1 \documentclass[10pt]{article}
2 \usepackage[ansinew]{inputenc}
3 \usepackage{makeidx,amsmath,amssymb,exscale,multicol,epsfig,graphics,verbatim,ulem}
4 \usepackage{epsfig,geometry,url,listings, subcaption}
5 \usepackage{boxedminipage}
6 \usepackage[T1]{fontenc}%required
8 \geometry{headsep=3ex,hscale=0.9}
10 \hypersetup{pdftitle={GNUnet C Tutorial},
12 pdfauthor={Christian Grothoff <christian@grothoff.org>},
13 pdfkeywords={p2p,search,gnunet,tutorial}
14 %,pdfpagemode={FullScreen}
23 literate={*}{{\char42}}1
27 \newcommand{\exercise}[1]{\noindent\begin{boxedminipage}{\textwidth}{\bf Exercise:} #1 \end{boxedminipage}}
32 \large {A Tutorial for GNUnet 0.9.x (C version)}
34 Christian Grothoff $\qquad$ Bart Polot $\qquad$ Matthias Wachs
38 This tutorials explains how to install GNUnet on a GNU/Linux system ond gives an introduction how
39 GNUnet can be used to develop a Peer-to-Peer application. Detailed installation instructions for
40 various operating systems and a detailed list of all dependencies can found on our website at
41 \url{https://gnunet.org/installation}.
43 \textbf{Please read this tutorial carefully since every single step is important and do not hesitate to contact the GNUnet team if you have any questions or problems! Check here how to contact the GNUnet team:
44 \url{https://gnunet.org/contact_information}}
47 \section{Installing GNUnet}
48 First of all you have to install a current version of GNUnet. You can download a
49 tarball of a stable version from GNU FTP mirrors or obtain the latest development
50 version from our Subversion repository.
52 Most of the time you should prefer to download the stable version since with the
53 latest development version things can be broken, functionality can be changed or tests
54 can fail. You should only use the development version if you know that you require a
55 certain feature or a certain issue has been fixed since the last release.
57 \subsection{Obtaining a stable version}
58 You can download the latest stable version of GNUnet from GNU FTP mirrors:
60 \url{ftp://ftp.gnu.org/gnu/gnunet/gnunet-0.9.5a.tar.gz}
62 You should also download the signature file and verify the integrity of the tarball.
64 \url{ftp://ftp.gnu.org/gnu/gnunet/gnunet-0.9.5a.tar.gz.sig}
66 To verify the signature you should first import the GPG key used to sign the tarball
68 $ gpg --keyserver keys.gnupg.net --recv-keys 48426C7E
70 And use this key to verify the tarball's signature
72 $ gpg --verify gnunet-0.9.5a.tar.gz.sig gnunet-0.9.5a.tar.gz
74 After successfully verifying the integrity you can extract the tarball using
76 $ tar xvzf gnunet-0.9.5a.tar.gz
77 $ mv gnunet-0.9.5a gnunet # we will use the directory "gnunet" in the reminder of this document
81 \subsection{Installing Build Tool Chain and Dependencies}
82 To successfully compile GNUnet you need the tools to build GNUnet and the required dependencies.
83 Please have a look at \url{https://gnunet.org/dependencies} for a list of required dependencies
84 and \url{https://gnunet.org/generic_installation} for specific instructions for your operating system.
86 Please check the notes at the end of the configure process about required dependencies.
88 For GNUNet bootstrapping support and the http(s) plugin you should install \texttt{libcurl}.
89 For the filesharing service you should install at least one of the datastore backends \texttt{mysql},
90 \texttt{sqlite} or \texttt{postgresql}.
92 \subsection{Obtaining the latest version from Subversion}
93 The latest development version can obtained from our Subversion (\textit{svn}) repository. To obtain
94 the code you need Subversion installed and checkout the repository using:
95 \lstset{language=bash}
97 $ svn checkout https://gnunet.org/svn/gnunet
99 After cloning the repository you have to execute
100 \lstset{language=bash}
106 The remainder of this tutorial assumes that you have SVN HEAD checked out.
108 \subsection{Compiling and Installing GNUnet}
110 First, you need to install the latest {\tt
111 libgnupgerror}\footnote{\url{ftp://ftp.gnupg.org/gcrypt/libgpg-error/libgpg-error-1.11.tar.bz2}}
112 and {\tt libgcrypt} version from Git. The current GNUnet code uses
113 ECC functions not available in any released version of libgcrypt.
115 \lstset{language=bash}
117 $ git clone git://git.gnupg.org/libgcrypt.git
126 Assuming all dependencies are installed, the following commands will compile and install GNUnet in your
127 home directory. You can specify the directory where GNUnet will be installed by changing the \lstinline|--prefix| value when calling \lstinline|./configure|. If you do not specifiy a prefix, GNUnet is installed in the directory \lstinline|/usr/local|. When developing new applications you may want to enable
128 verbose logging by adding \lstinline|--enable-logging=verbose|:
130 \lstset{language=bash}
132 $ ./configure --prefix=$HOME --enable-logging
137 After installing GNUnet you have to set the \lstinline|GNUNET_PREFIX| environmental variable used by GNUnet to detect it's installation directory and add your GNUnet installation to your path environmental variable.
138 This configuration is only valid for the current shell session, so you should add \lstinline|export GNUNET_PREFIX=$HOME| to your \lstinline|.bash_rc| or \lstinline|.profile| to be sure the environment variable is always set. In addition you have to create the \lstinline|.gnunet| directory in your home directory where GNUnet stores it's data and an empty GNUnet configuration file:
140 \lstset{language=bash}
142 $ export GNUNET_PREFIX=$HOME
143 $ export PATH=$PATH:$GNUNET_PREFIX/bin
144 $ echo export GNUNET_PREFIX=$HOME >> ~/.bashrc
145 $ echo export PATH=$GNUNET_PREFIX/bin:$PATH >> ~/.bashrc
147 $ touch ~/.gnunet/gnunet.conf
151 \subsection{Common Issues - Check your GNUnet installation}
152 You should check your installation to ensure that installing GNUnet was successful up to this point. You should be able to access GNUnet's binaries and run GNUnet's self check.
156 should return \lstinline|$GNUNET_PREFIX/bin/gnunet-arm|. It should be located in your GNUnet installation and the output should not be empty. If you see an output like:
161 check your {\tt PATH} variable to ensure GNUnet's {\tt bin} directory is included.
163 GNUnet provides tests for all of it's subcomponents. Run
167 to execute tests for all components. {\tt make check} traverses all subdirectories in {\tt src}.
168 For every subdirectory you should get a message like this:
171 make[2]: Entering directory `/home/mwachs/gnunet/contrib'
172 PASS: test_gnunet_prefix
178 If you see a message like this:
181 Mar 12 16:57:56-642482 resolver-api-19449 ERROR Must specify `HOSTNAME' for `resolver' in configuration!
182 Mar 12 16:57:56-642573 test_program-19449 ERROR Assertion failed at resolver_api.c:204.
183 /bin/bash: line 5: 19449 Aborted (core dumped) ${dir}$tst
186 double check your {\tt GNUNET\_PREFIX} environmental variable and double check the steps performed in ~\ref{sub:install}
188 \section{Background: GNUnet Architecture}
189 GNUnet is organized in layers and services. Each service is composed of a
190 main service implementation and a client library for other programs to use
191 the service's functionality, described by an API. This approach is shown in
192 figure~\ref{fig:service}. Some services provide an additional command line
193 tool to enable the user to interact with the service.
195 Very often it is other GNUnet services that will use these APIs to build the
196 higher layers of GNUnet on top of the lower ones. Each layer expands or extends
197 the functionality of the service below (for instance, to build a mesh on top of
198 a DHT). See figure ~\ref{fig:interaction} for an illustration of this approach.
203 \begin{subfigure}[b]{0.3\textwidth}
205 \includegraphics[width=\textwidth]{figs/Service.pdf}
206 \caption{Service with API and network protocol}
210 \begin{subfigure}[b]{0.3\textwidth}
212 \includegraphics[width=\textwidth]{figs/System.pdf}
213 \caption{Service interaction}
214 \label{fig:interaction}
217 \caption{GNUnet's layered system architecture}
220 The main service implementation runs as a standalone process in the operating
221 system and the client code runs as part of the client program, so crashes of a
222 client do not affect the service process or other clients. The service and the
223 clients communicate via a message protocol to be defined and implemented by
226 \section{First Steps with GNUnet}
228 \subsection{Configure your peer}
229 First of all we need to configure your peer. Each peer is started with a configuration containing settings for GNUnet itself and it's services. This configuration is based on the default configuration shipped with GNUnet and can be modified. The default configuration is located in the {\tt \$GNUNET\_PREFIX/share/gnunet/config.d} directory. When starting a peer, you can specify a customized configuration using the the {\tt$-c$} command line switch when starting the ARM service and all other services. When using a modified configuration the default values are loaded and only values specified in the configuration file will replace the default values.
231 Since we want to start additional peers later, we need
232 some modifications from the default configuration. We need to create a separate service home and a file containing our modifications for this peer:
238 Now add the following lines to peer1.conf to use this directory. For simplified usage we want to prevent
239 the peer to connect to the GNUnet network since this could lead to confusing output. This modifications will replace the default settings:
242 $ SERVICEHOME = ~/gnunet1/ # Use this directory to store GNUnet data
244 $ SERVERS = # prevent bootstrapping
247 \subsection{Start a peer}
248 Each GNUnet instance (called peer) has an identity (\textit{peer ID}) based on a
249 cryptographic public private key pair. The peer ID is the printable hash of the
250 public key. So before starting the peer, you may want to just generate the peer's private
251 key using the command
252 \lstset{language=bash}
254 $ gnunet-peerinfo -c ~/peer1.conf -s
256 You should see an output containing the peer ID similar to:
257 \lstset{language=bash}
259 I am peer `0PA02UVRKQTS2C .. JL5Q78F6H0B1ACPV1CJI59MEQUMQCC5G'.
262 GNUnet services are controlled by a master service the so called \textit{Automatic Restart Manager} (ARM).
263 ARM starts, stops and even restarts services automatically or on demand when a client connects.
264 You interact with the ARM service using the \lstinline|gnunet-arm| tool.
265 GNUnet can then be started with \lstinline|gnunet-arm -s| and stopped with
266 \lstinline|gnunet-arm -e|. An additional service not automatically started
267 can be started using \lstinline|gnunet-arm -i <service name>| and stopped
268 using \lstinline|gnunet-arm -k <servicename>|.
270 \subsection{Monitor a peer}
271 In this section, we will monitor the behaviour of our peer's DHT service with respect to a
272 specific key. First we will start GNUnet and then start the DHT service and use the DHT monitor tool
273 to monitor the PUT and GET commands we issue ussing the \lstinline|gnunet-dht-put| and
274 \lstinline|gnunet-dht-get| command. Using the ``monitor'' line given below, you can observe the behavior of
275 your own peer's DHT with respect to the specified KEY:
277 \lstset{language=bash}
279 $ gnunet-arm -c ~/peer1.conf -s # start gnunet with all default services
280 $ gnunet-arm -c ~/peer1.conf -i dht # start DHT service
281 $ cd ~/gnunet/src/dht;
282 $ ./gnunet-dht-monitor -c ~/peer1.conf -k KEY
284 Now open a separate terminal and change again to the \lstinline|gnunet/src/dht| directory:
286 $ cd ~/gnunet/src/dht
287 $ ./gnunet-dht-put -c ~/peer1.conf -k KEY -d VALUE # put VALUE under KEY in the DHT
288 $ ./gnunet/src/dht/gnunet-dht-get ~/peer1.conf -k KEY # get key KEY from the DHT
289 $ gnunet-statistics -c ~/peer1.conf # print statistics about current GNUnet state
290 $ gnunet-statistics -c ~/peer1.conf -s dht # print statistics about DHT service
293 \subsection{Starting Two Peers by Hand}
294 \subsubsection{Setup a second peer}
295 We will now start a second peer on your machine.
296 For the second peer, you will need to manually create a modified
297 configuration file to avoid conflicts with ports and directories.
298 A peers configuration file is by default located in {\tt ~/.gnunet/gnunet.conf}.
299 This file is typically very short or even empty as only the differences to the
300 defaults need to be specified. The defaults are located in
301 many files in the {\tt \$GNUNET\_PREFIX/share/gnunet/config.d} directory.
303 To configure the second peer, use the files {\tt
304 \$GNUNET\_PREFIX/share/gnunet/config.d} as a template for your main
307 \lstset{language=bash}
309 $ cat $GNUNET_PREFIX/share/gnunet/config.d/*.conf > peer2.conf
311 Now you have to edit {\tt peer2.conf} and change:
314 \item{\texttt{SERVICEHOME} under \texttt{PATHS}}
315 \item{Every (uncommented) value for ``\texttt{PORT}'' (add 10000) in any
316 section (the option may be commented out if \texttt{PORT} is
317 prefixed by "\#", in this case, UNIX domain sockets are used
318 and the PORT option does not need to be touched) }
319 \item{Every value for ``\texttt{UNIXPATH}'' in any section (e.g. by adding a "-p2" suffix)}
321 to a fresh, unique value. Make sure that the \texttt{PORT} numbers stay
322 below 65536. From now on, whenever you interact with the second
323 peer, you need to specify {\tt -c peer2.conf} as an additional
324 command line argument.
326 Now, generate the 2nd peer's private key:
328 \lstset{language=bash}
330 $ gnunet-peerinfo -s -c peer2.conf
334 This may take a while, generate entropy using your keyboard or mouse
335 as needed. Also, make sure the output is different from the {\tt
336 gnunet-peerinfo} output for the first peer (otherwise you made an
337 error in the configuration).
339 \subsubsection{Start the second peer and connect the peers}
340 Then, you can start a second peer using:
341 \lstset{language=bash}
343 $ gnunet-arm -c peer2.conf -s
344 $ gnunet-arm -c peer2.conf -i dht
345 $ ~/gnunet/src/dht/gnunet-dht-put -c peer2.conf -k KEY -d VALUE
346 $ ~/gnunet/src/dht/gnunet-dht-get -c peer2.conf -k KEY
348 If you want the two peers to connect, you have multiple options:
351 \item UDP neighbour discovery (automatic)
352 \item Setup a bootstrap server
353 \item Connect manually
355 To setup peer 1 as bootstrapping server change the configuration of the first one to be a hostlist server by adding the following lines to \texttt{peer1.conf} to enable bootstrapping server:
361 Then change {\tt peer2.conf} and replace the ``\texttt{SERVERS}'' line in the ``\texttt{[hostlist]}'' section with
362 ``\texttt{http://localhost:8080/}''. Restart both peers using:
364 $ gnunet-arm -c peer1.conf -e # stop first peer
365 $ gnunet-arm -c peer1.conf -s # start first peer
366 $ gnunet-arm -c peer2.conf -s # start second peer
369 Note that if you start your peers without changing these settings, they
370 will use the ``global'' hostlist servers of the GNUnet P2P network and
371 likely connect to those peers. At that point, debugging might become
372 tricky as you're going to be connected to many more peers and would
373 likely observe traffic and behaviors that are not explicitly controlled
376 \subsubsection{How to connect manually}
377 If you want to use the \texttt{peerinfo} tool to connect your peers, you should:
380 \item{Remove {\tt hostlist} from {\tt DEFAULTSERVICES} (to not connect to the global GNUnet)}
381 \item{Start both peers running {\tt gnunet-arm -c peer1.conf -s} and {\tt gnunet-arm -c peer2.conf -s}}
382 \item{Get \texttt{HELLO} message of the first peer running {\tt gnunet-peerinfo -c peer1.conf -g}}
383 \item{Give the output to the second peer by running {\tt gnunet-peerinfo -c peer2.conf -p '<output>'}}
386 Check that they are connected using {\tt gnunet-core -c peer1.conf}, which should give you the other peer's
389 $ gnunet-core -c peer1.conf
390 Peer `9TVUCS8P5A7ILLBGO6JSTSSN2B44H3D2MUIFJMLKAITC0I22UVFBFP1H8NRK2IA35VKAK16LLO0MFS7TAQ9M1KNBJ4NGCHP3JPVULDG'
393 \subsection{Starting Peers Using the Testbed Service}
395 GNUnet's testbed service is used for testing scenarios where a number of peers
396 are to be started. The testbed can manage peers on a single host or on multiple
397 hosts in a distributed fashion. On a single affordable computer, it should be
398 possible to run around tens of peers without drastically increasing the load on the
401 The testbed service can be access through its API
402 \texttt{include/gnunet\_testbed\_service.h}. The API provides many routines for
403 managing a group of peers. It also provides a helper function
404 \texttt{GNUNET\_TESTBED\_test\_run()} to quickly setup a minimalistic testing
405 environment on a single host.
407 This function takes a configuration file which will be used as a template
408 configuration for the peers. The testbed takes care of modifying relevant
409 options in the peers' configuration such as SERVICEHOME, PORT, UNIXPATH to
410 unique values so that peers run without running into conflicts. It also checks
411 and assigns the ports in configurations only if they are free.
413 Additionally, the testbed service also reads its options from the same
414 configuration file. Various available options and details about them can be
415 found in the testbed default configuration file \texttt{src/testbed/testbed.conf}.
417 With the testbed API, a sample test case can be structured as follows:
418 \lstinputlisting[language=C]{testbed_test.c}
419 The source code for the above listing can be found at
420 \url{https://gnunet.org/svn/gnunet/doc/testbed_test.c}. After installing GNUnet, the above source code can be compiled as:
421 \lstset{language=bash}
423 $ export CPPFLAGS="-I/path/to/gnunet/headers"
424 $ export LDFLAGS="-L/path/to/gnunet/libraries"
425 $ gcc $CPPFLAGS $LDFLAGS -o testbed-test testbed_test.c -lgnunettestbed -lgnunetdht -lgnunetutil
427 The \texttt{CPPFLAGS} and \texttt{LDFLAGS} are necessary if GNUnet is installed
428 into a different directory other than \texttt{/usr/local}.
430 All of testbed API's peer management functions treat management actions as
431 operations and return operation handles. It is expected that the operations
432 begin immediately, but they may get delayed (to balance out load on the system).
433 The program using the API then has to take care of marking the operation as
434 ``done'' so that its associated resources can be freed immediately and other
435 waiting operations can be executed. Operations will be canceled if they are
436 marked as ``done'' before their completion.
438 An operation is treated as completed when it succeeds or fails. Completion of
439 an operation is either conveyed as events through \textit{controller event
440 callback} or through respective operation completion callbacks. In functions
441 which support completion notification through both controller event callback and
442 operation completion callback, first the controller event callback will be
443 called. If the operation is not marked as done in that callback or if the
444 callback is given as NULL when creating the operation, the operation completion
445 callback will be called. The API documentation shows which event are to be
446 expected in the controller event notifications. It also documents any
447 exceptional behaviour.
449 Once the peers are started, test cases often need to connect some of the peers'
450 services. Normally, opening a connect to a peer's service requires the peer's
451 configuration. While using testbed, the testbed automatically generates
452 per-peer configuration. Accessing those configurations directly through file
453 system is discouraged as their locations are dynamically created and will be
454 different among various runs of testbed. To make access to these configurations
455 easy, testbed API provides the function
456 \texttt{GNUNET\_TESTBED\_service\_connect()}. This function fetches the
457 configuration of a given peer and calls the \textit{Connect Adapter}.
458 In the example code, it is the \texttt{dht\_ca}. A connect adapter is expected
459 to open the connection to the needed service by using the provided configuration
460 and return the created service connection handle. Successful connection to the
461 needed service is signaled through \texttt{service\_connect\_comp\_cb}.
463 A dual to connect adapter is the \textit{Disconnect Adapter}. This callback is
464 called after the connect adapter has been called when the operation from
465 \texttt{GNUNET\_TESTBED\_service\_connect()} is marked as ``done''. It has to
466 disconnect from the service with the provided service handle (\texttt{op\_result}).
468 \exercise{Find out how many peers you can run on your system.}
470 \exercise{Find out how to create a 2D torus topology by changing the
471 options in the configuration file.\footnote{See \url{https://gnunet.org/content/supported-topologies}}
472 Then use the DHT API to store and retrieve values in the
475 \section{Developing Applications}
476 \subsection{gnunet-ext}
477 To develop a new peer-to-peer application or to extend GNUnet we provide
478 a template build system for writing GNUnet extensions in C. It can be
481 \lstset{language=bash}
483 $ svn checkout https://gnunet.org/svn/gnunet-ext/
486 $ ./configure --prefix=$HOME --with-gnunet=$GNUNET_PREFIX
493 The GNUnet ext template includes examples and a working buildsystem for a new GNUnet service.
494 A common GNUnet service consists of the following parts which will be discussed in detail in the
495 remainder of this document. The functionality of a GNUnet service is implemented in:
499 \item the GNUnet service (\lstinline|gnunet-ext/src/ext/gnunet-service-ext.c|)
500 \item the client API (\lstinline|gnunet-ext/src/ext/ext_api.c|)
501 \item the client application using the service API (\lstinline|gnunet-ext/src/ext/gnunet-ext.c|)
506 The interfaces for these entities are defined in:
509 \item client API interface (\lstinline|gnunet-ext/src/ext/ext.h|)
510 \item the service interface (\lstinline|gnunet-ext/src/include/gnunet_service_SERVICE.h|)
511 \item the P2P protocol (\lstinline|gnunet-ext/src/include/gnunet_protocols_ext.h|)
515 In addition the \texttt{ext} systems provides:
518 \item a test testing the API (\lstinline|gnunet-ext/src/ext/test_ext_api.c|)
519 \item a configuration template for the service (\lstinline|gnunet-ext/src/ext/ext.conf.in|)
523 \subsection{Adapting the Template}
525 The first step for writing any extension with a new service is to
526 ensure that the {\tt ext.conf.in} file contains entries for the
527 \texttt{UNIXPATH}, \texttt{PORT} and \texttt{BINARY} for the service in a section named after
530 If you want to adapt the template rename the {\tt ext.conf.in} to match your
531 services name, you have to modify the \texttt{AC\_OUTPUT} section in {\tt configure.ac}
532 in the \texttt{gnunet-ext} root.
534 \section{Writing a Client Application}
536 When writing any client application (for example, a command-line
537 tool), the basic structure is to start with the {\tt
538 GNUNET\_PROGRAM\_run} function. This function will parse
539 command-line options, setup the scheduler and then invoke the {\tt
540 run} function (with the remaining non-option arguments) and a handle
541 to the parsed configuration (and the configuration file name that was
542 used, which is typically not needed):
546 #include <gnunet/platform.h>
547 #include <gnunet/gnunet_util_lib.h>
555 const struct GNUNET_CONFIGURATION_Handle *cfg)
562 main (int argc, char *const *argv)
564 static const struct GNUNET_GETOPT_CommandLineOption options[] = {
565 GNUNET_GETOPT_OPTION_END
568 GNUNET_PROGRAM_run (argc,
571 gettext_noop ("binary description text"),
572 options, &run, NULL)) ? ret : 1;
576 \subsection{Handling command-line options}
578 Options can then be added easily by adding global variables and
579 expanding the {\tt options} array. For example, the following would
580 add a string-option and a binary flag (defaulting to {\tt NULL} and
581 {\tt GNUNET\_NO} respectively):
584 static char *string_option;
588 static const struct GNUNET_GETOPT_CommandLineOption options[] = {
589 {'s', "name", "SOMESTRING",
590 gettext_noop ("text describing the string_option NAME"), 1,
591 &GNUNET_GETOPT_set_string, &string_option},
593 gettext_noop ("text describing the flag option"), 0,
594 &GNUNET_GETOPT_set_one, &a_flag},
595 GNUNET_GETOPT_OPTION_END
597 string_option = NULL;
598 a_flag = GNUNET_SYSERR;
602 Issues such as displaying some helpful text describing options using
603 the {\tt --help} argument and error handling are taken care of when
604 using this approach. Other {\tt GNUNET\_GETOPT\_}-functions can be used
605 to obtain integer value options, increment counters, etc. You can
606 even write custom option parsers for special circumstances not covered
607 by the available handlers. To check if an argument was specified by the
608 user you initialize the variable with a specific value (e.g. NULL for
609 a string and GNUNET\_SYSERR for a integer) and check after parsing
610 happened if the values were modified.
612 Inside the {\tt run} method, the program would perform the
613 application-specific logic, which typically involves initializing and
614 using some client library to interact with the service. The client
615 library is supposed to implement the IPC whereas the service provides
616 more persistent P2P functions.
618 \exercise{Add a few command-line options and print them inside
619 of {\tt run}. What happens if the user gives invalid arguments?}
621 \subsection{Writing a Client Library}
623 The first and most important step in writing a client library is to
624 decide on an API for the library. Typical API calls include
625 connecting to the service, performing application-specific requests
626 and cleaning up. Many examples for such service APIs can be found
627 in the {\tt gnunet/src/include/gnunet\_*\_service.h} files.
629 Then, a client-service protocol needs to be designed. This typically
630 involves defining various message formats in a header that will be
631 included by both the service and the client library (but is otherwise
632 not shared and hence located within the service's directory and not
633 installed by {\tt make install}). Each message must start with a {\tt
634 struct GNUNET\_MessageHeader} and must be shorter than 64k. By
635 convention, all fields in IPC (and P2P) messages must be in big-endian
636 format (and thus should be read using {\tt ntohl} and similar
637 functions and written using {\tt htonl} and similar functions).
638 Unique message types must be defined for each message struct in the
639 {\tt gnunet\_protocols.h} header (or an extension-specific include
642 \subsubsection{Connecting to the Service}
644 Before a client library can implement the application-specific protocol
645 with the service, a connection must be created:
649 struct GNUNET_CLIENT_Connection *client;
650 client = GNUNET_CLIENT_connect ("service-name", cfg);
653 As a result a {\tt GNUNET\_CLIENT\_Connection} handle is returned
654 which has to used in later API calls related to this service.
655 The complete client API can be found in {\tt gnunet\_client\_lib.h}
657 \subsubsection{GNUnet Messages}
659 In GNUnet, messages are always sent beginning with a {\tt struct GNUNET\_MessageHeader}
660 in big endian format. This header defines the size and the type of the
661 message, the payload follows after this header.
665 struct GNUNET_MessageHeader
669 * The length of the struct (in bytes, including the length field itself),
670 * in big-endian format.
672 uint16_t size GNUNET_PACKED;
675 * The type of the message (GNUNET_MESSAGE_TYPE_XXXX), in big-endian format.
677 uint16_t type GNUNET_PACKED;
682 Existing message types are defined in {\tt gnunet\_protocols.h}\\
683 A common way to create a message is:
687 struct GNUNET_MessageHeader *msg =
688 GNUNET_malloc(payload_size + sizeof(struct GNUNET_MessageHeader));
689 msg->size = htons(payload_size + sizeof(struct GNUNET_MessageHeader));
690 msg->type = htons(GNUNET_MY_MESSAGE_TYPE);
691 memcpy(&msg[1], &payload, payload_size);
695 \exercise{Define a message struct that includes a 32-bit
696 unsigned integer in addition to the standard GNUnet MessageHeader.
697 Add a C struct and define a fresh protocol number for your message.}
700 \subsubsection{Sending Requests to the Service}
702 Any client-service protocol must start with the client sending the
703 first message to the service, since services are only notified about
704 (new) clients upon receiving a the first message.
706 Clients can transmit messages to the service using the
707 {\tt GNUNET\_CLIENT\_notify\_transmit\_ready} API:
711 transmit_cb (void *cls, size_t size, void *buf)
714 if (NULL == buf) { /* handle error here */; return 0; }
715 GNUNET_assert (size >= msg_size);
716 memcpy (buf, my_msg, msg_size);
722 th = GNUNET_CLIENT_notify_transmit_ready (client,
730 The client-service protocoll calls {\tt GNUNET\_CLIENT\_notify\_transmit\_ready}
731 to be notified when the client is ready to send data to the service.
732 Besides other arguments, you have to pass the client returned
733 from the {\tt connect} call, the message size and the callback function to
734 call when the client is ready to send.
736 Only a single transmission request can be queued per client at the
737 same time using this API. The handle {\tt th} can be used to cancel
738 the request if necessary (for example, during shutdown).
740 When {\tt transmit\_cb} is called the message is copied in the buffer provided and
741 the number of bytes copied into the buffer is returned. {\tt transmit\_cb}
742 could also return 0 if for some reason no message
743 could be constructed; this is not an error and the connection to the
744 service will persist in this case.
746 \exercise{Define a helper function to transmit a 32-bit
747 unsigned integer (as payload) to a service using some given client
751 \subsubsection{Receiving Replies from the Service}
753 Clients can receive messages from the service using the
754 {\tt GNUNET\_CLIENT\_receive} API:
759 * Function called with messages from stats service.
762 * @param msg message received, NULL on timeout or fatal error
765 receive_message (void *cls, const struct GNUNET_MessageHeader *msg)
767 struct MyArg *arg = cls;
773 GNUNET_CLIENT_receive (client,
780 It should be noted that this receive call only receives a single
781 message. To receive additional messages, {\tt
782 GNUNET\_CLIENT\_receive} must be called again.
784 \exercise{Expand your helper function to receive a
785 response message (for example, containing just the GNUnet MessageHeader
786 without any payload). Upon receiving the service's response, you should
787 call a callback provided to your helper function's API. You'll need to
788 define a new 'struct' to hold your local context (``closure'').}
791 \subsection{Writing a user interface}
793 Given a client library, all it takes to access a service now is to
794 combine calls to the client library with parsing command-line
797 \exercise{Call your client API from your {\tt run} method
798 in your client application to send a request to the service.
799 For example, send a 32-bit integer value based on a number given
800 at the command-line to the service.}
804 \section{Writing a Service}
806 Before you can test the client you've written so far, you'll need to also
807 implement the corresponding service.
810 \subsection{Code Placement}
812 New services are placed in their own subdirectory under {\tt gnunet/src}.
813 This subdirectory should contain the API implementation file {\tt SERVICE\_api.c},
814 the description of the client-service protocol {\tt SERVICE.h} and P2P protocol
815 {\tt SERVICE\_protocol.h}, the implementation of the service itself
816 {\tt gnunet-service-SERVICE.h} and several files for tests, including test code
817 and configuration files.
819 \subsection{Starting a Service}
821 The key API definitions for starting services are:
824 typedef void (*GNUNET_SERVICE_Main) (void *cls,
825 struct GNUNET_SERVER_Handle *server,
826 const struct GNUNET_CONFIGURATION_Handle *cfg);
827 int GNUNET_SERVICE_run (int argc,
829 const char *serviceName,
830 enum GNUNET_SERVICE_Options opt,
831 GNUNET_SERVICE_Main task,
835 Here is a starting point for your main function for your service:
839 static void my_main (void *cls,
840 struct GNUNET_SERVER_Handle *server,
841 const struct GNUNET_CONFIGURATION_Handle *cfg)
846 int main (int argc, char *const*argv)
849 GNUNET_SERVICE_run (argc, argv, "my",
850 GNUNET_SERVICE_OPTION_NONE,
857 \exercise{Write a stub service that processes no messages at all
858 in your code. Create a default configuration for it, integrate it
859 with the build system and start the service from {\tt
860 gnunet-service-arm} using {\tt gnunet-arm -i NAME}.}
863 \subsection{Receiving Requests from Clients}
865 Inside of the {\tt my\_main} method, a service typically registers for
866 the various message types from clients that it supports by providing
867 a handler function, the message type itself and possibly a fixed
868 message size (or 0 for variable-size messages):
873 handle_set (void *cls,
874 struct GNUNET_SERVER_Client *client,
875 const struct GNUNET_MessageHeader *message)
877 GNUNET_SERVER_receive_done (client, GNUNET_OK);
880 handle_get (void *cls,
881 struct GNUNET_SERVER_Client *client,
882 const struct GNUNET_MessageHeader *message)
884 GNUNET_SERVER_receive_done (client, GNUNET_OK);
887 static void my_main (void *cls,
888 struct GNUNET_SERVER_Handle *server,
889 const struct GNUNET_CONFIGURATION_Handle *cfg)
891 static const struct GNUNET_SERVER_MessageHandler handlers[] = {
892 {&handle_set, NULL, GNUNET_MESSAGE_TYPE_MYNAME_SET, 0},
893 {&handle_get, NULL, GNUNET_MESSAGE_TYPE_MYNAME_GET, 0},
896 GNUNET_SERVER_add_handlers (server, handlers);
897 /* do more setup work */
901 Each handler function {\bf must} eventually (possibly in some
902 asynchronous continuation) call {\tt GNUNET\_SERVER\_receive\_done}.
903 Only after this call additional messages from the same client may
904 be processed. This way, the service can throttle processing messages
905 from the same client. By passing {\tt GNUNET\_SYSERR}, the service
906 can close the connection to the client, indicating an error.
908 Services must check that client requests are well-formed and must not
909 crash on protocol violations by the clients. Similarly, client
910 libraries must check replies from servers and should gracefully report
911 errors via their API.
914 \exercise{Change the service to ``handle'' the message from your
915 client (for now, by printing a message). What happens if you
916 forget to call {\tt GNUNET\_SERVER\_receive\_done}?}
919 \subsection{Responding to Clients}
921 Servers can send messages to clients using the
922 {\tt GNUNET\_SERVER\_notify\_transmit\_ready} API:
927 transmit_cb (void *cls, size_t size, void *buf)
930 if (NULL == buf) { handle_error(); return 0; }
931 GNUNET_assert (size >= msg_size);
932 memcpy (buf, my_msg, msg_size);
938 struct GNUNET_SERVER_TransmitHandle *th;
939 th = GNUNET_SERVER_notify_transmit_ready (client,
946 Only a single transmission request can be queued per client
947 at the same time using this API.
948 Additional APIs for sending messages to clients can be found
949 in the {\tt gnunet\_server\_lib.h} header.
952 \exercise{Change the service respond to the request from your
953 client. Make sure you handle malformed messages in both directions.}
956 \section{Interacting directly with other Peers using the CORE Service}
958 One of the most important services in GNUnet is the \texttt{CORE} service
959 managing connections between peers and handling encryption between peers.
961 One of the first things any service that extends the P2P protocol typically does
962 is connect to the \texttt{CORE} service using:
966 #include <gnunet/gnunet_core_service.h>
968 struct GNUNET_CORE_Handle *
969 GNUNET_CORE_connect (const struct GNUNET_CONFIGURATION_Handle *cfg,
971 GNUNET_CORE_StartupCallback init,
972 GNUNET_CORE_ConnectEventHandler connects,
973 GNUNET_CORE_DisconnectEventHandler disconnects,
974 GNUNET_CORE_MessageCallback inbound_notify,
975 int inbound_hdr_only,
976 GNUNET_CORE_MessageCallback outbound_notify,
977 int outbound_hdr_only,
978 const struct GNUNET_CORE_MessageHandler *handlers);
981 \subsection{New P2P connections}
983 Before any traffic with a different peer can be exchanged, the peer must be
984 known to the service. This is notified by the \texttt{CORE} {\tt connects} callback,
985 which communicates the identity of the new peer to the service:
991 const struct GNUNET_PeerIdentity * peer)
993 /* Save identity for later use */
994 /* Optional: start sending messages to peer */
998 \exercise{Create a service that connects to the \texttt{CORE}. Then
999 start (and connect) two peers and print a message once your connect
1000 callback is invoked.}
1002 \subsection{Receiving P2P Messages}
1004 To receive messages from \texttt{CORE}, services register a set of handlers
1005 (parameter {\tt *handlers} in the \lstinline|GNUNET_CORE_connect| call that are called by \texttt{CORE}
1006 when a suitable message arrives.
1011 callback_function_for_type_one(void *cls,
1012 const struct GNUNET_PeerIdentity *peer,
1013 const struct GNUNET_MessageHeader *message)
1016 return GNUNET_OK; /* or GNUNET_SYSERR to close the connection */
1020 * Functions to handle messages from core
1022 static struct GNUNET_CORE_MessageHandler core_handlers[] = {
1023 {&callback_function_for_type_one, GNUNET_MESSAGE_TYPE_MYSERVICE_TYPE_ONE, 0},
1029 \exercise{Start one peer with a new service that has a message
1030 handler and start a second peer that only has your ``old'' service
1031 without message handlers. Which ``connect'' handlers are invoked when
1032 the two peers are connected? Why?}
1035 \subsection{Sending P2P Messages}
1037 In response to events (connect, disconnect, inbound messages,
1038 timing, etc.) services can then use this API to transmit messages:
1043 (*GNUNET_CONNECTION_TransmitReadyNotify) (void *cls,
1047 /* Fill "*buf" with up to "size" bytes, must start with GNUNET_MessageHeader */
1048 return n; /* Total size of the message put in "*buf" */
1051 struct GNUNET_CORE_TransmitHandle *
1052 GNUNET_CORE_notify_transmit_ready (struct GNUNET_CORE_Handle *handle,
1053 int cork, uint32_t priority,
1054 struct GNUNET_TIME_Relative maxdelay,
1055 const struct GNUNET_PeerIdentity *target,
1057 GNUNET_CONNECTION_TransmitReadyNotify notify,
1061 \exercise{Write a service that upon connect sends messages as
1062 fast as possible to the other peer (the other peer should run a
1063 service that ``processes'' those messages). How fast is the
1064 transmission? Count using the STATISTICS service on both ends. Are
1065 messages lost? How can you transmit messages faster? What happens if
1066 you stop the peer that is receiving your messages?}
1069 \subsection{End of P2P connections}
1071 If a message handler returns {\tt GNUNET\_SYSERR}, the remote peer shuts down or
1072 there is an unrecoverable network disconnection, CORE notifies the service that
1073 the peer disconnected. After this notification no more messages will be received
1074 from the peer and the service is no longer allowed to send messages to the peer.
1075 The disconnect callback looks like the following:
1080 disconnects (void *cls,
1081 const struct GNUNET_PeerIdentity * peer)
1083 /* Remove peer's identity from known peers */
1084 /* Make sure no messages are sent to peer from now on */
1088 \exercise{Fix your service to handle peer disconnects.}
1090 \section{Using the DHT}
1091 The DHT allows to store data so other peers in the P2P network can
1092 access it and retrieve data stored by any peers in the network.
1093 This section will explain how to use the DHT. Of course, the first
1094 thing to do is to connect to the DHT service:
1097 dht_handle = GNUNET_DHT_connect (cfg, parallel_requests);
1099 The second parameter indicates how many requests in parallel to expect.
1100 It is not a hard limit, but a good approximation will make the DHT more
1103 \subsection{Storing data in the DHT}
1104 Since the DHT is a dynamic environment (peers join a leave frequently)
1105 the data that we put in the DHT does not stay there indefinitely. It is
1106 important to ``refresh'' the data periodically by simply storing it again,
1107 in order to make sure other peers can access it.
1109 The put API call offers a callback to signal that the PUT request has been
1110 sent. This does not guarantee that the data is accessible to others peers,
1111 or even that is has been stored, only that the service has requested to
1112 a neighboring peer the retransmission of the PUT request towards its final
1113 destination. Currently there is no feedback about whether or not the data
1114 has been sucessfully stored or where it has been stored. In order to improve
1115 the availablilty of the data and to compensate for possible errors, peers leaving
1116 and other unfavorable events, just make several PUT requests!
1121 message_sent_cont (void *cls, const struct GNUNET_SCHEDULER_TaskContext *tc)
1123 /* Request has left local node */
1126 struct GNUNET_DHT_PutHandle *
1127 GNUNET_DHT_put (struct GNUNET_DHT_Handle *handle,
1128 const struct GNUNET_HashCode * key,
1129 uint32_t desired_replication_level,
1130 enum GNUNET_DHT_RouteOption options, /* Route options, see next call */
1131 enum GNUNET_BLOCK_Type type, size_t size, const void *data,
1132 struct GNUNET_TIME_Absolute exp, /* When does the data expire? */
1133 struct GNUNET_TIME_Relative timeout, /* How long to try to send the request */
1134 GNUNET_DHT_PutContinuation cont,
1138 \exercise{Store a value in the DHT periodically to make sure it is available
1139 over time. You might consider using the function GNUNET\_SCHEDULER\_add\_delayed and
1140 call GNUNET\_DHT\_put from inside a helper function.}
1143 \subsection{Obtaining data from the DHT}
1144 As we saw in the previous example, the DHT works in an asynchronous mode.
1145 Each request to the DHT is executed ``in the background'' and the API
1146 calls return immediately. In order to receive results from the DHT, the
1147 API provides a callback. Once started, the request runs in the service,
1148 the service will try to get as many results as possible (filtering out
1149 duplicates) until the timeout expires or we explicitly stop the request.
1150 It is possible to give a ``forever'' timeout with
1151 {\tt GNUNET\_TIME\_UNIT\_FOREVER\_REL}.
1153 If we give a route option {\tt GNUNET\_DHT\_RO\_RECORD\_ROUTE} the callback
1154 will get a list of all the peers the data has travelled, both on the PUT
1155 path and on the GET path.
1159 get_result_iterator (void *cls, struct GNUNET_TIME_Absolute expiration,
1160 const struct GNUNET_HashCode * key,
1161 const struct GNUNET_PeerIdentity *get_path,
1162 unsigned int get_path_length,
1163 const struct GNUNET_PeerIdentity *put_path,
1164 unsigned int put_path_length,
1165 enum GNUNET_BLOCK_Type type, size_t size, const void *data)
1167 /* Do stuff with the data and/or route */
1169 GNUNET_DHT_get_stop (get_handle);
1173 GNUNET_DHT_get_start (dht_handle,
1177 GNUNET_DHT_RO_NONE, /* Route options */
1178 NULL, /* xquery: not used here */
1179 0, /* xquery size */
1180 &get_result_iterator,
1184 \exercise{Store a value in the DHT and after a while retrieve it. Show the IDs of all
1185 the peers the requests have gone through. In order to convert a peer ID to a string, use
1186 the function GNUNET\_i2s. Pay attention to the route option parameters in both calls!}
1188 \subsection{Implementing a block plugin}
1190 In order to store data in the DHT, it is necessary to provide a block
1191 plugin. The DHT uses the block plugin to ensure that only well-formed
1192 requests and replies are transmitted over the network.
1194 The block plugin should be put in a file {\tt
1195 plugin\_block\_SERVICE.c} in the service's respective directory. The
1196 mandatory functions that need to be implemented for a block plugin are
1197 described in the following sections.
1199 \subsubsection{Validating requests and replies}
1201 The evaluate function should validate a reply or a request. It returns
1202 a {\tt GNUNET\_BLOCK\_EvaluationResult}, which is an enumeration. All
1203 possible answers are in {\tt gnunet\_block\_lib.h}. The function will
1204 be called with a {\tt reply\_block} argument of {\tt NULL} for
1205 requests. Note that depending on how {\tt evaluate} is called, only
1206 some of the possible return values are valid. The specific meaning of
1207 the {\tt xquery} argument is application-specific. Applications that
1208 do not use an extended query should check that the {\tt xquery\_size}
1209 is zero. The Bloom filter is typically used to filter duplicate
1214 static enum GNUNET_BLOCK_EvaluationResult
1215 block_plugin_SERVICE_evaluate (void *cls,
1216 enum GNUNET_BLOCK_Type type,
1217 const GNUNET_HashCode * query,
1218 struct GNUNET_CONTAINER_BloomFilter **bf,
1222 const void *reply_block,
1223 size_t reply_block_size)
1225 /* Verify type, block and bloomfilter */
1229 Note that it is mandatory to detect duplicate replies in this
1230 function and return the respective status code. Duplicate
1231 detection should be done by setting the respective bits in
1232 the Bloom filter {\tt bf}. Failure to do so may cause replies
1233 to circle in the network.
1235 \subsubsection{Deriving a key from a reply}
1237 The DHT can operate more efficiently if it is possible to derive a key
1238 from the value of the corresponding block. The {\tt get\_key}
1239 function is used to obtain the key of a block --- for example, by
1240 means of hashing. If deriving the key is not possible, the function
1241 should simply return {\tt GNUNET\_SYSERR} (the DHT will still work
1242 just fine with such blocks).
1247 block_plugin_SERVICE_get_key (void *cls, enum GNUNET_BLOCK_Type type,
1248 const void *block, size_t block_size,
1249 GNUNET_HashCode * key)
1251 /* Store the key in the key argument, return GNUNET_OK on success. */
1255 \subsubsection{Initialization of the plugin}
1257 The plugin is realized as a shared C library. The library must export
1258 an initialization function which should initialize the plugin. The
1259 initialization function specifies what block types the plugin cares
1260 about and returns a struct with the functions that are to be used for
1261 validation and obtaining keys (the ones just defined above).
1266 libgnunet_plugin_block_SERVICE_init (void *cls)
1268 static enum GNUNET_BLOCK_Type types[] =
1270 GNUNET_BLOCK_TYPE_SERVICE_BLOCKYPE, /* list of blocks we care about, from gnunet_block_lib.h */
1271 GNUNET_BLOCK_TYPE_ANY /* end of list */
1273 struct GNUNET_BLOCK_PluginFunctions *api;
1275 api = GNUNET_malloc (sizeof (struct GNUNET_BLOCK_PluginFunctions));
1276 api->evaluate = &block_plugin_SERICE_evaluate;
1277 api->get_key = &block_plugin_SERVICE_get_key;
1283 \subsubsection{Shutdown of the plugin}
1285 Following GNUnet's general plugin API concept, the plugin must
1286 export a second function for cleaning up. It usually does very
1292 libgnunet_plugin_block_SERVICE_done (void *cls)
1294 struct GNUNET_TRANSPORT_PluginFunctions *api = cls;
1302 \subsubsection{Integration of the plugin with the build system}
1304 In order to compile the plugin, the {\tt Makefile.am} file for the
1305 service should contain a rule similar to this:
1307 \lstset{language=make}
1309 plugin_LTLIBRARIES = \
1310 libgnunet_plugin_block_SERVICE.la
1311 libgnunet_plugin_block_SERVICE_la_SOURCES = \
1312 plugin_block_SERVICE.c
1313 libgnunet_plugin_block_SERVICE_la_LIBADD = \
1314 $(top_builddir)/src/hello/libgnunethello.la \
1315 $(top_builddir)/src/block/libgnunetblock.la \
1316 $(top_builddir)/src/util/libgnunetutil.la
1317 libgnunet_plugin_block_SERVICE_la_LDFLAGS = \
1318 $(GN_PLUGIN_LDFLAGS)
1319 libgnunet_plugin_block_SERVICE_la_DEPENDENCIES = \
1320 $(top_builddir)/src/block/libgnunetblock.la
1325 \exercise{Write a block plugin that accepts all queries
1326 and all replies but prints information about queries and replies
1327 when the respective validation hooks are called.}
1331 \subsection{Monitoring the DHT}
1332 It is possible to monitor the functioning of the local DHT service. When monitoring
1333 the DHT, the service will alert the monitoring program of any events,
1334 both started locally or received for routing from another peer. The are three different
1335 types of events possible: a GET request, a PUT request or a response (a reply to
1338 Since the different events have different associated data, the API gets 3
1339 different callbacks (one for each message type) and optional type and key parameters,
1340 to allow for filtering of messages. When an event happens, the appropiate callback
1341 is called with all the information about the event.
1345 get_callback (void *cls,
1346 enum GNUNET_DHT_RouteOption options,
1347 enum GNUNET_BLOCK_Type type,
1349 uint32_t desired_replication_level,
1350 unsigned int path_length,
1351 const struct GNUNET_PeerIdentity *path,
1352 const struct GNUNET_HashCode * key)
1357 get_resp_callback (void *cls,
1358 enum GNUNET_BLOCK_Type type,
1359 const struct GNUNET_PeerIdentity *get_path,
1360 unsigned int get_path_length,
1361 const struct GNUNET_PeerIdentity *put_path,
1362 unsigned int put_path_length,
1363 struct GNUNET_TIME_Absolute exp,
1364 const struct GNUNET_HashCode * key,
1371 put_callback (void *cls,
1372 enum GNUNET_DHT_RouteOption options,
1373 enum GNUNET_BLOCK_Type type,
1375 uint32_t desired_replication_level,
1376 unsigned int path_length,
1377 const struct GNUNET_PeerIdentity *path,
1378 struct GNUNET_TIME_Absolute exp,
1379 const struct GNUNET_HashCode * key,
1385 monitor_handle = GNUNET_DHT_monitor_start (dht_handle,
1386 block_type, /* GNUNET_BLOCK_TYPE_ANY for all */
1387 key, /* NULL for all */
1395 \section{Debugging with {\tt gnunet-arm}}
1397 Even if services are managed by {\tt gnunet-arm}, you can start them with
1398 {\tt gdb} or {\tt valgrind}. For example, you could add the following lines
1399 to your configuration file to start the DHT service in a {\tt gdb} session in a
1404 PREFIX=xterm -e gdb --args
1407 Alternatively, you can stop a service that was started via ARM and run it manually:
1409 \lstset{language=bash}
1412 $ gdb --args gnunet-service-dht -L DEBUG
1413 $ valgrind gnunet-service-dht -L DEBUG
1417 Assuming other services are well-written, they will automatically re-integrate the
1418 restarted service with the peer.
1420 GNUnet provides a powerful logging mechanism providing log levels \texttt{ERROR},
1421 \texttt{WARNING}, \texttt{INFO} and \texttt{DEBUG}. The current log level is
1422 configured using the \lstinline|$GNUNET_FORCE_LOG| environmental variable.
1423 The \texttt{DEBUG} level is only available if \lstinline|--enable-logging=verbose| was used when
1424 running \texttt{configure}. More details about logging can be found under
1425 \url{https://gnunet.org/logging}.
1427 You should also probably enable the creation of core files, by setting
1428 {\tt ulimit}, and echo'ing 1 into {\tt /proc/sys/kernel/core\_uses\_pid}.
1429 Then you can investigate the core dumps with {\tt gdb}, which is often
1430 the fastest method to find simple errors.
1432 \exercise{Add a memory leak to your service and obtain a trace
1433 pointing to the leak using {\tt valgrind} while running the service
1434 from {\tt gnunet-service-arm}.}