Building a Custom Linux kernel for your VOIP System – Part1



Is your VOIP system based on userspace applications ? What about building kernel based VOIP system ?. This is not new but we are going to go through it here. I am assuming Linux as an operating system. The Linux kernel is about core and modules (micro kernel + loadable modules). You might not touch the core but you might need to build a module that support your VOIP system in the kernel for performance purposes. The new module supposed to do VOIP jobs (filtering, QoS, protection, media relay). There are already some useful kernel modules that can support your system like NATing, IPv6ing, and so on . Use them!

In this article i will just introduce what you need to know to follow this series of articles: how to compile and install a new, fresh and clean Linux kernel and what you need to know to compile and install a new kernel module.

We will not touch the current running kernel but building a new one, install it, boot from it, write a new module, add it to the kernel and test it. You need to be very careful when you work in the kernel.

Compilation and Installation Steps

This is a traditional common way to all Linux distributions to compile and install a Linux kernel:

Download the source:

I will work on the kernel 3.19.5 as an example. For other versions, go to and see the list there.

# wget -c

  Decompress the gz file and go to the directory “linux-3.19.5”:

# tar -xzvf linux-3.19.5.tar.gz

# cd linux-3.19.5

Configure the kernel:

Fresh configuration: traditional way: # make menuconfig or the modern way: # make nconfig

Note: To detect memory leaks in the kernel modules, you can use the tool ‘kmemleak’. To use it, you need to compile the kernel with ‘CONFIG_DEBUG_KMEMLEAK’ option enabled. Enable it as following: make menuconfig –> ‘kernel Hacking’ –> ‘Memory Debugging’ –> ‘Kernel Memory Leak Detector’. Enable this option and do save. You can check this file “/sys/kernel/debug/kmemleak” while testing.

You can use the config file of the current running kernel. In Fedora this file exists as /boot/config-$(KernelName). For example: /boot/config-3.19.5-100.fc20.x86_64. Then update it with ‘make menuconfig’ and save.


Prepare for the compilation: # make mrproper

Compile the kernel: # make

Install the compiled modules:

# make modules_install

The modules will be installed in /lib/modules/3.19.5/

Install the kernel: Now we want to copy and rename the image bzImage (arch/x86_64/boot/bzImage) to /boot + creating the initial RAM disk + creating the map/kernel symbols , and update the grub config file). This done by: # make install

After compiling and installing the kernel, we can boot from it (appear in the GRUB startup List).

Header Files needed to compile a new module:

If you want to build your own kernel module, you need to have the header files.  Check the folder /lib/modules/3.19.5/build/ where 3.19.5 is the new kernel version.

The module is header and source files with Makefile file. When it is built we got .ko file. In the Makefile you need to specify the kernel source and the obj-m.

After building the module, it will be ready to be installed. You need to boot from the new kernel and do ‘modprob $ModuleName’

Again this article is just an introduction to this series. More information is coming next.


OpenSIPS and SIP PATH Extension



RFC 3327 (SIP Path Extension) specifies a way to discover and record the sequence of proxies between the UA and the registrar. The constructed routing path will be stored in the database of the registrar along with the corresponding contact address. Each intermediate proxy (NOT terminating routing point) that wants to stay in the routing path, inserts its URI as a new “Path” header field (or as value in the “Path” header field). This mechanism is applied only to requests that are transiting or originating in the home domain. The constructed routing path will be carried as pre-loaded Route set in the upcoming initial requests. This mechanism is similar to the operation of Record-Route mechanism in dialog-initiating requests (In-dialog/Sequential requests) and it can be combined with it (Example – Invite Transaction).

In this article i will talk about SIP Path extension in OpenSIPS context.


Assume we have this scenario:

UA1 – – – – P1 (Outbound Proxy) – – – – P2 – – – – Registrar/Home-Proxy (OpenSIPS) – – – – UA2

The outgoing proxy is the first hop after client’s NAT . The home proxy is a terminal point for SIP routing. The requested path starts by the outgoing proxy and ends at the registrar/home proxy. It is not common that the UA adds a “Path” header field in the request.

OpenSIPS As Registrar/Home-Proxy

OpenSIPS “Registrar” module has “Path” support. When OpenSIPS receives a SIP Register request, it calls the function save(..), among others. Upon calling this function, the values of the Path header(s) along with the contact address will be stored in the database (User Location Table). The contact address will be stored in the column “contact” and the values of Path header(s) will be stored in the column “path”.

In our scenario, if the proxies P1 and P2 receive a Register request and want to be in the path vector, each one adds its own SIP address as a top most value in the “Path” header field and forward the request. When the request reached the registrar, the path vector will be: Path: <;lr>,<;lr>.

The function save(…) takes some flags as a single string parameter. Among others, we have these flags that are related to the “Path” support:

  • p0‘ (OFF Mode): The path vector will be saved in the user location table but it will not be included in the reply of the Register request.
  • p1‘ (Lazy Mode): The path vector will be saved in the user location table and it will be included in the reply only if the UA indicated path support in the request (add the tag “path” as a value in the “Supported” header field of the request: “Supported: path”).
  • p2‘ (Strict Mode): The path vector will be saved in the user location table only if the the UA indicated path support in the registration request (add the tag “path” as a value in the “Supported” header field of the request). If it is not indicated, the request is rejected with “420 – Bad Extension” and the header “Unsupported: path” will be included in the reply along with the received path vector.

The flags above enable “path” support in “registrar” module and specify the mode of this support as well.

According to all above and if the path vector {p2,p1} is sent in the response, the UA can use the path (as preloaded Route set) in the future requests if that is allowed by the policy. In addition to this, any request reached the home proxy can be routed based on that path {p2,p1} which is stored in the database of the registrar.

According to RFC 3327:

The intermediate proxies (not terminating points) should not add a “Path” header field to a request unless the UA add the tag “path” as a value in the “Supported” header field in that request. Here we say the UA indicated support for PATH extension. If the UA indicated path support and the intermediate proxy requires path support in the registrar, then the proxy adds the tag “path” as a value in the “Requires” header field in that request. If the UA has not indicated support for path and the proxy requires support for it in the registrar, the proxy should reject the request with 421 response. This nearly corresponds to using the flag ‘p2’ in the “save” function: save(“location”,”p2″).

The “save(…)” function has also flag ‘v‘ (path received). If this flag is set, the “received” parameter of the first Path URI of a registration {p1} is set as received-uri and the NAT branch flag is set for this contact. This will be explained further down.

To use multiple flags in the “save(…)” function, mix them as one string. Example: save(“location”,”p2v”).

The routing script will have these lines, among others:


if (method == “REGISTER”) {

save(“location”, “p2v”);

Registrar’s Lookup Operation

When for example UA2 wants to call UA1 (Send Invite request). When the request reached OpenSIPS that operates as a registrar/home proxy, it looks up (Call the function lookup(..)) the AOR of UA1 ( and returns:

  • Contact = <sip:UA1@>

  • Path vector= <;lr>,<;lr>

The Request-URI is overwritten by the registered contact address and the Path vector {p2,p1} is inserted as Route set in the same order {p2,p1} in the outgoing Invite request. The topmost Route header field (proxy: p2) is used for routing decision (next hop) in conjunction with the local policy. Calling this function, also sets the destination URI to the first Path URI (outbound proxy: p1).

OpenSIPS As Outbound Proxy with NAT Detection

OpenSIPS has a module named “path” which is designed to be used at intermediate SIP proxies (e.g. outgoing proxy, load balancer,..etc.). Assume we have this scenario:

UA1 – – – NAT – – – P1 (Outbound Proxy -OpenSIPS) – – – P2 – – – Registrar/Home-Proxy

P1 is OpenSIPS that operates as outgoing proxy with NAT detection. To use “path” module, it must be configured in the routing script to be loaded at startup (loadmodule “”). The module “rr” must be loaded before so it must be configured to be loaded before. To insert a path header into Register message, two functions are exported by this module:

  • The function “add_path(…)” which adds a path header in this form “Path: <sip:;lr>” .The IP address is the IP address of the proxy’s outgoing interface. A port is only added if it is not the default (5060). If it is called with username as parameter, then the username will be added to the URI “Path: <sip:username@;lr>“.
  • The function “add_path_received(…)” which adds a path header in this form:

Path: <sip:;received=sip:;lr>

In our scenario, the outbound proxy p1 will call the function “add_path_received” upon receiving a Register message. The received parameter of its Path URI contains the address where the request has been received (NAT’ed address). This is used to pass the NAT’ed address to the registrar through the next proxy if exists. If it is called with username then the username is added to the URI: “Path: <sip:username@;received=sip:;lr>“. Then the proxy will forward the message further to next hop on a way to the registrar.

If the parameter “use_received” exported from “path” module is set to 1 by the outgoing proxy (modparam(“path”, “use_received”, 1), then when the proxy receives a request, the “received” parameter of the its Route URI  (first Route URI ) is evaluated and used as destination URI(route the request to this address).

This behaviour of outbound proxy is considered as complement to registrar behaviour (SIP Path Extension-aware behaviour of Save(…) and Lookup(…)). So Keep the outbound proxy in the constructed path.

The routing script will have these lines, among others:

loadmodule “”

modparam(“path”, “use_received”, 1)

if (method == “REGISTER”) {
if (!add_path_received(“Outbound”))
sl_send_reply(“503”, “Internal Path Error”);



  • SIP “Path” extension is very useful for SIP loadbalancers that are in front of registrars and proxies. As it helps in drawing SIP routing paths.
  • If you like to manipulate the “received” parameter of the Path URI by yourself in the script (not recommended), you can use the script core variables: $Ri which is the IP address of the interface where the request has been received and $Rp which is the port number. Informations like these are detected by the transport of OpenSIPS (under the SIP layer).
  • If you do not understand the structure of the SIP URI. Go Here.

 More Information