1<refentry id="ping"> 2 3<refmeta> 4<refentrytitle>ping</refentrytitle> 5<manvolnum>8</manvolnum> 6<refmiscinfo>iputils-&snapshot;</refmiscinfo> 7</refmeta> 8 9<refnamediv> 10<refname>ping, ping6</refname> 11<refpurpose>send ICMP ECHO_REQUEST to network hosts</refpurpose> 12</refnamediv> 13 14<refsynopsisdiv> 15<cmdsynopsis> 16<command>ping</command> 17<arg choice="opt"><option>-aAbBdDfhLnOqrRUvV</option></arg> 18<arg choice="opt">-c <replaceable/count/</arg> 19<arg choice="opt">-F <replaceable/flowlabel/</arg> 20<arg choice="opt">-i <replaceable/interval/</arg> 21<arg choice="opt">-I <replaceable/interface/</arg> 22<arg choice="opt">-l <replaceable/preload/</arg> 23<arg choice="opt">-m <replaceable/mark/</arg> 24<arg choice="opt">-M <replaceable/pmtudisc_option/</arg> 25<arg choice="opt">-N <replaceable/nodeinfo_option/</arg> 26<arg choice="opt">-w <replaceable/deadline/</arg> 27<arg choice="opt">-W <replaceable/timeout/</arg> 28<arg choice="opt">-p <replaceable/pattern/</arg> 29<arg choice="opt">-Q <replaceable/tos/</arg> 30<arg choice="opt">-s <replaceable/packetsize/</arg> 31<arg choice="opt">-S <replaceable/sndbuf/</arg> 32<arg choice="opt">-t <replaceable/ttl/</arg> 33<arg choice="opt">-T <replaceable/timestamp option/</arg> 34<arg choice="opt" rep="repeat"><replaceable/hop/</arg> 35<arg choice="req"><replaceable/destination/</arg> 36</cmdsynopsis> 37</refsynopsisdiv> 38 39<refsect1><title>DESCRIPTION</title> 40<para> 41<command/ping/ uses the ICMP protocol's mandatory ECHO_REQUEST 42datagram to elicit an ICMP ECHO_RESPONSE from a host or gateway. 43ECHO_REQUEST datagrams (``pings'') have an IP and ICMP 44header, followed by a <structname/struct timeval/ and then an arbitrary 45number of ``pad'' bytes used to fill out the packet. 46</para> 47<para> 48<command/ping6/ is IPv6 version of <command/ping/, and can also send Node Information Queries (RFC4620). 49Intermediate <replaceable/hop/s may not be allowed, because IPv6 source routing was deprecated (RFC5095). 50</para> 51</refsect1> 52 53<refsect1><title>OPTIONS</title> 54 55<variablelist> 56 <varlistentry> 57 <term><option/-a/</term> 58 <listitem><para> 59Audible ping. 60 </para></listitem> 61 </varlistentry> 62 <varlistentry> 63 <term><option/-A/</term> 64 <listitem><para> 65Adaptive ping. Interpacket interval adapts to round-trip time, so that 66effectively not more than one (or more, if preload is set) unanswered probe 67is present in the network. Minimal interval is 200msec for not super-user. 68On networks with low rtt this mode is essentially equivalent to flood mode. 69 </para></listitem> 70 </varlistentry> 71 <varlistentry> 72 <term><option/-b/</term> 73 <listitem><para> 74Allow pinging a broadcast address. 75 </para></listitem> 76 </varlistentry> 77 <varlistentry> 78 <term><option/-B/</term> 79 <listitem><para> 80Do not allow <command/ping/ to change source address of probes. 81The address is bound to one selected when <command/ping/ starts. 82 </para></listitem> 83 </varlistentry> 84 <varlistentry> 85 <term><option><anchor id="ping.count">-c <replaceable/count/</option></term> 86 <listitem><para> 87Stop after sending <replaceable/count/ ECHO_REQUEST 88packets. With 89<link linkend="ping.deadline"><replaceable/deadline/</link> 90option, <command/ping/ waits for 91<replaceable/count/ ECHO_REPLY packets, until the timeout expires. 92 </para></listitem> 93 </varlistentry> 94 <varlistentry> 95 <term><option/-d/</term> 96 <listitem><para> 97Set the <constant/SO_DEBUG/ option on the socket being used. 98Essentially, this socket option is not used by Linux kernel. 99 </para></listitem> 100 </varlistentry> 101 <varlistentry> 102 <term><option/-D/</term> 103 <listitem><para> 104Print timestamp (unix time + microseconds as in gettimeofday) before 105each line. 106 </para></listitem> 107 </varlistentry> 108 <varlistentry> 109 <term><option/-f/</term> 110 <listitem><para> 111Flood ping. For every ECHO_REQUEST sent a period ``.'' is printed, 112while for ever ECHO_REPLY received a backspace is printed. 113This provides a rapid display of how many packets are being dropped. 114If interval is not given, it sets interval to zero and 115outputs packets as fast as they come back or one hundred times per second, 116whichever is more. 117Only the super-user may use this option with zero interval. 118 </para></listitem> 119 </varlistentry> 120 <varlistentry> 121 <term><option>-F <replaceable/flow label/</option></term> 122 <listitem><para> 123<command/ping6/ only. 124Allocate and set 20 bit flow label (in hex) on echo request packets. 125If value is zero, kernel allocates random flow label. 126 </para></listitem> 127 </varlistentry> 128 <varlistentry> 129 <term><option/-h/</term> 130 <listitem><para> 131Show help. 132 </para></listitem> 133 </varlistentry> 134 <varlistentry> 135 <term><option>-i <replaceable/interval/</option></term> 136 <listitem><para> 137Wait <replaceable/interval/ seconds between sending each packet. 138The default is to wait for one second between each packet normally, 139or not to wait in flood mode. Only super-user may set interval 140to values less 0.2 seconds. 141 </para></listitem> 142 </varlistentry> 143 <varlistentry> 144 <term><option>-I <replaceable/interface/</option></term> 145 <listitem><para> 146<replaceable/interface/ is either an address, or an interface name. 147If <replaceable/interface/ is an address, it sets source address 148to specified interface address. 149If <replaceable/interface/ in an interface name, it sets 150source interface to specified interface. 151For <command/ping6/, when doing ping to a link-local scope 152address, link specification (by the '%'-notation in 153<replaceable/destination/, or by this option) is required. 154 </para></listitem> 155 </varlistentry> 156 <varlistentry> 157 <term><option>-l <replaceable/preload/</option></term> 158 <listitem><para> 159If <replaceable/preload/ is specified, 160<command/ping/ sends that many packets not waiting for reply. 161Only the super-user may select preload more than 3. 162 </para></listitem> 163 </varlistentry> 164 <varlistentry> 165 <term><option/-L/</term> 166 <listitem><para> 167Suppress loopback of multicast packets. This flag only applies if the ping 168destination is a multicast address. 169 </para></listitem> 170 </varlistentry> 171 <varlistentry> 172 <term><option>-m <replaceable/mark/</option></term> 173 <listitem><para> 174use <replaceable/mark/ to tag the packets going out. This is useful 175for variety of reasons within the kernel such as using policy 176routing to select specific outbound processing. 177 </para></listitem> 178 </varlistentry> 179 <varlistentry> 180 <term><option>-M <replaceable/pmtudisc_opt/</option></term> 181 <listitem><para> 182Select Path MTU Discovery strategy. 183<replaceable/pmtudisc_option/ may be either <replaceable/do/ 184(prohibit fragmentation, even local one), 185<replaceable/want/ (do PMTU discovery, fragment locally when packet size 186is large), or <replaceable/dont/ (do not set DF flag). 187 </para></listitem> 188 </varlistentry> 189 <varlistentry> 190 <term><option>-N <replaceable/nodeinfo_option/</option></term> 191 <listitem><para> 192<command/ping6/ only. 193Send ICMPv6 Node Information Queries (RFC4620), instead of Echo Request. 194 <variablelist> 195 <varlistentry> 196 <term><option>help</option></term> 197 <listitem><para>Show help for NI support.</para></listitem> 198 </varlistentry> 199 </variablelist> 200 <variablelist> 201 <varlistentry> 202 <term><option>name</option></term> 203 <listitem><para>Queries for Node Names.</para></listitem> 204 </varlistentry> 205 </variablelist> 206 <variablelist> 207 <varlistentry> 208 <term><option>ipv6</option></term> 209 <listitem><para>Queries for IPv6 Addresses. There are several IPv6 specific flags. 210 <variablelist> 211 <varlistentry> 212 <term><option>ipv6-global</option></term> 213 <listitem><para>Request IPv6 global-scope addresses.</para></listitem> 214 </varlistentry> 215 </variablelist> 216 <variablelist> 217 <varlistentry> 218 <term><option>ipv6-sitelocal</option></term> 219 <listitem><para>Request IPv6 site-local addresses.</para></listitem> 220 </varlistentry> 221 </variablelist> 222 <variablelist> 223 <varlistentry> 224 <term><option>ipv6-linklocal</option></term> 225 <listitem><para>Request IPv6 link-local addresses.</para></listitem> 226 </varlistentry> 227 </variablelist> 228 <variablelist> 229 <varlistentry> 230 <term><option>ipv6-all</option></term> 231 <listitem><para>Request IPv6 addresses on other interfaces.</para></listitem> 232 </varlistentry> 233 </variablelist> 234 </para></listitem> 235 </varlistentry> 236 </variablelist> 237 <variablelist> 238 <varlistentry> 239 <term><option>ipv4</option></term> 240 <listitem><para>Queries for IPv4 Addresses. There is one IPv4 specific flag. 241 <variablelist> 242 <varlistentry> 243 <term><option>ipv4-all</option></term> 244 <listitem><para>Request IPv4 addresses on other interfaces.</para></listitem> 245 </varlistentry> 246 </variablelist> 247 </para></listitem> 248 </varlistentry> 249 </variablelist> 250 <variablelist> 251 <varlistentry> 252 <term><option>subject-ipv6=<replaceable/ipv6addr/</option></term> 253 <listitem><para>IPv6 subject address.</para></listitem> 254 </varlistentry> 255 </variablelist> 256 <variablelist> 257 <varlistentry> 258 <term><option>subject-ipv4=<replaceable/ipv4addr/</option></term> 259 <listitem><para>IPv4 subject address.</para></listitem> 260 </varlistentry> 261 </variablelist> 262 <variablelist> 263 <varlistentry> 264 <term><option>subject-name=<replaceable/nodename/</option></term> 265 <listitem><para>Subject name. If it contains more than one dot, 266 fully-qualified domain name is assumed.</para></listitem> 267 </varlistentry> 268 </variablelist> 269 <variablelist> 270 <varlistentry> 271 <term><option>subject-fqdn=<replaceable/nodename/</option></term> 272 <listitem><para>Subject name. Fully-qualified domain name is 273 always assumed.</para></listitem> 274 </varlistentry> 275 </variablelist> 276 </para></listitem> 277 </varlistentry> 278 <varlistentry> 279 <term><option/-n/</term> 280 <listitem><para> 281Numeric output only. 282No attempt will be made to lookup symbolic names for host addresses. 283 </para></listitem> 284 </varlistentry> 285 <varlistentry> 286 <term><option/-O/</term> 287 <listitem><para> 288Report outstanding ICMP ECHO reply before sending next packet. 289This is useful together with the timestamp <option>-D</option> to 290log output to a diagnostic file and search for missing answers. 291 </para></listitem> 292 </varlistentry> 293 <varlistentry> 294 <term><option>-p <replaceable/pattern/</option></term> 295 <listitem><para> 296You may specify up to 16 ``pad'' bytes to fill out the packet you send. 297This is useful for diagnosing data-dependent problems in a network. 298For example, <option>-p ff</option> will cause the sent packet 299to be filled with all ones. 300 </para></listitem> 301 </varlistentry> 302 <varlistentry> 303 <term><option/-q/</term> 304 <listitem><para> 305Quiet output. 306Nothing is displayed except the summary lines at startup time and 307when finished. 308 </para></listitem> 309 </varlistentry> 310 <varlistentry> 311 <term><option>-Q <replaceable/tos/</option></term> 312 <listitem><para> 313 Set Quality of Service -related bits in ICMP datagrams. 314 <replaceable/tos/ can be decimal (<command/ping/ only) or hex number. 315 </para> 316 <para> 317 In RFC2474, these fields are interpreted as 8-bit Differentiated 318 Services (DS), consisting of: bits 0-1 (2 lowest bits) of separate 319 data, and bits 2-7 (highest 6 bits) of Differentiated Services 320 Codepoint (DSCP). In RFC2481 and RFC3168, bits 0-1 are used for ECN. 321 </para> 322 <para> 323 Historically (RFC1349, obsoleted by RFC2474), these were interpreted 324 as: bit 0 (lowest bit) for reserved (currently being redefined as 325 congestion control), 1-4 for Type of Service and bits 5-7 326 (highest bits) for Precedence. 327 </para> 328 </listitem> 329 </varlistentry> 330 <varlistentry> 331 <term><option/-r/</term> 332 <listitem><para> 333Bypass the normal routing tables and send directly to a host on an attached 334interface. 335If the host is not on a directly-attached network, an error is returned. 336This option can be used to ping a local host through an interface 337that has no route through it provided the option <option/-I/ is also 338used. 339 </para></listitem> 340 </varlistentry> 341 <varlistentry> 342 <term><option/-R/</term> 343 <listitem><para> 344<command/ping/ only. 345Record route. 346Includes the RECORD_ROUTE option in the ECHO_REQUEST 347packet and displays the route buffer on returned packets. 348Note that the IP header is only large enough for nine such routes. 349Many hosts ignore or discard this option. 350 </para></listitem> 351 </varlistentry> 352 <varlistentry> 353 <term><option>-s <replaceable/packetsize/</option></term> 354 <listitem><para> 355Specifies the number of data bytes to be sent. 356The default is 56, which translates into 64 ICMP 357data bytes when combined with the 8 bytes of ICMP header data. 358 </para></listitem> 359 </varlistentry> 360 <varlistentry> 361 <term><option>-S <replaceable/sndbuf/</option></term> 362 <listitem><para> 363Set socket sndbuf. If not specified, it is selected to buffer 364not more than one packet. 365 </para></listitem> 366 </varlistentry> 367 <varlistentry> 368 <term><option>-t <replaceable/ttl/</option></term> 369 <listitem><para> 370<command/ping/ only. 371Set the IP Time to Live. 372 </para></listitem> 373 </varlistentry> 374 <varlistentry> 375 <term><option>-T <replaceable/timestamp option/</option></term> 376 <listitem><para> 377Set special IP timestamp options. 378<replaceable/timestamp option/ may be either 379<replaceable/tsonly/ (only timestamps), 380<replaceable/tsandaddr/ (timestamps and addresses) or 381<replaceable/tsprespec host1 [host2 [host3 [host4]]]/ 382(timestamp prespecified hops). 383 </para></listitem> 384 </varlistentry> 385 <varlistentry> 386 <term><option/-U/</term> 387 <listitem><para> 388Print full user-to-user latency (the old behaviour). Normally 389<command/ping/ 390prints network round trip time, which can be different 391f.e. due to DNS failures. 392 </para></listitem> 393 </varlistentry> 394 <varlistentry> 395 <term><option/-v/</term> 396 <listitem><para> 397Verbose output. 398 </para></listitem> 399 </varlistentry> 400 <varlistentry> 401 <term><option/-V/</term> 402 <listitem><para> 403Show version and exit. 404 </para></listitem> 405 </varlistentry> 406 <varlistentry> 407 <term><option><anchor id="ping.deadline">-w <replaceable/deadline/</option></term> 408 <listitem><para> 409Specify a timeout, in seconds, before 410<command/ping/ 411exits regardless of how many 412packets have been sent or received. In this case 413<command/ping/ 414does not stop after 415<link linkend="ping.count"><replaceable/count/</link> 416packet are sent, it waits either for 417<link linkend="ping.deadline"><replaceable/deadline/</link> 418expire or until 419<link linkend="ping.count"><replaceable/count/</link> 420probes are answered or for some error notification from network. 421 </para></listitem> 422 </varlistentry> 423 <varlistentry> 424 <term><option>-W <replaceable/timeout/</option></term> 425 <listitem><para> 426Time to wait for a response, in seconds. The option affects only timeout 427in absence of any responses, otherwise <command/ping/ waits for two RTTs. 428 </para></listitem> 429 </varlistentry> 430</variablelist> 431 432<para> 433When using <command/ping/ for fault isolation, it should first be run 434on the local host, to verify that the local network interface is up 435and running. Then, hosts and gateways further and further away should be 436``pinged''. Round-trip times and packet loss statistics are computed. 437If duplicate packets are received, they are not included in the packet 438loss calculation, although the round trip time of these packets is used 439in calculating the minimum/average/maximum round-trip time numbers. 440When the specified number of packets have been sent (and received) or 441if the program is terminated with a 442<constant/SIGINT/, a brief summary is displayed. Shorter current statistics 443can be obtained without termination of process with signal 444<constant/SIGQUIT/. 445</para> 446 447<para> 448If <command/ping/ does not receive any reply packets at all it will 449exit with code 1. If a packet 450<link linkend="ping.count"><replaceable/count/</link> 451and 452<link linkend="ping.deadline"><replaceable/deadline/</link> 453are both specified, and fewer than 454<link linkend="ping.count"><replaceable/count/</link> 455packets are received by the time the 456<link linkend="ping.deadline"><replaceable/deadline/</link> 457has arrived, it will also exit with code 1. 458On other error it exits with code 2. Otherwise it exits with code 0. This 459makes it possible to use the exit code to see if a host is alive or 460not. 461</para> 462 463 464<para> 465This program is intended for use in network testing, measurement and 466management. 467Because of the load it can impose on the network, it is unwise to use 468<command/ping/ during normal operations or from automated scripts. 469</para> 470 471</refsect1> 472 473 474<refsect1><title>ICMP PACKET DETAILS</title> 475 476<para> 477An IP header without options is 20 bytes. 478An ICMP ECHO_REQUEST packet contains an additional 8 bytes worth 479of ICMP header followed by an arbitrary amount of data. 480When a <replaceable/packetsize/ is given, this indicated the size of this 481extra piece of data (the default is 56). Thus the amount of data received 482inside of an IP packet of type ICMP ECHO_REPLY will always be 8 bytes 483more than the requested data space (the ICMP header). 484</para> 485 486<para> 487If the data space is at least of size of <structname/struct timeval/ 488<command/ping/ uses the beginning bytes of this space to include 489a timestamp which it uses in the computation of round trip times. 490If the data space is shorter, no round trip times are given. 491</para> 492 493</refsect1> 494 495<refsect1><title>DUPLICATE AND DAMAGED PACKETS</title> 496 497<para> 498<command/ping/ will report duplicate and damaged packets. 499Duplicate packets should never occur, and seem to be caused by 500inappropriate link-level retransmissions. 501Duplicates may occur in many situations and are rarely (if ever) a 502good sign, although the presence of low levels of duplicates may not 503always be cause for alarm. 504</para> 505 506<para> 507Damaged packets are obviously serious cause for alarm and often 508indicate broken hardware somewhere in the 509<command/ping/ packet's path (in the network or in the hosts). 510</para> 511 512</refsect1> 513 514<refsect1><title>TRYING DIFFERENT DATA PATTERNS</title> 515 516<para> 517The (inter)network layer should never treat packets differently depending 518on the data contained in the data portion. 519Unfortunately, data-dependent problems have been known to sneak into 520networks and remain undetected for long periods of time. 521In many cases the particular pattern that will have problems is something 522that doesn't have sufficient ``transitions'', such as all ones or all 523zeros, or a pattern right at the edge, such as almost all zeros. 524It isn't necessarily enough to specify a data pattern of all zeros (for 525example) on the command line because the pattern that is of interest is 526at the data link level, and the relationship between what you type and 527what the controllers transmit can be complicated. 528</para> 529 530<para> 531This means that if you have a data-dependent problem you will probably 532have to do a lot of testing to find it. 533If you are lucky, you may manage to find a file that either can't be sent 534across your network or that takes much longer to transfer than other 535similar length files. 536You can then examine this file for repeated patterns that you can test 537using the <option/-p/ option of <command/ping/. 538</para> 539 540</refsect1> 541 542<refsect1><title>TTL DETAILS</title> 543 544<para> 545The TTL value of an IP packet represents the maximum number of IP routers 546that the packet can go through before being thrown away. 547In current practice you can expect each router in the Internet to decrement 548the TTL field by exactly one. 549</para> 550 551<para> 552The TCP/IP specification states that the TTL field for TCP 553packets should be set to 60, but many systems use smaller values 554(4.3 BSD uses 30, 4.2 used 15). 555</para> 556 557<para> 558The maximum possible value of this field is 255, and most Unix systems set 559the TTL field of ICMP ECHO_REQUEST packets to 255. 560This is why you will find you can ``ping'' some hosts, but not reach them 561with 562<citerefentry><refentrytitle/telnet/<manvolnum/1/</citerefentry> 563or 564<citerefentry><refentrytitle/ftp/<manvolnum/1/</citerefentry>. 565</para> 566 567<para> 568In normal operation ping prints the TTL value from the packet it receives. 569When a remote system receives a ping packet, it can do one of three things 570with the TTL field in its response: 571</para> 572 573<itemizedlist> 574 <listitem><para> 575Not change it; this is what Berkeley Unix systems did before the 5764.3BSD Tahoe release. In this case the TTL value in the received packet 577will be 255 minus the number of routers in the round-trip path. 578 </para></listitem> 579 <listitem><para> 580Set it to 255; this is what current Berkeley Unix systems do. 581In this case the TTL value in the received packet will be 255 minus the 582number of routers in the path <emphasis/from/ 583the remote system <emphasis/to/ the <command/ping/ing host. 584 </para></listitem> 585 <listitem><para> 586Set it to some other value. Some machines use the same value for 587ICMP packets that they use for TCP packets, for example either 30 or 60. 588Others may use completely wild values. 589 </para></listitem> 590</itemizedlist> 591 592</refsect1> 593 594<refsect1><title>BUGS</title> 595 596<itemizedlist> 597 <listitem><para> 598Many Hosts and Gateways ignore the RECORD_ROUTE option. 599 </para></listitem> 600 <listitem><para> 601The maximum IP header length is too small for options like 602RECORD_ROUTE to be completely useful. 603There's not much that can be done about this, however. 604 </para></listitem> 605 <listitem><para> 606Flood pinging is not recommended in general, and flood pinging the 607broadcast address should only be done under very controlled conditions. 608 </para></listitem> 609</itemizedlist> 610 611</refsect1> 612 613<refsect1><title>SEE ALSO</title> 614<para> 615<citerefentry><refentrytitle/netstat/<manvolnum/1/</citerefentry>, 616<citerefentry><refentrytitle/ifconfig/<manvolnum/8/</citerefentry>. 617</para> 618</refsect1> 619 620<refsect1><title>HISTORY</title> 621<para> 622The <command/ping/ command appeared in 4.3BSD. 623</para> 624<para> 625The version described here is its descendant specific to Linux. 626</para> 627</refsect1> 628 629<refsect1><title>SECURITY</title> 630<para> 631<command/ping/ requires <constant/CAP_NET_RAW/ capability 632to be executed. It may be used as set-uid root. 633</para> 634</refsect1> 635 636<refsect1><title>AVAILABILITY</title> 637<para> 638<command/ping/ is part of <filename/iputils/ package 639and the latest versions are available in source form at 640<ulink url="http://www.skbuff.net/iputils/iputils-current.tar.bz2"> 641http://www.skbuff.net/iputils/iputils-current.tar.bz2</ulink>. 642</para> 643</refsect1> 644 645<![IGNORE[ 646<refsect1><title>COPYING</title> 647<para> 648<literallayout> 649Copyright (c) 1989 The Regents of the University of California. 650All rights reserved. 651 652This code is derived from software contributed to Berkeley by 653Mike Muuss. 654 655Redistribution and use in source and binary forms, with or without 656modification, are permitted provided that the following conditions 657are met: 6581. Redistributions of source code must retain the above copyright 659 notice, this list of conditions and the following disclaimer. 6602. Redistributions in binary form must reproduce the above copyright 661 notice, this list of conditions and the following disclaimer in the 662 documentation and/or other materials provided with the distribution. 6633. All advertising materials mentioning features or use of this software 664 must display the following acknowledgement: 665 This product includes software developed by the University of 666 California, Berkeley and its contributors. 6674. Neither the name of the University nor the names of its contributors 668 may be used to endorse or promote products derived from this software 669 without specific prior written permission. 670 671THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 672ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 673IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 674ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 675FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 676DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 677OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 678HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 679LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 680OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 681SUCH DAMAGE. 682</literallayout> 683</para> 684</refsect1> 685]]> 686 687 688</refentry> 689 690