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 0.10.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 and 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
44 important and do not hesitate to contact the GNUnet team if you have
45 any questions or problems! Check here how to contact the GNUnet
46 team: \url{https://gnunet.org/contact_information}}
49 \section{Installing GNUnet}
50 First of all you have to install a current version of GNUnet. You can download a
51 tarball of a stable version from GNU FTP mirrors or obtain the latest development
52 version from our Subversion repository.
54 Most of the time you should prefer to download the stable version since with the
55 latest development version things can be broken, functionality can be changed or tests
56 can fail. You should only use the development version if you know that you require a
57 certain feature or a certain issue has been fixed since the last release.
59 \subsection{Obtaining a stable version}
60 You can download the latest stable version of GNUnet from GNU FTP mirrors:
62 \url{ftp://ftp.gnu.org/gnu/gnunet/gnunet-0.10.x.tar.gz}
64 You should also download the signature file and verify the integrity of the tarball.
66 \url{ftp://ftp.gnu.org/gnu/gnunet/gnunet-0.10.x.tar.gz.sig}
68 To verify the signature you should first import the GPG key used to sign the tarball
70 $ gpg --keyserver keys.gnupg.net --recv-keys 48426C7E
72 And use this key to verify the tarball's signature
74 $ gpg --verify gnunet-0.10.x.tar.gz.sig gnunet-0.10.x.tar.gz
76 After successfully verifying the integrity you can extract the tarball using
78 $ tar xvzf gnunet-0.10.x.tar.gz
79 $ mv gnunet-0.10.x gnunet # we will use the directory "gnunet" in the remainder of this document
83 \subsection{Installing Build Tool Chain and Dependencies}
84 To successfully compile GNUnet you need the tools to build GNUnet and the required dependencies.
85 Please have a look at \url{https://gnunet.org/dependencies} for a list of required dependencies
86 and \url{https://gnunet.org/generic_installation} for specific instructions for your operating system.
88 Please check the notes at the end of the configure process about required dependencies.
90 For GNUNet bootstrapping support and the http(s) plugin you should install \texttt{libcurl}.
91 For the filesharing service you should install at least one of the datastore backends \texttt{mysql},
92 \texttt{sqlite} or \texttt{postgresql}.
94 \subsection{Obtaining the latest version from Subversion}
95 The latest development version can obtained from our Subversion (\textit{svn}) repository. To obtain
96 the code you need Subversion installed and checkout the repository using:
97 \lstset{language=bash}
99 $ svn checkout https://gnunet.org/svn/gnunet
101 After cloning the repository you have to execute
102 \lstset{language=bash}
108 The remainder of this tutorial assumes that you have SVN HEAD checked out.
110 \subsection{Compiling and Installing GNUnet}
112 First, you need to install at least {\tt libgnupgerror} version
113 1.12\footnote{\url{ftp://ftp.gnupg.org/gcrypt/libgpg-error/libgpg-error-1.12.tar.bz2}}
114 and {\tt libgcrypt} version
115 1.6\footnote{\url{ftp://ftp.gnupg.org/gcrypt/libgcrypt/libgcrypt-1.6.0.tar.bz2}}.
117 \lstset{language=bash}
119 $ wget ftp://ftp.gnupg.org/gcrypt/libgpg-error/libgpg-error-1.12.tar.bz2
120 $ tar xf libgpg-error-1.12.tar.bz2
121 $ cd libgpg-error-1.12
127 \lstset{language=bash}
129 $ wget ftp://ftp.gnupg.org/gcrypt/libgcrypt/libgcrypt-1.6.0.tar.bz2
130 $ tar xf libgcrypt-1.6.0.tar.bz2
138 Assuming all dependencies are installed, the following commands will
139 compile and install GNUnet in your home directory. You can specify the
140 directory where GNUnet will be installed by changing the
141 \lstinline|--prefix| value when calling \lstinline|./configure|. If
142 you do not specifiy a prefix, GNUnet is installed in the directory
143 \lstinline|/usr/local|. When developing new applications you may want
144 to enable verbose logging by adding
145 \lstinline|--enable-logging=verbose|:
147 \lstset{language=bash}
149 $ ./configure --prefix=$PREFIX --enable-logging
154 After installing GNUnet you have to add your GNUnet installation to your path
155 environmental variable. In addition you have to create the \lstinline|.gnunet|
156 directory in your home directory where GNUnet stores it's data and an empty
157 GNUnet configuration file:
159 \lstset{language=bash}
161 $ export PATH=$PATH:$PREFIX/bin
162 $ echo export PATH=$PREFIX/bin:\\$PATH >> ~/.bashrc
164 $ touch ~/.gnunet/gnunet.conf
168 \subsection{Common Issues - Check your GNUnet installation}
169 You should check your installation to ensure that installing GNUnet
170 was successful up to this point. You should be able to access GNUnet's
171 binaries and run GNUnet's self check.
175 should return \lstinline|$PREFIX/bin/gnunet-arm|. It should be
176 located in your GNUnet installation and the output should not be
177 empty. If you see an output like:
182 check your {\tt PATH} variable to ensure GNUnet's {\tt bin} directory is included.
184 GNUnet provides tests for all of it's subcomponents. Run
188 to execute tests for all components. {\tt make check} traverses all subdirectories in {\tt src}.
189 For every subdirectory you should get a message like this:
192 make[2]: Entering directory `/home/mwachs/gnunet/contrib'
193 PASS: test_gnunet_prefix
199 If you see a message like this:
202 Mar 12 16:57:56-642482 resolver-api-19449 ERROR Must specify `HOSTNAME' for `resolver' in configuration!
203 Mar 12 16:57:56-642573 test_program-19449 ERROR Assertion failed at resolver_api.c:204.
204 /bin/bash: line 5: 19449 Aborted (core dumped) ${dir}$tst
207 double check your {\tt GNUNET\_PREFIX} environmental variable and double check the steps performed in ~\ref{sub:install}
209 \section{Background: GNUnet Architecture}
210 GNUnet is organized in layers and services. Each service is composed of a
211 main service implementation and a client library for other programs to use
212 the service's functionality, described by an API. This approach is shown in
213 figure~\ref{fig:service}. Some services provide an additional command line
214 tool to enable the user to interact with the service.
216 Very often it is other GNUnet services that will use these APIs to build the
217 higher layers of GNUnet on top of the lower ones. Each layer expands or extends
218 the functionality of the service below (for instance, to build a mesh on top of
219 a DHT). See figure ~\ref{fig:interaction} for an illustration of this approach.
224 \begin{subfigure}[b]{0.3\textwidth}
226 \includegraphics[width=\textwidth]{figs/Service.pdf}
227 \caption{Service with API and network protocol}
231 \begin{subfigure}[b]{0.3\textwidth}
233 \includegraphics[width=\textwidth]{figs/System.pdf}
234 \caption{Service interaction}
235 \label{fig:interaction}
238 \caption{GNUnet's layered system architecture}
241 The main service implementation runs as a standalone process in the operating
242 system and the client code runs as part of the client program, so crashes of a
243 client do not affect the service process or other clients. The service and the
244 clients communicate via a message protocol to be defined and implemented by
247 \section{First Steps with GNUnet}
249 \subsection{Configure your peer}
250 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.
252 Since we want to start additional peers later, we need
253 some modifications from the default configuration. We need to create a separate service home and a file containing our modifications for this peer:
259 Now add the following lines to peer1.conf to use this directory. For
260 simplified usage we want to prevent the peer to connect to the GNUnet
261 network since this could lead to confusing output. This modifications
262 will replace the default settings:
265 GNUNET_HOME = ~/gnunet1/ # Use this directory to store GNUnet data
267 SERVERS = # prevent bootstrapping
270 \subsection{Start a peer}
271 Each GNUnet instance (called peer) has an identity (\textit{peer ID}) based on a
272 cryptographic public private key pair. The peer ID is the printable hash of the
273 public key. So before starting the peer, you may want to just generate the peer's private
274 key using the command
275 \lstset{language=bash}
277 $ gnunet-peerinfo -c ~/peer1.conf -s
279 You should see an output containing the peer ID similar to:
280 \lstset{language=bash}
282 I am peer `0PA02UVRKQTS2C .. JL5Q78F6H0B1ACPV1CJI59MEQUMQCC5G'.
285 GNUnet services are controlled by a master service the so called \textit{Automatic Restart Manager} (ARM).
286 ARM starts, stops and even restarts services automatically or on demand when a client connects.
287 You interact with the ARM service using the \lstinline|gnunet-arm| tool.
288 GNUnet can then be started with \lstinline|gnunet-arm -s| and stopped with
289 \lstinline|gnunet-arm -e|. An additional service not automatically started
290 can be started using \lstinline|gnunet-arm -i <service name>| and stopped
291 using \lstinline|gnunet-arm -k <servicename>|.
293 \subsection{Monitor a peer}
294 In this section, we will monitor the behaviour of our peer's DHT service with respect to a
295 specific key. First we will start GNUnet and then start the DHT service and use the DHT monitor tool
296 to monitor the PUT and GET commands we issue ussing the \lstinline|gnunet-dht-put| and
297 \lstinline|gnunet-dht-get| command. Using the ``monitor'' line given below, you can observe the behavior of
298 your own peer's DHT with respect to the specified KEY:
300 \lstset{language=bash}
302 $ gnunet-arm -c ~/peer1.conf -s # start gnunet with all default services
303 $ gnunet-arm -c ~/peer1.conf -i dht # start DHT service
304 $ cd ~/gnunet/src/dht;
305 $ ./gnunet-dht-monitor -c ~/peer1.conf -k KEY
307 Now open a separate terminal and change again to the \lstinline|gnunet/src/dht| directory:
309 $ cd ~/gnunet/src/dht
310 $ ./gnunet-dht-put -c ~/peer1.conf -k KEY -d VALUE # put VALUE under KEY in the DHT
311 $ ./gnunet/src/dht/gnunet-dht-get -c ~/peer1.conf -k KEY # get key KEY from the DHT
312 $ gnunet-statistics -c ~/peer1.conf # print statistics about current GNUnet state
313 $ gnunet-statistics -c ~/peer1.conf -s dht # print statistics about DHT service
316 \subsection{Starting Two Peers by Hand}
317 \subsubsection{Setup a second peer}
318 We will now start a second peer on your machine.
319 For the second peer, you will need to manually create a modified
320 configuration file to avoid conflicts with ports and directories.
321 A peers configuration file is by default located in {\tt ~/.gnunet/gnunet.conf}.
322 This file is typically very short or even empty as only the differences to the
323 defaults need to be specified. The defaults are located in
324 many files in the {\tt \$GNUNET\_PREFIX/share/gnunet/config.d} directory.
326 To configure the second peer, use the files {\tt
327 \$GNUNET\_PREFIX/share/gnunet/config.d} as a template for your main
330 \lstset{language=bash}
332 $ cat $PREFIX/share/gnunet/config.d/*.conf > peer2.conf
334 Now you have to edit {\tt peer2.conf} and change:
337 \item{\texttt{SERVICEHOME} under \texttt{PATHS}}
338 \item{Every (uncommented) value for ``\texttt{PORT}'' (add 10000) in any
339 section (the option may be commented out if \texttt{PORT} is
340 prefixed by "\#", in this case, UNIX domain sockets are used
341 and the PORT option does not need to be touched) }
342 \item{Every value for ``\texttt{UNIXPATH}'' in any section (e.g. by adding a "-p2" suffix)}
344 to a fresh, unique value. Make sure that the \texttt{PORT} numbers stay
345 below 65536. From now on, whenever you interact with the second
346 peer, you need to specify {\tt -c peer2.conf} as an additional
347 command line argument.
349 Now, generate the 2nd peer's private key:
351 \lstset{language=bash}
353 $ gnunet-peerinfo -s -c peer2.conf
357 This may take a while, generate entropy using your keyboard or mouse
358 as needed. Also, make sure the output is different from the {\tt
359 gnunet-peerinfo} output for the first peer (otherwise you made an
360 error in the configuration).
362 \subsubsection{Start the second peer and connect the peers}
363 Then, you can start a second peer using:
364 \lstset{language=bash}
366 $ gnunet-arm -c peer2.conf -s
367 $ gnunet-arm -c peer2.conf -i dht
368 $ ~/gnunet/src/dht/gnunet-dht-put -c peer2.conf -k KEY -d VALUE
369 $ ~/gnunet/src/dht/gnunet-dht-get -c peer2.conf -k KEY
371 If you want the two peers to connect, you have multiple options:
374 \item UDP neighbour discovery (automatic)
375 \item Setup a bootstrap server
376 \item Connect manually
378 To setup peer 1 as bootstrapping server change the configuration of
379 the first one to be a hostlist server by adding the following lines to
380 \texttt{peer1.conf} to enable bootstrapping server:
386 Then change {\tt peer2.conf} and replace the ``\texttt{SERVERS}'' line in the ``\texttt{[hostlist]}'' section with
387 ``\texttt{http://localhost:8080/}''. Restart both peers using:
389 $ gnunet-arm -c peer1.conf -e # stop first peer
390 $ gnunet-arm -c peer1.conf -s # start first peer
391 $ gnunet-arm -c peer2.conf -s # start second peer
394 Note that if you start your peers without changing these settings, they
395 will use the ``global'' hostlist servers of the GNUnet P2P network and
396 likely connect to those peers. At that point, debugging might become
397 tricky as you're going to be connected to many more peers and would
398 likely observe traffic and behaviors that are not explicitly controlled
401 \subsubsection{How to connect manually}
402 If you want to use the \texttt{peerinfo} tool to connect your peers, you should:
405 \item{Remove {\tt hostlist} from {\tt DEFAULTSERVICES} (to not connect to the global GNUnet)}
406 \item{Start both peers running {\tt gnunet-arm -c peer1.conf -s} and {\tt gnunet-arm -c peer2.conf -s}}
407 \item{Get \texttt{HELLO} message of the first peer running {\tt gnunet-peerinfo -c peer1.conf -g}}
408 \item{Give the output to the second peer by running {\tt gnunet-peerinfo -c peer2.conf -p '<output>'}}
411 Check that they are connected using {\tt gnunet-core -c peer1.conf}, which should give you the other peer's
414 $ gnunet-core -c peer1.conf
415 Peer `9TVUCS8P5A7ILLBGO6JSTSSN2B44H3D2MUIFJMLKAITC0I22UVFBFP1H8NRK2IA35VKAK16LLO0MFS7TAQ9M1KNBJ4NGCHP3JPVULDG'
418 \subsection{Starting Peers Using the Testbed Service}
420 GNUnet's testbed service is used for testing scenarios where a number of peers
421 are to be started. The testbed can manage peers on a single host or on multiple
422 hosts in a distributed fashion. On a single affordable computer, it should be
423 possible to run around tens of peers without drastically increasing the load on the
426 The testbed service can be access through its API
427 \texttt{include/gnunet\_testbed\_service.h}. The API provides many routines for
428 managing a group of peers. It also provides a helper function
429 \texttt{GNUNET\_TESTBED\_test\_run()} to quickly setup a minimalistic testing
430 environment on a single host.
432 This function takes a configuration file which will be used as a template
433 configuration for the peers. The testbed takes care of modifying relevant
434 options in the peers' configuration such as SERVICEHOME, PORT, UNIXPATH to
435 unique values so that peers run without running into conflicts. It also checks
436 and assigns the ports in configurations only if they are free.
438 Additionally, the testbed service also reads its options from the same
439 configuration file. Various available options and details about them can be
440 found in the testbed default configuration file \texttt{src/testbed/testbed.conf}.
442 With the testbed API, a sample test case can be structured as follows:
443 \lstinputlisting[language=C]{testbed_test.c}
444 The source code for the above listing can be found at
445 \url{https://gnunet.org/svn/gnunet/doc/testbed_test.c}. After installing GNUnet, the above source code can be compiled as:
446 \lstset{language=bash}
448 $ export CPPFLAGS="-I/path/to/gnunet/headers"
449 $ export LDFLAGS="-L/path/to/gnunet/libraries"
450 $ gcc $CPPFLAGS $LDFLAGS -o testbed-test testbed_test.c -lgnunettestbed -lgnunetdht -lgnunetutil
452 The \texttt{CPPFLAGS} and \texttt{LDFLAGS} are necessary if GNUnet is installed
453 into a different directory other than \texttt{/usr/local}.
455 All of testbed API's peer management functions treat management actions as
456 operations and return operation handles. It is expected that the operations
457 begin immediately, but they may get delayed (to balance out load on the system).
458 The program using the API then has to take care of marking the operation as
459 ``done'' so that its associated resources can be freed immediately and other
460 waiting operations can be executed. Operations will be canceled if they are
461 marked as ``done'' before their completion.
463 An operation is treated as completed when it succeeds or fails. Completion of
464 an operation is either conveyed as events through \textit{controller event
465 callback} or through respective operation completion callbacks. In functions
466 which support completion notification through both controller event callback and
467 operation completion callback, first the controller event callback will be
468 called. If the operation is not marked as done in that callback or if the
469 callback is given as NULL when creating the operation, the operation completion
470 callback will be called. The API documentation shows which event are to be
471 expected in the controller event notifications. It also documents any
472 exceptional behaviour.
474 Once the peers are started, test cases often need to connect some of the peers'
475 services. Normally, opening a connect to a peer's service requires the peer's
476 configuration. While using testbed, the testbed automatically generates
477 per-peer configuration. Accessing those configurations directly through file
478 system is discouraged as their locations are dynamically created and will be
479 different among various runs of testbed. To make access to these configurations
480 easy, testbed API provides the function
481 \texttt{GNUNET\_TESTBED\_service\_connect()}. This function fetches the
482 configuration of a given peer and calls the \textit{Connect Adapter}.
483 In the example code, it is the \texttt{dht\_ca}. A connect adapter is expected
484 to open the connection to the needed service by using the provided configuration
485 and return the created service connection handle. Successful connection to the
486 needed service is signaled through \texttt{service\_connect\_comp\_cb}.
488 A dual to connect adapter is the \textit{Disconnect Adapter}. This callback is
489 called after the connect adapter has been called when the operation from
490 \texttt{GNUNET\_TESTBED\_service\_connect()} is marked as ``done''. It has to
491 disconnect from the service with the provided service handle (\texttt{op\_result}).
493 \exercise{Find out how many peers you can run on your system.}
495 \exercise{Find out how to create a 2D torus topology by changing the
496 options in the configuration file.\footnote{See \url{https://gnunet.org/content/supported-topologies}}
497 Then use the DHT API to store and retrieve values in the
500 \section{Developing Applications}
501 \subsection{gnunet-ext}
502 To develop a new peer-to-peer application or to extend GNUnet we provide
503 a template build system for writing GNUnet extensions in C. It can be
506 \lstset{language=bash}
508 $ svn checkout https://gnunet.org/svn/gnunet-ext/
511 $ ./configure --prefix=$PREFIX --with-gnunet=$PREFIX
518 The GNUnet ext template includes examples and a working buildsystem for a new GNUnet service.
519 A common GNUnet service consists of the following parts which will be discussed in detail in the
520 remainder of this document. The functionality of a GNUnet service is implemented in:
524 \item the GNUnet service (\lstinline|gnunet-ext/src/ext/gnunet-service-ext.c|)
525 \item the client API (\lstinline|gnunet-ext/src/ext/ext_api.c|)
526 \item the client application using the service API (\lstinline|gnunet-ext/src/ext/gnunet-ext.c|)
531 The interfaces for these entities are defined in:
534 \item client API interface (\lstinline|gnunet-ext/src/ext/ext.h|)
535 \item the service interface (\lstinline|gnunet-ext/src/include/gnunet_service_SERVICE.h|)
536 \item the P2P protocol (\lstinline|gnunet-ext/src/include/gnunet_protocols_ext.h|)
540 In addition the \texttt{ext} systems provides:
543 \item a test testing the API (\lstinline|gnunet-ext/src/ext/test_ext_api.c|)
544 \item a configuration template for the service (\lstinline|gnunet-ext/src/ext/ext.conf.in|)
548 \subsection{Adapting the Template}
550 The first step for writing any extension with a new service is to
551 ensure that the {\tt ext.conf.in} file contains entries for the
552 \texttt{UNIXPATH}, \texttt{PORT} and \texttt{BINARY} for the service in a section named after
555 If you want to adapt the template rename the {\tt ext.conf.in} to match your
556 services name, you have to modify the \texttt{AC\_OUTPUT} section in {\tt configure.ac}
557 in the \texttt{gnunet-ext} root.
559 \section{Writing a Client Application}
561 When writing any client application (for example, a command-line
562 tool), the basic structure is to start with the {\tt
563 GNUNET\_PROGRAM\_run} function. This function will parse
564 command-line options, setup the scheduler and then invoke the {\tt
565 run} function (with the remaining non-option arguments) and a handle
566 to the parsed configuration (and the configuration file name that was
567 used, which is typically not needed):
571 #include <gnunet/platform.h>
572 #include <gnunet/gnunet_util_lib.h>
580 const struct GNUNET_CONFIGURATION_Handle *cfg)
587 main (int argc, char *const *argv)
589 static const struct GNUNET_GETOPT_CommandLineOption options[] = {
590 GNUNET_GETOPT_OPTION_END
593 GNUNET_PROGRAM_run (argc,
596 gettext_noop ("binary description text"),
597 options, &run, NULL)) ? ret : 1;
601 \subsection{Handling command-line options}
603 Options can then be added easily by adding global variables and
604 expanding the {\tt options} array. For example, the following would
605 add a string-option and a binary flag (defaulting to {\tt NULL} and
606 {\tt GNUNET\_NO} respectively):
609 static char *string_option;
613 static const struct GNUNET_GETOPT_CommandLineOption options[] = {
614 {'s', "name", "SOMESTRING",
615 gettext_noop ("text describing the string_option NAME"), 1,
616 &GNUNET_GETOPT_set_string, &string_option},
618 gettext_noop ("text describing the flag option"), 0,
619 &GNUNET_GETOPT_set_one, &a_flag},
620 GNUNET_GETOPT_OPTION_END
622 string_option = NULL;
623 a_flag = GNUNET_SYSERR;
627 Issues such as displaying some helpful text describing options using
628 the {\tt --help} argument and error handling are taken care of when
629 using this approach. Other {\tt GNUNET\_GETOPT\_}-functions can be used
630 to obtain integer value options, increment counters, etc. You can
631 even write custom option parsers for special circumstances not covered
632 by the available handlers. To check if an argument was specified by the
633 user you initialize the variable with a specific value (e.g. NULL for
634 a string and GNUNET\_SYSERR for a integer) and check after parsing
635 happened if the values were modified.
637 Inside the {\tt run} method, the program would perform the
638 application-specific logic, which typically involves initializing and
639 using some client library to interact with the service. The client
640 library is supposed to implement the IPC whereas the service provides
641 more persistent P2P functions.
643 \exercise{Add a few command-line options and print them inside
644 of {\tt run}. What happens if the user gives invalid arguments?}
646 \subsection{Writing a Client Library}
648 The first and most important step in writing a client library is to
649 decide on an API for the library. Typical API calls include
650 connecting to the service, performing application-specific requests
651 and cleaning up. Many examples for such service APIs can be found
652 in the {\tt gnunet/src/include/gnunet\_*\_service.h} files.
654 Then, a client-service protocol needs to be designed. This typically
655 involves defining various message formats in a header that will be
656 included by both the service and the client library (but is otherwise
657 not shared and hence located within the service's directory and not
658 installed by {\tt make install}). Each message must start with a {\tt
659 struct GNUNET\_MessageHeader} and must be shorter than 64k. By
660 convention, all fields in IPC (and P2P) messages must be in big-endian
661 format (and thus should be read using {\tt ntohl} and similar
662 functions and written using {\tt htonl} and similar functions).
663 Unique message types must be defined for each message struct in the
664 {\tt gnunet\_protocols.h} header (or an extension-specific include
667 \subsubsection{Connecting to the Service}
669 Before a client library can implement the application-specific protocol
670 with the service, a connection must be created:
674 struct GNUNET_CLIENT_Connection *client;
675 client = GNUNET_CLIENT_connect ("service-name", cfg);
678 As a result a {\tt GNUNET\_CLIENT\_Connection} handle is returned
679 which has to used in later API calls related to this service.
680 The complete client API can be found in {\tt gnunet\_client\_lib.h}
682 \subsubsection{GNUnet Messages}
684 In GNUnet, messages are always sent beginning with a {\tt struct GNUNET\_MessageHeader}
685 in big endian format. This header defines the size and the type of the
686 message, the payload follows after this header.
690 struct GNUNET_MessageHeader
694 * The length of the struct (in bytes, including the length field itself),
695 * in big-endian format.
697 uint16_t size GNUNET_PACKED;
700 * The type of the message (GNUNET_MESSAGE_TYPE_XXXX), in big-endian format.
702 uint16_t type GNUNET_PACKED;
707 Existing message types are defined in {\tt gnunet\_protocols.h}\\
708 A common way to create a message is:
712 struct GNUNET_MessageHeader *msg =
713 GNUNET_malloc(payload_size + sizeof(struct GNUNET_MessageHeader));
714 msg->size = htons(payload_size + sizeof(struct GNUNET_MessageHeader));
715 msg->type = htons(GNUNET_MY_MESSAGE_TYPE);
716 memcpy(&msg[1], &payload, payload_size);
720 \exercise{Define a message struct that includes a 32-bit
721 unsigned integer in addition to the standard GNUnet MessageHeader.
722 Add a C struct and define a fresh protocol number for your message.
723 (Protocol numbers in gnunet-ext are defined in \lstinline|gnunet-ext/src/include/gnunet_protocols_ext.h|)}
726 \subsubsection{Sending Requests to the Service}
728 Any client-service protocol must start with the client sending the
729 first message to the service, since services are only notified about
730 (new) clients upon receiving a the first message.
732 Clients can transmit messages to the service using the
733 {\tt GNUNET\_CLIENT\_notify\_transmit\_ready} API:
737 transmit_cb (void *cls, size_t size, void *buf)
740 if (NULL == buf) { /* handle error here */; return 0; }
741 GNUNET_assert (size >= msg_size);
742 memcpy (buf, my_msg, msg_size);
748 th = GNUNET_CLIENT_notify_transmit_ready (client,
756 The client-service protocoll calls {\tt GNUNET\_CLIENT\_notify\_transmit\_ready}
757 to be notified when the client is ready to send data to the service.
758 Besides other arguments, you have to pass the client returned
759 from the {\tt connect} call, the message size and the callback function to
760 call when the client is ready to send.
762 Only a single transmission request can be queued per client at the
763 same time using this API. The handle {\tt th} can be used to cancel
764 the request if necessary (for example, during shutdown).
766 When {\tt transmit\_cb} is called the message is copied in the buffer provided and
767 the number of bytes copied into the buffer is returned. {\tt transmit\_cb}
768 could also return 0 if for some reason no message
769 could be constructed; this is not an error and the connection to the
770 service will persist in this case.
772 \exercise{Define a helper function to transmit a 32-bit
773 unsigned integer (as payload) to a service using some given client
777 \subsubsection{Receiving Replies from the Service}
779 Clients can receive messages from the service using the
780 {\tt GNUNET\_CLIENT\_receive} API:
785 * Function called with messages from stats service.
788 * @param msg message received, NULL on timeout or fatal error
791 receive_message (void *cls, const struct GNUNET_MessageHeader *msg)
793 struct MyArg *arg = cls;
799 GNUNET_CLIENT_receive (client,
806 It should be noted that this receive call only receives a single
807 message. To receive additional messages, {\tt
808 GNUNET\_CLIENT\_receive} must be called again.
810 \exercise{Expand your helper function to receive a
811 response message (for example, containing just the GNUnet MessageHeader
812 without any payload). Upon receiving the service's response, you should
813 call a callback provided to your helper function's API. You'll need to
814 define a new 'struct' to hold your local context (``closure'').}
817 \subsection{Writing a user interface}
819 Given a client library, all it takes to access a service now is to
820 combine calls to the client library with parsing command-line
823 \exercise{Call your client API from your {\tt run} method
824 in your client application to send a request to the service.
825 For example, send a 32-bit integer value based on a number given
826 at the command-line to the service.}
830 \section{Writing a Service}
832 Before you can test the client you've written so far, you'll need to also
833 implement the corresponding service.
836 \subsection{Code Placement}
838 New services are placed in their own subdirectory under {\tt gnunet/src}.
839 This subdirectory should contain the API implementation file {\tt SERVICE\_api.c},
840 the description of the client-service protocol {\tt SERVICE.h} and P2P protocol
841 {\tt SERVICE\_protocol.h}, the implementation of the service itself
842 {\tt gnunet-service-SERVICE.h} and several files for tests, including test code
843 and configuration files.
845 \subsection{Starting a Service}
847 The key API definitions for starting services are:
850 typedef void (*GNUNET_SERVICE_Main) (void *cls,
851 struct GNUNET_SERVER_Handle *server,
852 const struct GNUNET_CONFIGURATION_Handle *cfg);
853 int GNUNET_SERVICE_run (int argc,
855 const char *serviceName,
856 enum GNUNET_SERVICE_Options opt,
857 GNUNET_SERVICE_Main task,
861 Here is a starting point for your main function for your service:
865 static void my_main (void *cls,
866 struct GNUNET_SERVER_Handle *server,
867 const struct GNUNET_CONFIGURATION_Handle *cfg)
872 int main (int argc, char *const*argv)
875 GNUNET_SERVICE_run (argc, argv, "my",
876 GNUNET_SERVICE_OPTION_NONE,
883 \exercise{Write a stub service that processes no messages at all
884 in your code. Create a default configuration for it, integrate it
885 with the build system and start the service from {\tt
886 gnunet-service-arm} using {\tt gnunet-arm -i NAME}.}
889 \subsection{Receiving Requests from Clients}
891 Inside of the {\tt my\_main} method, a service typically registers for
892 the various message types from clients that it supports by providing
893 a handler function, the message type itself and possibly a fixed
894 message size (or 0 for variable-size messages):
899 handle_set (void *cls,
900 struct GNUNET_SERVER_Client *client,
901 const struct GNUNET_MessageHeader *message)
903 GNUNET_SERVER_receive_done (client, GNUNET_OK);
906 handle_get (void *cls,
907 struct GNUNET_SERVER_Client *client,
908 const struct GNUNET_MessageHeader *message)
910 GNUNET_SERVER_receive_done (client, GNUNET_OK);
913 static void my_main (void *cls,
914 struct GNUNET_SERVER_Handle *server,
915 const struct GNUNET_CONFIGURATION_Handle *cfg)
917 static const struct GNUNET_SERVER_MessageHandler handlers[] = {
918 {&handle_set, NULL, GNUNET_MESSAGE_TYPE_MYNAME_SET, 0},
919 {&handle_get, NULL, GNUNET_MESSAGE_TYPE_MYNAME_GET, 0},
922 GNUNET_SERVER_add_handlers (server, handlers);
923 /* do more setup work */
927 Each handler function {\bf must} eventually (possibly in some
928 asynchronous continuation) call {\tt GNUNET\_SERVER\_receive\_done}.
929 Only after this call additional messages from the same client may
930 be processed. This way, the service can throttle processing messages
931 from the same client. By passing {\tt GNUNET\_SYSERR}, the service
932 can close the connection to the client, indicating an error.
934 Services must check that client requests are well-formed and must not
935 crash on protocol violations by the clients. Similarly, client
936 libraries must check replies from servers and should gracefully report
937 errors via their API.
940 \exercise{Change the service to ``handle'' the message from your
941 client (for now, by printing a message). What happens if you
942 forget to call {\tt GNUNET\_SERVER\_receive\_done}?}
945 \subsection{Responding to Clients}
947 Servers can send messages to clients using the
948 {\tt GNUNET\_SERVER\_notify\_transmit\_ready} API:
953 transmit_cb (void *cls, size_t size, void *buf)
956 if (NULL == buf) { handle_error(); return 0; }
957 GNUNET_assert (size >= msg_size);
958 memcpy (buf, my_msg, msg_size);
964 struct GNUNET_SERVER_TransmitHandle *th;
965 th = GNUNET_SERVER_notify_transmit_ready (client,
972 Only a single transmission request can be queued per client
973 at the same time using this API.
974 Additional APIs for sending messages to clients can be found
975 in the {\tt gnunet\_server\_lib.h} header.
978 \exercise{Change the service respond to the request from your
979 client. Make sure you handle malformed messages in both directions.}
982 \section{Interacting directly with other Peers using the CORE Service}
984 One of the most important services in GNUnet is the \texttt{CORE} service
985 managing connections between peers and handling encryption between peers.
987 One of the first things any service that extends the P2P protocol typically does
988 is connect to the \texttt{CORE} service using:
992 #include <gnunet/gnunet_core_service.h>
994 struct GNUNET_CORE_Handle *
995 GNUNET_CORE_connect (const struct GNUNET_CONFIGURATION_Handle *cfg,
997 GNUNET_CORE_StartupCallback init,
998 GNUNET_CORE_ConnectEventHandler connects,
999 GNUNET_CORE_DisconnectEventHandler disconnects,
1000 GNUNET_CORE_MessageCallback inbound_notify,
1001 int inbound_hdr_only,
1002 GNUNET_CORE_MessageCallback outbound_notify,
1003 int outbound_hdr_only,
1004 const struct GNUNET_CORE_MessageHandler *handlers);
1007 \subsection{New P2P connections}
1009 Before any traffic with a different peer can be exchanged, the peer must be
1010 known to the service. This is notified by the \texttt{CORE} {\tt connects} callback,
1011 which communicates the identity of the new peer to the service:
1016 connects (void *cls,
1017 const struct GNUNET_PeerIdentity * peer)
1019 /* Save identity for later use */
1020 /* Optional: start sending messages to peer */
1024 \exercise{Create a service that connects to the \texttt{CORE}. Then
1025 start (and connect) two peers and print a message once your connect
1026 callback is invoked.}
1028 \subsection{Receiving P2P Messages}
1030 To receive messages from \texttt{CORE}, services register a set of handlers
1031 (parameter {\tt *handlers} in the \lstinline|GNUNET_CORE_connect| call that are called by \texttt{CORE}
1032 when a suitable message arrives.
1037 callback_function_for_type_one(void *cls,
1038 const struct GNUNET_PeerIdentity *peer,
1039 const struct GNUNET_MessageHeader *message)
1042 return GNUNET_OK; /* or GNUNET_SYSERR to close the connection */
1046 * Functions to handle messages from core
1048 static struct GNUNET_CORE_MessageHandler core_handlers[] = {
1049 {&callback_function_for_type_one, GNUNET_MESSAGE_TYPE_MYSERVICE_TYPE_ONE, 0},
1055 \exercise{Start one peer with a new service that has a message
1056 handler and start a second peer that only has your ``old'' service
1057 without message handlers. Which ``connect'' handlers are invoked when
1058 the two peers are connected? Why?}
1061 \subsection{Sending P2P Messages}
1063 In response to events (connect, disconnect, inbound messages,
1064 timing, etc.) services can then use this API to transmit messages:
1069 (*GNUNET_CONNECTION_TransmitReadyNotify) (void *cls,
1073 /* Fill "*buf" with up to "size" bytes, must start with GNUNET_MessageHeader */
1074 return n; /* Total size of the message put in "*buf" */
1077 struct GNUNET_CORE_TransmitHandle *
1078 GNUNET_CORE_notify_transmit_ready (struct GNUNET_CORE_Handle *handle,
1079 int cork, uint32_t priority,
1080 struct GNUNET_TIME_Relative maxdelay,
1081 const struct GNUNET_PeerIdentity *target,
1083 GNUNET_CONNECTION_TransmitReadyNotify notify,
1087 \exercise{Write a service that upon connect sends messages as
1088 fast as possible to the other peer (the other peer should run a
1089 service that ``processes'' those messages). How fast is the
1090 transmission? Count using the STATISTICS service on both ends. Are
1091 messages lost? How can you transmit messages faster? What happens if
1092 you stop the peer that is receiving your messages?}
1095 \subsection{End of P2P connections}
1097 If a message handler returns {\tt GNUNET\_SYSERR}, the remote peer shuts down or
1098 there is an unrecoverable network disconnection, CORE notifies the service that
1099 the peer disconnected. After this notification no more messages will be received
1100 from the peer and the service is no longer allowed to send messages to the peer.
1101 The disconnect callback looks like the following:
1106 disconnects (void *cls,
1107 const struct GNUNET_PeerIdentity * peer)
1109 /* Remove peer's identity from known peers */
1110 /* Make sure no messages are sent to peer from now on */
1114 \exercise{Fix your service to handle peer disconnects.}
1116 \section{Using the DHT}
1117 The DHT allows to store data so other peers in the P2P network can
1118 access it and retrieve data stored by any peers in the network.
1119 This section will explain how to use the DHT. Of course, the first
1120 thing to do is to connect to the DHT service:
1123 dht_handle = GNUNET_DHT_connect (cfg, parallel_requests);
1125 The second parameter indicates how many requests in parallel to expect.
1126 It is not a hard limit, but a good approximation will make the DHT more
1129 \subsection{Storing data in the DHT}
1130 Since the DHT is a dynamic environment (peers join and leave frequently)
1131 the data that we put in the DHT does not stay there indefinitely. It is
1132 important to ``refresh'' the data periodically by simply storing it again,
1133 in order to make sure other peers can access it.
1135 The put API call offers a callback to signal that the PUT request has been
1136 sent. This does not guarantee that the data is accessible to others peers,
1137 or even that is has been stored, only that the service has requested to
1138 a neighboring peer the retransmission of the PUT request towards its final
1139 destination. Currently there is no feedback about whether or not the data
1140 has been sucessfully stored or where it has been stored. In order to improve
1141 the availablilty of the data and to compensate for possible errors, peers leaving
1142 and other unfavorable events, just make several PUT requests!
1147 message_sent_cont (void *cls, const struct GNUNET_SCHEDULER_TaskContext *tc)
1149 /* Request has left local node */
1152 struct GNUNET_DHT_PutHandle *
1153 GNUNET_DHT_put (struct GNUNET_DHT_Handle *handle,
1154 const struct GNUNET_HashCode * key,
1155 uint32_t desired_replication_level,
1156 enum GNUNET_DHT_RouteOption options, /* Route options, see next call */
1157 enum GNUNET_BLOCK_Type type, size_t size, const void *data,
1158 struct GNUNET_TIME_Absolute exp, /* When does the data expire? */
1159 struct GNUNET_TIME_Relative timeout, /* How long to try to send the request */
1160 GNUNET_DHT_PutContinuation cont,
1164 \exercise{Store a value in the DHT periodically to make sure it is available
1165 over time. You might consider using the function GNUNET\_SCHEDULER\_add\_delayed and
1166 call GNUNET\_DHT\_put from inside a helper function.}
1169 \subsection{Obtaining data from the DHT}
1170 As we saw in the previous example, the DHT works in an asynchronous mode.
1171 Each request to the DHT is executed ``in the background'' and the API
1172 calls return immediately. In order to receive results from the DHT, the
1173 API provides a callback. Once started, the request runs in the service,
1174 the service will try to get as many results as possible (filtering out
1175 duplicates) until the timeout expires or we explicitly stop the request.
1176 It is possible to give a ``forever'' timeout with
1177 {\tt GNUNET\_TIME\_UNIT\_FOREVER\_REL}.
1179 If we give a route option {\tt GNUNET\_DHT\_RO\_RECORD\_ROUTE} the callback
1180 will get a list of all the peers the data has travelled, both on the PUT
1181 path and on the GET path.
1185 get_result_iterator (void *cls, struct GNUNET_TIME_Absolute expiration,
1186 const struct GNUNET_HashCode * key,
1187 const struct GNUNET_PeerIdentity *get_path,
1188 unsigned int get_path_length,
1189 const struct GNUNET_PeerIdentity *put_path,
1190 unsigned int put_path_length,
1191 enum GNUNET_BLOCK_Type type, size_t size, const void *data)
1193 /* Do stuff with the data and/or route */
1195 GNUNET_DHT_get_stop (get_handle);
1199 GNUNET_DHT_get_start (dht_handle,
1203 GNUNET_DHT_RO_NONE, /* Route options */
1204 NULL, /* xquery: not used here */
1205 0, /* xquery size */
1206 &get_result_iterator,
1210 \exercise{Store a value in the DHT and after a while retrieve it. Show the IDs of all
1211 the peers the requests have gone through. In order to convert a peer ID to a string, use
1212 the function GNUNET\_i2s. Pay attention to the route option parameters in both calls!}
1214 \subsection{Implementing a block plugin}
1216 In order to store data in the DHT, it is necessary to provide a block
1217 plugin. The DHT uses the block plugin to ensure that only well-formed
1218 requests and replies are transmitted over the network.
1220 The block plugin should be put in a file {\tt
1221 plugin\_block\_SERVICE.c} in the service's respective directory. The
1222 mandatory functions that need to be implemented for a block plugin are
1223 described in the following sections.
1225 \subsubsection{Validating requests and replies}
1227 The evaluate function should validate a reply or a request. It returns
1228 a {\tt GNUNET\_BLOCK\_EvaluationResult}, which is an enumeration. All
1229 possible answers are in {\tt gnunet\_block\_lib.h}. The function will
1230 be called with a {\tt reply\_block} argument of {\tt NULL} for
1231 requests. Note that depending on how {\tt evaluate} is called, only
1232 some of the possible return values are valid. The specific meaning of
1233 the {\tt xquery} argument is application-specific. Applications that
1234 do not use an extended query should check that the {\tt xquery\_size}
1235 is zero. The Bloom filter is typically used to filter duplicate
1240 static enum GNUNET_BLOCK_EvaluationResult
1241 block_plugin_SERVICE_evaluate (void *cls,
1242 enum GNUNET_BLOCK_Type type,
1243 const GNUNET_HashCode * query,
1244 struct GNUNET_CONTAINER_BloomFilter **bf,
1248 const void *reply_block,
1249 size_t reply_block_size)
1251 /* Verify type, block and bloomfilter */
1255 Note that it is mandatory to detect duplicate replies in this
1256 function and return the respective status code. Duplicate
1257 detection should be done by setting the respective bits in
1258 the Bloom filter {\tt bf}. Failure to do so may cause replies
1259 to circle in the network.
1261 \subsubsection{Deriving a key from a reply}
1263 The DHT can operate more efficiently if it is possible to derive a key
1264 from the value of the corresponding block. The {\tt get\_key}
1265 function is used to obtain the key of a block --- for example, by
1266 means of hashing. If deriving the key is not possible, the function
1267 should simply return {\tt GNUNET\_SYSERR} (the DHT will still work
1268 just fine with such blocks).
1273 block_plugin_SERVICE_get_key (void *cls, enum GNUNET_BLOCK_Type type,
1274 const void *block, size_t block_size,
1275 GNUNET_HashCode * key)
1277 /* Store the key in the key argument, return GNUNET_OK on success. */
1281 \subsubsection{Initialization of the plugin}
1283 The plugin is realized as a shared C library. The library must export
1284 an initialization function which should initialize the plugin. The
1285 initialization function specifies what block types the plugin cares
1286 about and returns a struct with the functions that are to be used for
1287 validation and obtaining keys (the ones just defined above).
1292 libgnunet_plugin_block_SERVICE_init (void *cls)
1294 static enum GNUNET_BLOCK_Type types[] =
1296 GNUNET_BLOCK_TYPE_SERVICE_BLOCKYPE, /* list of blocks we care about, from gnunet_block_lib.h */
1297 GNUNET_BLOCK_TYPE_ANY /* end of list */
1299 struct GNUNET_BLOCK_PluginFunctions *api;
1301 api = GNUNET_malloc (sizeof (struct GNUNET_BLOCK_PluginFunctions));
1302 api->evaluate = &block_plugin_SERICE_evaluate;
1303 api->get_key = &block_plugin_SERVICE_get_key;
1309 \subsubsection{Shutdown of the plugin}
1311 Following GNUnet's general plugin API concept, the plugin must
1312 export a second function for cleaning up. It usually does very
1318 libgnunet_plugin_block_SERVICE_done (void *cls)
1320 struct GNUNET_TRANSPORT_PluginFunctions *api = cls;
1328 \subsubsection{Integration of the plugin with the build system}
1330 In order to compile the plugin, the {\tt Makefile.am} file for the
1331 service \texttt{SERVICE} should contain a rule similar to this:
1333 \lstset{language=make}
1335 plugindir = $(libdir)/gnunet
1337 plugin_LTLIBRARIES = \
1338 libgnunet_plugin_block_ext.la
1339 libgnunet_plugin_block_ext_la_SOURCES = \
1341 libgnunet_plugin_block_ext_la_LIBADD = \
1342 $(prefix)/lib/libgnunethello.la \
1343 $(prefix)/lib/libgnunetblock.la \
1344 $(prefix)/lib/libgnunetutil.la
1345 libgnunet_plugin_block_ext_la_LDFLAGS = \
1346 $(GN_PLUGIN_LDFLAGS)
1347 libgnunet_plugin_block_ext_la_DEPENDENCIES = \
1348 $(prefix)/lib/libgnunetblock.la
1353 \exercise{Write a block plugin that accepts all queries
1354 and all replies but prints information about queries and replies
1355 when the respective validation hooks are called.}
1359 \subsection{Monitoring the DHT}
1360 It is possible to monitor the functioning of the local DHT service. When monitoring
1361 the DHT, the service will alert the monitoring program of any events,
1362 both started locally or received for routing from another peer. The are three different
1363 types of events possible: a GET request, a PUT request or a response (a reply to
1366 Since the different events have different associated data, the API gets 3
1367 different callbacks (one for each message type) and optional type and key parameters,
1368 to allow for filtering of messages. When an event happens, the appropiate callback
1369 is called with all the information about the event.
1373 get_callback (void *cls,
1374 enum GNUNET_DHT_RouteOption options,
1375 enum GNUNET_BLOCK_Type type,
1377 uint32_t desired_replication_level,
1378 unsigned int path_length,
1379 const struct GNUNET_PeerIdentity *path,
1380 const struct GNUNET_HashCode * key)
1385 get_resp_callback (void *cls,
1386 enum GNUNET_BLOCK_Type type,
1387 const struct GNUNET_PeerIdentity *get_path,
1388 unsigned int get_path_length,
1389 const struct GNUNET_PeerIdentity *put_path,
1390 unsigned int put_path_length,
1391 struct GNUNET_TIME_Absolute exp,
1392 const struct GNUNET_HashCode * key,
1399 put_callback (void *cls,
1400 enum GNUNET_DHT_RouteOption options,
1401 enum GNUNET_BLOCK_Type type,
1403 uint32_t desired_replication_level,
1404 unsigned int path_length,
1405 const struct GNUNET_PeerIdentity *path,
1406 struct GNUNET_TIME_Absolute exp,
1407 const struct GNUNET_HashCode * key,
1413 monitor_handle = GNUNET_DHT_monitor_start (dht_handle,
1414 block_type, /* GNUNET_BLOCK_TYPE_ANY for all */
1415 key, /* NULL for all */
1423 \section{Debugging with {\tt gnunet-arm}}
1425 Even if services are managed by {\tt gnunet-arm}, you can start them with
1426 {\tt gdb} or {\tt valgrind}. For example, you could add the following lines
1427 to your configuration file to start the DHT service in a {\tt gdb} session in a
1432 PREFIX=xterm -e gdb --args
1435 Alternatively, you can stop a service that was started via ARM and run it manually:
1437 \lstset{language=bash}
1440 $ gdb --args gnunet-service-dht -L DEBUG
1441 $ valgrind gnunet-service-dht -L DEBUG
1445 Assuming other services are well-written, they will automatically re-integrate the
1446 restarted service with the peer.
1448 GNUnet provides a powerful logging mechanism providing log levels \texttt{ERROR},
1449 \texttt{WARNING}, \texttt{INFO} and \texttt{DEBUG}. The current log level is
1450 configured using the \lstinline|$GNUNET_FORCE_LOG| environmental variable.
1451 The \texttt{DEBUG} level is only available if \lstinline|--enable-logging=verbose| was used when
1452 running \texttt{configure}. More details about logging can be found under
1453 \url{https://gnunet.org/logging}.
1455 You should also probably enable the creation of core files, by setting
1456 {\tt ulimit}, and echo'ing 1 into {\tt /proc/sys/kernel/core\_uses\_pid}.
1457 Then you can investigate the core dumps with {\tt gdb}, which is often
1458 the fastest method to find simple errors.
1460 \exercise{Add a memory leak to your service and obtain a trace
1461 pointing to the leak using {\tt valgrind} while running the service
1462 from {\tt gnunet-service-arm}.}