doc/user/installation.rst

   1 .. _installation:
   2
   3 Installation
   4 ============
   5
   6 .. index:: How to install FRR
   7 .. index:: Installation
   8 .. index:: Installing FRR
   9 .. index:: Building the system
  10 .. index:: Making FRR
  11
  12 This section covers the basics of building, installing and setting up FRR.
  13
  14 From Packages
  15 -------------
  16
  17 The project publishes packages for Red Hat, Centos, Debian and Ubuntu on the
  18 `GitHub releases <https://github.com/FRRouting/frr/releases>`_. page. External
  19 contributors offer packages for many other platforms including \*BSD, Alpine,
  20 Gentoo, Docker, and others. There is currently no documentation on how to use
  21 those but we hope to add it soon.
  22
  23 From Snapcraft
  24 --------------
  25
  26 In addition to traditional packages the project also builds and publishes
  27 universal Snap images, available at https://snapcraft.io/frr.
  28
  29 From Source
  30 -----------
  31
  32 Building FRR from source is the best way to ensure you have the latest features
  33 and bug fixes. Details for each supported platform, including dependency
  34 package listings, permissions, and other gotchas, are in the developer's
  35 documentation. This section provides a brief overview on the process.
  36
  37 Getting the Source
  38 ^^^^^^^^^^^^^^^^^^
  39
  40 FRR's source is available on the project
  41 `GitHub page <https://github.com/FRRouting/frr>`_.
  42
  43 .. code-block:: shell
  44
  45    git clone https://github.com/FRRouting/frr.git
  46
  47 When building from Git there are several branches to choose from. The
  48 ``master`` branch is the primary development branch. It should be considered
  49 unstable. Each release has its own branch named ``stable/X.X``, where ``X.X``
  50 is the release version.
  51
  52 In addition, release tarballs are published on the GitHub releases page
  53 `here <https://github.com/FRRouting/frr/releases>`_.
  54
  55 Configuration
  56 ^^^^^^^^^^^^^
  57
  58 .. index:: Configuration options
  59 .. index:: Options for configuring
  60 .. index:: Build options
  61 .. index:: Distribution configuration
  62 .. index:: Options to `./configure`
  63
  64 FRR has an excellent configure script which automatically detects most host
  65 configurations. There are several additional configure options to customize the
  66 build to include or exclude specific features and dependencies.
  67
  68 First, update the build system. Change into your FRR source directory and issue:
  69
  70 .. code-block:: shell
  71
  72    ./bootstrap.sh
  73
  74 This will install any missing build scripts and update the Autotools
  75 configuration. Once this is done you can move on to choosing your configuration
  76 options from the list below.
  77
  78 .. _frr-configuration:
  79
  80 .. program:: configure
  81
  82 .. option:: --disable-zebra
  83
  84    Do not build zebra daemon.
  85
  86 .. option:: --disable-ripd
  87
  88    Do not build ripd.
  89
  90 .. option:: --disable-ripngd
  91
  92    Do not build ripngd.
  93
  94 .. option:: --disable-ospfd
  95
  96    Do not build ospfd.
  97
  98 .. option:: --disable-ospf6d
  99
 100    Do not build ospf6d.
 101
 102 .. option:: --disable-bgpd
 103
 104    Do not build bgpd.
 105
 106 .. option:: --disable-bfdd
 107
 108    Do not build bfdd.
 109
 110 .. option:: --disable-bgp-announce
 111
 112    Make *bgpd* which does not make bgp announcements at all.  This
 113    feature is good for using *bgpd* as a BGP announcement listener.
 114
 115 .. option:: --enable-datacenter
 116
 117    Enable system defaults to work as if in a Data Center. See defaults.h
 118    for what is changed by this configure option.
 119
 120 .. option:: --enable-snmp
 121
 122    Enable SNMP support.  By default, SNMP support is disabled.
 123
 124 .. option:: --disable-ospfapi
 125
 126    Disable support for OSPF-API, an API to interface directly with ospfd.
 127    OSPF-API is enabled if --enable-opaque-lsa is set.
 128
 129 .. option:: --disable-ospfclient
 130
 131    Disable building of the example OSPF-API client.
 132
 133 .. option:: --disable-ospf-ri
 134
 135    Disable support for OSPF Router Information (RFC4970 & RFC5088) this
 136    requires support for Opaque LSAs and Traffic Engineering.
 137
 138 .. option:: --disable-isisd
 139
 140    Do not build isisd.
 141
 142 .. option:: --disable-fabricd
 143
 144    Do not build fabricd.
 145
 146 .. option:: --enable-isis-topology
 147
 148    Enable IS-IS topology generator.
 149
 150 .. option:: --enable-isis-te
 151
 152    Enable Traffic Engineering Extension for ISIS (RFC5305)
 153
 154 .. option:: --enable-realms
 155
 156    Enable the support of Linux Realms. Convert tag values from 1-255 into a
 157    realm value when inserting into the Linux kernel. Then routing policy can be
 158    assigned to the realm. See the tc man page.
 159
 160 .. option:: --disable-rtadv
 161
 162    Disable support IPV6 router advertisement in zebra.
 163
 164 .. option:: --enable-gcc-rdynamic
 165
 166    Pass the ``-rdynamic`` option to the linker driver.  This is in most cases
 167    necessary for getting usable backtraces.  This option defaults to on if the
 168    compiler is detected as gcc, but giving an explicit enable/disable is
 169    suggested.
 170
 171 .. option:: --disable-backtrace
 172
 173    Controls backtrace support for the crash handlers. This is autodetected by
 174    default. Using the switch will enforce the requested behaviour, failing with
 175    an error if support is requested but not available.  On BSD systems, this
 176    needs libexecinfo, while on glibc support for this is part of libc itself.
 177
 178 .. option:: --enable-dev-build
 179
 180    Turn on some options for compiling FRR within a development environment in
 181    mind.  Specifically turn on -g3 -O0 for compiling options and add inclusion
 182    of grammar sandbox.
 183
 184 .. option:: --enable-fuzzing
 185
 186    Turn on some compile options to allow you to run fuzzing tools against the
 187    system. This flag is intended as a developer only tool and should not be
 188    used for normal operations.
 189
 190 .. option:: --disable-snmp
 191
 192    Build without SNMP support.
 193
 194 .. option:: --disable-vtysh
 195
 196    Build without VTYSH.
 197
 198 .. option:: --enable-fpm
 199
 200    Build with FPM module support.
 201
 202 .. option:: --enable-numeric-version
 203
 204    Alpine Linux does not allow non-numeric characters in the version string.
 205    With this option, we provide a way to strip out these characters for APK dev
 206    package builds.
 207
 208 .. option:: --enable-multipath=X
 209
 210    Compile FRR with up to X way ECMP supported.  This number can be from 0-999.
 211    For backwards compatability with older configure options when setting X = 0,
 212    we will build FRR with 64 way ECMP.  This is needed because there are
 213    hardcoded arrays that FRR builds towards, so we need to know how big to
 214    make these arrays at build time.
 215
 216 .. option:: --enable-gcov
 217
 218    Code coverage reports from gcov require adjustments to the C and LD flags.
 219    With this option, gcov instrumentation is added to the build and coverage
 220    reports are created during execution.  The check-coverage make target is
 221    also created to ease report uploading to codecov.io.  The upload requires
 222    the COMMIT (git hash) and TOKEN (codecov upload token) environment variables
 223    be set.
 224
 225 .. option:: --enable-config-rollbacks
 226
 227    Build with configuration rollback support. Requires SQLite3.
 228
 229 .. option:: --enable-confd=<dir>
 230
 231    Build the ConfD northbound plugin. Look for the libconfd libs and headers
 232    in `dir`.
 233
 234 You may specify any combination of the above options to the configure
 235 script. By default, the executables are placed in :file:`/usr/local/sbin`
 236 and the configuration files in :file:`/usr/local/etc`. The :file:`/usr/local/`
 237 installation prefix and other directories may be changed using the following
 238 options to the configuration script.
 239
 240 .. option:: --prefix <prefix>
 241
 242    Install architecture-independent files in `prefix` [/usr/local].
 243
 244 .. option:: --sysconfdir <dir>
 245
 246    Look for configuration files in `dir` [`prefix`/etc]. Note that sample
 247    configuration files will be installed here.
 248
 249 .. option:: --localstatedir <dir>
 250
 251    Configure zebra to use `dir` for local state files, such as pid files and
 252    unix sockets.
 253
 254 .. _least-privilege-support:
 255
 256 Least-Privilege Support
 257 """""""""""""""""""""""
 258
 259 .. index:: FRR Least-Privileges
 260 .. index:: FRR Privileges
 261
 262 Additionally, you may configure zebra to drop its elevated privileges
 263 shortly after startup and switch to another user. The configure script will
 264 automatically try to configure this support. There are three configure
 265 options to control the behaviour of FRR daemons.
 266
 267 .. option:: --enable-user <user>
 268
 269    Switch to user `user shortly after startup, and run as user `user` in normal
 270    operation.
 271
 272 .. option:: --enable-group <user>
 273
 274    Switch real and effective group to `group` shortly after startup.
 275
 276 .. option:: --enable-vty-group <group>
 277
 278    Create Unix Vty sockets (for use with vtysh) with group ownership set to
 279    `group`. This allows one to create a separate group which is restricted to
 280    accessing only the vty sockets, hence allowing one to delegate this group to
 281    individual users, or to run vtysh setgid to this group.
 282
 283 The default user and group which will be configured is 'frr' if no user or
 284 group is specified. Note that this user or group requires write access to the
 285 local state directory (see :option:`--localstatedir`) and requires at least
 286 read access, and write access if you wish to allow daemons to write out their
 287 configuration, to the configuration directory (see :option:`--sysconfdir`).
 288
 289 On systems which have the 'libcap' capabilities manipulation library (currently
 290 only Linux), FRR will retain only minimal capabilities required and will only
 291 raise these capabilities for brief periods. On systems without libcap, FRR will
 292 run as the user specified and only raise its UID to 0 for brief periods.
 293
 294 Linux Notes
 295 """""""""""
 296
 297 .. index:: Building on Linux boxes
 298 .. index:: Linux configurations
 299
 300 There are several options available only to GNU/Linux systems.  If you use
 301 GNU/Linux, make sure that the current kernel configuration is what you want.
 302 FRR will run with any kernel configuration but some recommendations do exist.
 303
 304 :makevar:`CONFIG_NETLINK`
 305    Kernel/User Netlink socket. This is a enables an advanced interface between
 306    the Linux kernel and *zebra* (:ref:`kernel-interface`).
 307
 308 :makevar:`CONFIG_RTNETLINK`
 309    This makes it possible to receive Netlink routing messages.  If you specify
 310    this option, *zebra* can detect routing information updates directly from
 311    the kernel (:ref:`kernel-interface`).
 312
 313 :makevar:`CONFIG_IP_MULTICAST`
 314    This option enables IP multicast and should be specified when you use *ripd*
 315    (:ref:`rip`) or *ospfd* (:ref:`ospfv2`) because these protocols use
 316    multicast.
 317
 318 Linux sysctl settings and kernel modules
 319 ````````````````````````````````````````
 320
 321 There are several kernel parameters that impact overall operation of FRR when
 322 using Linux as a router. Generally these parameters should be set in a
 323 sysctl related configuration file, e.g., :file:`/etc/sysctl.conf` on
 324 Ubuntu based systems and a new file
 325 :file:`/etc/sysctl.d/90-routing-sysctl.conf` on Centos based systems.
 326 Additional kernel modules are also needed to support MPLS forwarding.
 327
 328 :makevar:`IPv4 and IPv6 forwarding`
 329    The following are set to enable IP forwarding in the kernel:
 330
 331    .. code-block:: shell
 332
 333       net.ipv4.conf.all.forwarding=1
 334       net.ipv6.conf.all.forwarding=1
 335
 336 :makevar:`MPLS forwarding`
 337    Basic MPLS kernel support was introduced 4.1, additional capability
 338    was introduced in 4.3 and 4.5. For some general information on Linux
 339    MPLS support see
 340    https://www.netdevconf.org/1.1/proceedings/slides/prabhu-mpls-tutorial.pdf.
 341    The following modules should be loaded to support MPLS forwarding,
 342    and are generally added to a configuration file such as
 343    :file:`/etc/modules-load.d/modules.conf`:
 344
 345    .. code-block:: shell
 346
 347       # Load MPLS Kernel Modules
 348       mpls_router
 349       mpls_iptunnel
 350
 351    The following is an example to enable MPLS forwarding in the kernel:
 352
 353    .. code-block:: shell
 354
 355       # Enable MPLS Label processing on all interfaces
 356       net.mpls.conf.eth0.input=1
 357       net.mpls.conf.eth1.input=1
 358       net.mpls.conf.eth2.input=1
 359       net.mpls.platform_labels=100000
 360
 361    Make sure to add a line equal to :file:`net.mpls.conf.<if>.input` for
 362    each interface *'<if>'* used with MPLS and to set labels to an
 363    appropriate value.
 364
 365 :makevar:`VRF forwarding`
 366    General information on Linux VRF support can be found in
 367    https://www.kernel.org/doc/Documentation/networking/vrf.txt. Kernel
 368    support for VRFs was introduced in 4.3 and improved upon through
 369    4.13, which is the version most used in FRR testing (as of June
 370    2018).  Additional background on using Linux VRFs and kernel specific
 371    features can be found in
 372    http://schd.ws/hosted_files/ossna2017/fe/vrf-tutorial-oss.pdf.
 373
 374    The following impacts how BGP TCP sockets are managed across VRFs:
 375
 376    .. code-block:: shell
 377
 378       net.ipv4.tcp_l3mdev_accept=0
 379
 380    With this setting a BGP TCP socket is opened per VRF.  This setting
 381    ensures that other TCP services, such as SSH, provided for non-VRF
 382    purposes are blocked from VRF associated Linux interfaces.
 383
 384    .. code-block:: shell
 385
 386       net.ipv4.tcp_l3mdev_accept=1
 387
 388    With this setting a single BGP TCP socket is shared across the
 389    system.  This setting exposes any TCP service running on the system,
 390    e.g., SSH, to all VRFs.  Generally this setting is not used in
 391    environments where VRFs are used to support multiple administrative
 392    groups.
 393
 394    **Important note** as of June 2018, Kernel versions 4.14-4.18 have a
 395    known bug where VRF-specific TCP sockets are not properly handled. When
 396    running these kernel versions, if unable to establish any VRF BGP
 397    adjacencies, either downgrade to 4.13 or set
 398    'net.ipv4.tcp_l3mdev_accept=1'. The fix for this issue is planned to be
 399    included in future kernel versions so upgrading your kernel may also
 400    address this issue.
 401
 402
 403 Building
 404 ^^^^^^^^
 405
 406 Once you have chosen your configure options, run the configure script and pass
 407 the options you chose:
 408
 409 .. code-block:: shell
 410
 411    ./configure \
 412        --prefix=/usr \
 413        --enable-exampledir=/usr/share/doc/frr/examples/ \
 414        --localstatedir=/var/run/frr \
 415        --sbindir=/usr/lib/frr \
 416        --sysconfdir=/etc/frr \
 417        --enable-pimd \
 418        --enable-watchfrr \
 419        ...
 420
 421 After configuring the software, you are ready to build and install it for your
 422 system.
 423
 424 .. code-block:: shell
 425
 426    make && sudo make install
 427
 428 If everything finishes successfully, FRR should be installed. You should now
 429 skip to the section on :ref:`basic-setup`.