_nio.c modules/corex/corex_nio.h git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml # 2. making commit seems to work, git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h 3. git format-patch does not work, git format-patch [no-output] 4. Also tried, git format-patch master --stdout [no-output] git format-patch master --stdout > /tmp/nio.patch [no-output] I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy. i miss subversion so badly... Thank you. On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use: - git add -- to add new files - git commit -- to commit changes - git format-patch -- to get the commit in a file Cheers, Daniel On 05/08/14 01:14, Muhammad Shahzad wrote:Done all changes as you suggested. 1. The code is embedded in corex module. Two new files added named, corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc. The code is available for review and download at, http://webrtc.voip-demos.com/corex.tbz2 Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on. Thank you. On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad <shaheryarkh@gmail.com <mailto:shaheryarkh@gmail.com>> wrote: Thank you for your valuable suggestions, i appreciate it. I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it. Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc. There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends. Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone. Thank you. On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: Hello, I looked at the code and my conclusion was that the name is a bit confusing. While indeed you can do obfuscation using some embedded language or other meanings from the config file, the module itself doesn't do anything related. It executes an event route for network packet in/out. Such functionaly is of course useful, just to remind the given example with obfuscation. But I think a better name to reflect the behaviour should be given. Maybe obfuscate should be a module that implement some algorithms for obfuscations, like you mentions: - simply compressing (gz) - compressing and encapsulation as body in a http request/reply - itv My proposal would be: - either rename the module or embed in another module (e.g., corex) what was developed for these event route execution. The event route name should be related to network reception/transmission (e.g., network:msg). Also the is_obfuscated_msg() should be something like is_incoming() - have a new module obfuscate to implement some algorithm as proof of concept (e.g., gzip -- i guess plenty of code from gzcompress module is very useful just to take from there). Then others may contribute more, as they have new ideas or from what you listed (http ecapsulation is at least interesting, considering many allow port 80 and inspect for http). Of course, these are my opinions, so the discussion can go on for deciding on how to proceed. Cheers, Daniel On 02/08/14 23:57, Muhammad Shahzad wrote: Hi, As already discussed in detail in following email thread, https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html The new Kamailio module obfuscate is ready for testing and can be downloaded at, http://webrtc.voip-demos.com/obfuscate.tbz2 It contains full code, with documentation and usage scenarios etc. Adding Kamailio dev team, can you guys review the code and add this module to Kamailio trunk if possible? I will be responsible for its maintenance and bug fixes etc. if any. Thank you. -- Daniel-Constantin Mierla - http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> - http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla -http://www.asipto.com http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda-- Daniel-Constantin Mierla http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 -http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
humm, original function must have got lost while moving the code to corex. Anyways, lets just remove this. I send you new patch. Is there anything else need to be changed/removed etc.?
Thank you.
On Fri, Aug 8, 2014 at 4:31 PM, Daniel-Constantin Mierla miconda@gmail.com wrote:
Here is the new function you added:
+int pv_get_raw_msg(sip_msg_t *msg, pv_param_t *param,
- pv_value_t *res)
+{
- str sb;
- if(msg==NULL || res==NULL)
return -1;- sb.s = msg->buf;
- sb.len = msg->len;
- return pv_get_strval(msg, param, res, &sb);
+}
And next is the existing function for $mb:
int pv_get_msg_buf(struct sip_msg *msg, pv_param_t *param, pv_value_t *res) { str s; if(msg==NULL) return -1;
s.s = msg->buf; s.len = msg->len; return pv_get_strval(msg, param, res, &s);}
They are the same apart of variables, so no matter where they will be used (before or after event route processing), they point to the same buffer, therefore they will return the same.
Cheers, Daniel
On 08/08/14 12:23, Muhammad Shahzad wrote:
humm, it is suppose to contain the original message that was feed to event_route[network:msg], it would be same as $mb till this route changes the contents of it stores the new contents in $mb, then the $mb and $raw_msg would contain different contents (original vs modified), this may be useful in other routes, such as error_route or local_route e.g. for comparison etc.
Of course, you can remove this variable, since i don't have any concrete use of it in my mind (other then comparing any changes to message were done by event_route[network:msg] in other routes).
Thank you.
On Fri, Aug 8, 2014 at 11:07 AM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
I see that the code introduces a new variable $raw_msg. It looks like being message buffer which is returned by $mb, thus redundant.
Cheers, Daniel
On 05/08/14 12:06, Muhammad Shahzad wrote:
yup, i always download kamailio from official repo mentioned here,
http://www.kamailio.org/wiki/install/devel/git
Thank you.
On Tue, Aug 5, 2014 at 3:03 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
On 05/08/14 11:55, Muhammad Shahzad wrote:
OK, created the patch finally...
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
I will look over it and push it to the repository if everything is ok.
I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.
Strange, I never had to do it -- did you clone the git repository from sip-router.org?
I will just let you know i hit my head on wall just 17 times to make it work. :-(
Hopefully it was facebook wall :-)
Daniel
Thank you.
On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
You should use:
git format-patch -1 <sha>
The <sha> in your case appears to be f415b0c . If it is last commit, should work with:
git format-patch -1 HEAD
Cheers, Daniel
On 05/08/14 11:25, Muhammad Shahzad wrote:
These commands do not seem to work for me. Can you please do the patch?
- adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
- making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
- git format-patch does not work,
git format-patch [no-output]
- Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files
- git commit -- to commit changes
- git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad < shaheryarkh@gmail.com> wrote:
Thank you for your valuable suggestions, i appreciate it.
I agree about module name, in fact a couple friends have already sent me email privately suggesting the same. May be something like "nio_intercept" would be better, since module only intercepts the network in/output and leaves it to script writer what to do with it.
Regarding embedding in corex, that is interesting, however i need to go through its code to understand it a bit and then attempt to embed it. Later on, i will create separate module which import this functionality from corex and do specific tasks such as compression and faking http etc.
There is a good progress on doubango SDK side as well and at least idoubs would be ready soon (7 to 10 days approx). Then mobile voip developers can fully test this functionality at both ends.
Anyways let me look at corex module and see if and how can i embed this. In the meantime more suggestions and comments on this are warmly welcome from everyone.
Thank you.
On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
> Hello, > > I looked at the code and my conclusion was that the name is a bit > confusing. > > While indeed you can do obfuscation using some embedded language or > other meanings from the config file, the module itself doesn't do anything > related. It executes an event route for network packet in/out. > > Such functionaly is of course useful, just to remind the given > example with obfuscation. But I think a better name to reflect the > behaviour should be given. > > Maybe obfuscate should be a module that implement some algorithms > for obfuscations, like you mentions: > - simply compressing (gz) > - compressing and encapsulation as body in a http request/reply > - itv > > My proposal would be: > - either rename the module or embed in another module (e.g., corex) > what was developed for these event route execution. The event route name > should be related to network reception/transmission (e.g., network:msg). > Also the is_obfuscated_msg() should be something like is_incoming() > - have a new module obfuscate to implement some algorithm as proof > of concept (e.g., gzip -- i guess plenty of code from gzcompress module is > very useful just to take from there). Then others may contribute more, as > they have new ideas or from what you listed (http ecapsulation is at least > interesting, considering many allow port 80 and inspect for http). > > Of course, these are my opinions, so the discussion can go on for > deciding on how to proceed. > > Cheers, > Daniel > > > > > On 02/08/14 23:57, Muhammad Shahzad wrote: > >> Hi, >> >> As already discussed in detail in following email thread, >> >> >> https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html >> >> The new Kamailio module obfuscate is ready for testing and can be >> downloaded at, >> >> http://webrtc.voip-demos.com/obfuscate.tbz2 >> >> It contains full code, with documentation and usage scenarios etc. >> >> Adding Kamailio dev team, can you guys review the code and add this >> module to Kamailio trunk if possible? I will be responsible for its >> maintenance and bug fixes etc. if any. >> >> Thank you. >> >> >> > -- > Daniel-Constantin Mierla - http://www.asipto.com > http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda > >
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
On Fri, Aug 8, 2014 at 4:37 PM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
humm, original function must have got lost while moving the code to corex. Anyways, lets just remove this. I send you new patch. Is there anything else need to be changed/removed etc.?
Thank you.
On Fri, Aug 8, 2014 at 4:31 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Here is the new function you added:
+int pv_get_raw_msg(sip_msg_t *msg, pv_param_t *param,
- pv_value_t *res)
+{
- str sb;
- if(msg==NULL || res==NULL)
return -1;- sb.s = msg->buf;
- sb.len = msg->len;
- return pv_get_strval(msg, param, res, &sb);
+}
And next is the existing function for $mb:
int pv_get_msg_buf(struct sip_msg *msg, pv_param_t *param, pv_value_t *res) { str s; if(msg==NULL) return -1;
s.s = msg->buf; s.len = msg->len; return pv_get_strval(msg, param, res, &s);}
They are the same apart of variables, so no matter where they will be used (before or after event route processing), they point to the same buffer, therefore they will return the same.
Cheers, Daniel
On 08/08/14 12:23, Muhammad Shahzad wrote:
humm, it is suppose to contain the original message that was feed to event_route[network:msg], it would be same as $mb till this route changes the contents of it stores the new contents in $mb, then the $mb and $raw_msg would contain different contents (original vs modified), this may be useful in other routes, such as error_route or local_route e.g. for comparison etc.
Of course, you can remove this variable, since i don't have any concrete use of it in my mind (other then comparing any changes to message were done by event_route[network:msg] in other routes).
Thank you.
On Fri, Aug 8, 2014 at 11:07 AM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
I see that the code introduces a new variable $raw_msg. It looks like being message buffer which is returned by $mb, thus redundant.
Cheers, Daniel
On 05/08/14 12:06, Muhammad Shahzad wrote:
yup, i always download kamailio from official repo mentioned here,
http://www.kamailio.org/wiki/install/devel/git
Thank you.
On Tue, Aug 5, 2014 at 3:03 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
On 05/08/14 11:55, Muhammad Shahzad wrote:
OK, created the patch finally...
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
I will look over it and push it to the repository if everything is ok.
I had to add all file i wanted to include in the commit using "git add", even the ones that are already included in git repo, e.g. corex.xml and corex_mod.c.
Strange, I never had to do it -- did you clone the git repository from sip-router.org?
I will just let you know i hit my head on wall just 17 times to make it work. :-(
Hopefully it was facebook wall :-)
Daniel
Thank you.
On Tue, Aug 5, 2014 at 2:40 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
You should use:
git format-patch -1 <sha>
The <sha> in your case appears to be f415b0c . If it is last commit, should work with:
git format-patch -1 HEAD
Cheers, Daniel
On 05/08/14 11:25, Muhammad Shahzad wrote:
These commands do not seem to work for me. Can you please do the patch?
- adding files work,
git add modules/corex/corex_nio.c modules/corex/corex_nio.h
git status # On branch master # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: modules/corex/corex_nio.c # new file: modules/corex/corex_nio.h # # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # # modified: modules/corex/README # modified: modules/corex/corex_mod.c # modified: modules/corex/doc/corex.xml # modified: modules/corex/doc/corex_admin.xml #
- making commit seems to work,
git commit -m "added support for network io intercept." [master f415b0c] added support for network io intercept. 2 files changed, 268 insertions(+) create mode 100644 modules/corex/corex_nio.c create mode 100644 modules/corex/corex_nio.h
- git format-patch does not work,
git format-patch [no-output]
- Also tried,
git format-patch master --stdout [no-output]
git format-patch master --stdout > /tmp/nio.patch [no-output]
I don't know what is wrong, i have git version 1.7.10.4 on debian wheezy.
i miss subversion so badly...
Thank you.
On Tue, Aug 5, 2014 at 1:26 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
Can you do a commit on your local git clone and send the patch? It is easier to review the changes. Practically use:
- git add -- to add new files
- git commit -- to commit changes
- git format-patch -- to get the commit in a file
Cheers, Daniel
On 05/08/14 01:14, Muhammad Shahzad wrote:
Done all changes as you suggested.
- The code is embedded in corex module. Two new files added named,
corex_nio.h and corex_nio.c. Some minor changes to corex_mod.c to call the methods defined in corex_nio.h. 2. Event route name changed to event_route[network:msg]. 3. function name changed from is_msg_obfuscated to is_incoming. 4. one additional module parameter is added, named network_io_intercept, to enable disable the whole functionality. (default is disabled) 5. The README file and doc folder is updated with relevant documentation, examples and use cases etc.
The code is available for review and download at,
http://webrtc.voip-demos.com/corex.tbz2
Regarding the actual encryption / compression etc., i am planning to add some example PERL / LUA scripts later on.
Thank you.
On Mon, Aug 4, 2014 at 8:19 PM, Muhammad Shahzad < shaheryarkh@gmail.com> wrote:
> Thank you for your valuable suggestions, i appreciate it. > > I agree about module name, in fact a couple friends have already > sent me email privately suggesting the same. May be something like > "nio_intercept" would be better, since module only intercepts the network > in/output and leaves it to script writer what to do with it. > > Regarding embedding in corex, that is interesting, however i need > to go through its code to understand it a bit and then attempt to embed it. > Later on, i will create separate module which import this functionality > from corex and do specific tasks such as compression and faking http etc. > > There is a good progress on doubango SDK side as well and at least > idoubs would be ready soon (7 to 10 days approx). Then mobile voip > developers can fully test this functionality at both ends. > > Anyways let me look at corex module and see if and how can i embed > this. In the meantime more suggestions and comments on this are warmly > welcome from everyone. > > Thank you. > > > > > On Mon, Aug 4, 2014 at 7:55 PM, Daniel-Constantin Mierla < > miconda@gmail.com> wrote: > >> Hello, >> >> I looked at the code and my conclusion was that the name is a bit >> confusing. >> >> While indeed you can do obfuscation using some embedded language or >> other meanings from the config file, the module itself doesn't do anything >> related. It executes an event route for network packet in/out. >> >> Such functionaly is of course useful, just to remind the given >> example with obfuscation. But I think a better name to reflect the >> behaviour should be given. >> >> Maybe obfuscate should be a module that implement some algorithms >> for obfuscations, like you mentions: >> - simply compressing (gz) >> - compressing and encapsulation as body in a http request/reply >> - itv >> >> My proposal would be: >> - either rename the module or embed in another module (e.g., corex) >> what was developed for these event route execution. The event route name >> should be related to network reception/transmission (e.g., network:msg). >> Also the is_obfuscated_msg() should be something like is_incoming() >> - have a new module obfuscate to implement some algorithm as proof >> of concept (e.g., gzip -- i guess plenty of code from gzcompress module is >> very useful just to take from there). Then others may contribute more, as >> they have new ideas or from what you listed (http ecapsulation is at least >> interesting, considering many allow port 80 and inspect for http). >> >> Of course, these are my opinions, so the discussion can go on for >> deciding on how to proceed. >> >> Cheers, >> Daniel >> >> >> >> >> On 02/08/14 23:57, Muhammad Shahzad wrote: >> >>> Hi, >>> >>> As already discussed in detail in following email thread, >>> >>> >>> https://www.mail-archive.com/sr-users@lists.sip-router.org/msg19922.html >>> >>> The new Kamailio module obfuscate is ready for testing and can be >>> downloaded at, >>> >>> http://webrtc.voip-demos.com/obfuscate.tbz2 >>> >>> It contains full code, with documentation and usage scenarios etc. >>> >>> Adding Kamailio dev team, can you guys review the code and add >>> this module to Kamailio trunk if possible? I will be responsible for its >>> maintenance and bug fixes etc. if any. >>> >>> Thank you. >>> >>> >>> >> -- >> Daniel-Constantin Mierla - http://www.asipto.com >> http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda >> >> >
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
-- Daniel-Constantin Mierla http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile.
Cheers, Daniel
On 08/08/14 15:08, Muhammad Shahzad wrote:
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
OK, no problem.
Thank you.
On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla <miconda@gmail.com
wrote:
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile.
Cheers, Daniel
On 08/08/14 15:08, Muhammad Shahzad wrote:
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
There are variable declared in corex_nio.h and this file is included in two .c files. That results in separate global variables in both .c files.
You have to declare them in a .c file and in the .h with 'extern'.
Cheers, Daniel
On 13/08/14 15:05, Muhammad Shahzad wrote:
OK, no problem.
Thank you.
On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla <miconda@gmail.com mailto:miconda@gmail.com> wrote:
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile. Cheers, Daniel On 08/08/14 15:08, Muhammad Shahzad wrote:patched updated, as discussed. http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patch Thank you.-- Daniel-Constantin Mierla http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 -http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
Sorry for late relay, we have had internet blockage due to mass protests here.
Anyways, i think the variables are declared static, so there should be only one copy of them in memory. But anyways let me analyze it in gdb.
Thank you.
On Thu, Aug 14, 2014 at 12:21 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
There are variable declared in corex_nio.h and this file is included in two .c files. That results in separate global variables in both .c files.
You have to declare them in a .c file and in the .h with 'extern'.
Cheers, Daniel
On 13/08/14 15:05, Muhammad Shahzad wrote:
OK, no problem.
Thank you.
On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile.
Cheers, Daniel
On 08/08/14 15:08, Muhammad Shahzad wrote:
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
oops, spoke too soon. These are not declared static.
Anyways, let me apply your suggestion.
Thank you.
On Fri, Aug 15, 2014 at 12:13 AM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
Sorry for late relay, we have had internet blockage due to mass protests here.
Anyways, i think the variables are declared static, so there should be only one copy of them in memory. But anyways let me analyze it in gdb.
Thank you.
On Thu, Aug 14, 2014 at 12:21 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
There are variable declared in corex_nio.h and this file is included in two .c files. That results in separate global variables in both .c files.
You have to declare them in a .c file and in the .h with 'extern'.
Cheers, Daniel
On 13/08/14 15:05, Muhammad Shahzad wrote:
OK, no problem.
Thank you.
On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile.
Cheers, Daniel
On 08/08/14 15:08, Muhammad Shahzad wrote:
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
Fixed as discussed,
webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patch
Thank you.
On Fri, Aug 15, 2014 at 12:15 AM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
oops, spoke too soon. These are not declared static.
Anyways, let me apply your suggestion.
Thank you.
On Fri, Aug 15, 2014 at 12:13 AM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
Sorry for late relay, we have had internet blockage due to mass protests here.
Anyways, i think the variables are declared static, so there should be only one copy of them in memory. But anyways let me analyze it in gdb.
Thank you.
On Thu, Aug 14, 2014 at 12:21 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
There are variable declared in corex_nio.h and this file is included in two .c files. That results in separate global variables in both .c files.
You have to declare them in a .c file and in the .h with 'extern'.
Cheers, Daniel
On 13/08/14 15:05, Muhammad Shahzad wrote:
OK, no problem.
Thank you.
On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile.
Cheers, Daniel
On 08/08/14 15:08, Muhammad Shahzad wrote:
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
OK, I've just pushed the patch to master branch -- thanks for this contribution.
Cheers, Daniel
On 14/08/14 21:45, Muhammad Shahzad wrote:
Fixed as discussed,
webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patch http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patch
Thank you.
On Fri, Aug 15, 2014 at 12:15 AM, Muhammad Shahzad <shaheryarkh@gmail.com mailto:shaheryarkh@gmail.com> wrote:
oops, spoke too soon. These are not declared static. Anyways, let me apply your suggestion. Thank you. On Fri, Aug 15, 2014 at 12:13 AM, Muhammad Shahzad <shaheryarkh@gmail.com <mailto:shaheryarkh@gmail.com>> wrote: Sorry for late relay, we have had internet blockage due to mass protests here. Anyways, i think the variables are declared static, so there should be only one copy of them in memory. But anyways let me analyze it in gdb. Thank you. On Thu, Aug 14, 2014 at 12:21 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: There are variable declared in corex_nio.h and this file is included in two .c files. That results in separate global variables in both .c files. You have to declare them in a .c file and in the .h with 'extern'. Cheers, Daniel On 13/08/14 15:05, Muhammad Shahzad wrote:OK, no problem. Thank you. On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla <miconda@gmail.com <mailto:miconda@gmail.com>> wrote: I will get back to it and push it if all ok -- got caught by some other stuff meanwhile. Cheers, Daniel On 08/08/14 15:08, Muhammad Shahzad wrote:patched updated, as discussed. http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patch Thank you.-- Daniel-Constantin Mierla http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 -http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA-- Daniel-Constantin Mierla http://twitter.com/#!/miconda <http://twitter.com/#%21/miconda> -http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 -http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
That's great. Again thank you so much for you help and cooperation.
Thank you.
On Fri, Aug 15, 2014 at 4:25 PM, Daniel-Constantin Mierla <miconda@gmail.com
wrote:
OK, I've just pushed the patch to master branch -- thanks for this contribution.
Cheers, Daniel
On 14/08/14 21:45, Muhammad Shahzad wrote:
Fixed as discussed,
webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.patch
Thank you.
On Fri, Aug 15, 2014 at 12:15 AM, Muhammad Shahzad shaheryarkh@gmail.com wrote:
oops, spoke too soon. These are not declared static.
Anyways, let me apply your suggestion.
Thank you.
On Fri, Aug 15, 2014 at 12:13 AM, Muhammad Shahzad < shaheryarkh@gmail.com> wrote:
Sorry for late relay, we have had internet blockage due to mass protests here.
Anyways, i think the variables are declared static, so there should be only one copy of them in memory. But anyways let me analyze it in gdb.
Thank you.
On Thu, Aug 14, 2014 at 12:21 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
There are variable declared in corex_nio.h and this file is included in two .c files. That results in separate global variables in both .c files.
You have to declare them in a .c file and in the .h with 'extern'.
Cheers, Daniel
On 13/08/14 15:05, Muhammad Shahzad wrote:
OK, no problem.
Thank you.
On Wed, Aug 13, 2014 at 2:52 PM, Daniel-Constantin Mierla < miconda@gmail.com> wrote:
I will get back to it and push it if all ok -- got caught by some other stuff meanwhile.
Cheers, Daniel
On 08/08/14 15:08, Muhammad Shahzad wrote:
patched updated, as discussed.
http://webrtc.voip-demos.com/0001-added-support-for-network-io-intercept.pat...
Thank you.
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-- Daniel-Constantin Mierlahttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Next Kamailio Advanced Trainings 2014 - http://www.asipto.com Sep 22-25, Berlin, Germany ::: Oct 15-17, San Francisco, USA
-
Daniel-Constantin Mierla -
Muhammad Shahzad