1PING(8) System Manager's Manual: iputils PING(8)
2
6 ping, ping6 - send ICMP ECHO_REQUEST to network hosts
7
9 ping [ -LRUbdfnqrvVaAB] [ -c count] [ -i interval] [ -l preload] [
10 -p pattern] [ -s packetsize] [ -t ttl] [ -w deadline] [ -F flowla‐
11 bel] [ -I interface] [ -M hint] [ -Q tos] [ -S sndbuf] [ -T time‐
12 stamp option] [ -W timeout] [ hop ...] destination
13
16 ping uses the ICMP protocol's mandatory ECHO_REQUEST datagram to elicit
17 an ICMP ECHO_RESPONSE from a host or gateway. ECHO_REQUEST datagrams
18 (``pings'') have an IP and ICMP header, followed by a struct timeval
19 and then an arbitrary number of ``pad'' bytes used to fill out the
20 packet.
21
23 -a Audible ping.
24 -A Adaptive ping. Interpacket interval adapts to round-trip time,
26 so that effectively not more than one (or more, if preload is
27 set) unanswered probes present in the network. Minimal interval
28 is 200msec for not super-user. On networks with low rtt this
29 mode is essentially equivalent to flood mode.
30 -b Allow pinging a broadcast address.
32 -B Do not allow ping to change source address of probes. The
34 address is bound to one selected when ping starts.
35 -c count
37 Stop after sending count ECHO_REQUEST packets. With deadline
38 option, ping waits for count ECHO_REPLY packets, until the time‐
39 out expires.
40 -d Set the SO_DEBUG option on the socket being used. Essentially,
42 this socket option is not used by Linux kernel.
43 -F flow label
45 Allocate and set 20 bit flow label on echo request packets.
46 (Only ping6). If value is zero, kernel allocates random flow
47 label.
48 -f Flood ping. For every ECHO_REQUEST sent a period ``.'' is
50 printed, while for ever ECHO_REPLY received a backspace is
51 printed. This provides a rapid display of how many packets are
52 being dropped. If interval is not given, it sets interval to
53 zero and outputs packets as fast as they come back or one hun‐
54 dred times per second, whichever is more. Only the super-user
55 may use this option with zero interval.
56 -i interval
58 Wait interval seconds between sending each packet. The default
59 is to wait for one second between each packet normally, or not
60 to wait in flood mode. Only super-user may set interval to val‐
61 ues less 0.2 seconds.
62 -I interface address
64 Set source address to specified interface address. Argument may
65 be numeric IP address or name of device. When pinging IPv6 link-
66 local address this option is required.
67 -l preload
69 If preload is specified, ping sends that many packets not wait‐
70 ing for reply. Only the super-user may select preload more than
71 3.
72 -L Suppress loopback of multicast packets. This flag only applies
74 if the ping destination is a multicast address.
75 -n Numeric output only. No attempt will be made to lookup symbolic
77 names for host addresses.
78 -p pattern
80 You may specify up to 16 ``pad'' bytes to fill out the packet
81 you send. This is useful for diagnosing data-dependent problems
82 in a network. For example, -p ff will cause the sent packet to
83 be filled with all ones.
84 -Q tos Set Quality of Service -related bits in ICMP datagrams. tos can
86 be either decimal or hex number. Traditionally (RFC1349), these
87 have been interpreted as: 0 for reserved (currently being rede‐
88 fined as congestion control), 1-4 for Type of Service and 5-7
89 for Precedence. Possible settings for Type of Service are: min‐
90 imal cost: 0x02, reliability: 0x04, throughput: 0x08, low delay:
91 0x10. Multiple TOS bits should not be set simultaneously. Pos‐
92 sible settings for special Precedence range from priority (0x20)
93 to net control (0xe0). You must be root (CAP_NET_ADMIN capabil‐
94 ity) to use Critical or higher precedence value. You cannot set
95 bit 0x01 (reserved) unless ECN has been enabled in the kernel.
96 In RFC2474, these fields has been redefined as 8-bit Differenti‐
97 ated Services (DS), consisting of: bits 0-1 of separate data
98 (ECN will be used, here), and bits 2-7 of Differentiated Ser‐
99 vices Codepoint (DSCP).
100 -q Quiet output. Nothing is displayed except the summary lines at
102 startup time and when finished.
103 -R Record route. Includes the RECORD_ROUTE option in the
105 ECHO_REQUEST packet and displays the route buffer on returned
106 packets. Note that the IP header is only large enough for nine
107 such routes. Many hosts ignore or discard this option.
108 -r Bypass the normal routing tables and send directly to a host on
110 an attached interface. If the host is not on a directly-
111 attached network, an error is returned. This option can be used
112 to ping a local host through an interface that has no route
113 through it provided the option -I is also used.
114 -s packetsize
116 Specifies the number of data bytes to be sent. The default is
117 56, which translates into 64 ICMP data bytes when combined with
118 the 8 bytes of ICMP header data.
119 -S sndbuf
121 Set socket sndbuf. If not specified, it is selected to buffer
122 not more than one packet.
123 -t ttl Set the IP Time to Live.
125 -T timestamp option
127 Set special IP timestamp options. timestamp option may be
128 either tsonly (only timestamps), tsandaddr (timestamps and
129 addresses) or tsprespec host1 [host2 [host3 [host4]]] (timestamp
130 prespecified hops).
131 -M hint
133 Select Path MTU Discovery strategy. hint may be either do (pro‐
134 hibit fragmentation, even local one), want (do PMTU discovery,
135 fragment locally when packet size is large), or dont (do not set
136 DF flag).
137 -U Print full user-to-user latency (the old behaviour). Normally
139 ping prints network round trip time, which can be different f.e.
140 due to DNS failures.
141 -v Verbose output.
143 -V Show version and exit.
145 -w deadline
147 Specify a timeout, in seconds, before ping exits regardless of
148 how many packets have been sent or received. In this case ping
149 does not stop after count packet are sent, it waits either for
150 deadline expire or until count probes are answered or for some
151 error notification from network.
152 -W timeout
154 Time to wait for a response, in seconds. The option affects only
155 timeout in absense of any responses, otherwise ping waits for
156 two RTTs.
157 When using ping for fault isolation, it should first be run on the
159 local host, to verify that the local network interface is up and run‐
160 ning. Then, hosts and gateways further and further away should be
161 ``pinged''. Round-trip times and packet loss statistics are computed.
162 If duplicate packets are received, they are not included in the packet
163 loss calculation, although the round trip time of these packets is used
164 in calculating the minimum/average/maximum round-trip time numbers.
165 When the specified number of packets have been sent (and received) or
166 if the program is terminated with a SIGINT, a brief summary is dis‐
167 played. Shorter current statistics can be obtained without termination
168 of process with signal SIGQUIT.
169 If ping does not receive any reply packets at all it will exit with
171 code 1. If a packet count and deadline are both specified, and fewer
172 than count packets are received by the time the deadline has arrived,
173 it will also exit with code 1. On other error it exits with code 2.
174 Otherwise it exits with code 0. This makes it possible to use the exit
175 code to see if a host is alive or not.
176 This program is intended for use in network testing, measurement and
178 management. Because of the load it can impose on the network, it is
179 unwise to use ping during normal operations or from automated scripts.
180
182 An IP header without options is 20 bytes. An ICMP ECHO_REQUEST packet
183 contains an additional 8 bytes worth of ICMP header followed by an
184 arbitrary amount of data. When a packetsize is given, this indicated
185 the size of this extra piece of data (the default is 56). Thus the
186 amount of data received inside of an IP packet of type ICMP ECHO_REPLY
187 will always be 8 bytes more than the requested data space (the ICMP
188 header).
189 If the data space is at least of size of struct timeval ping uses the
191 beginning bytes of this space to include a timestamp which it uses in
192 the computation of round trip times. If the data space is shorter, no
193 round trip times are given.
194
196 ping will report duplicate and damaged packets. Duplicate packets
197 should never occur, and seem to be caused by inappropriate link-level
198 retransmissions. Duplicates may occur in many situations and are
199 rarely (if ever) a good sign, although the presence of low levels of
200 duplicates may not always be cause for alarm.
201 Damaged packets are obviously serious cause for alarm and often indi‐
203 cate broken hardware somewhere in the ping packet's path (in the net‐
204 work or in the hosts).
205
207 The (inter)network layer should never treat packets differently depend‐
208 ing on the data contained in the data portion. Unfortunately, data-
209 dependent problems have been known to sneak into networks and remain
210 undetected for long periods of time. In many cases the particular pat‐
211 tern that will have problems is something that doesn't have sufficient
212 ``transitions'', such as all ones or all zeros, or a pattern right at
213 the edge, such as almost all zeros. It isn't necessarily enough to
214 specify a data pattern of all zeros (for example) on the command line
215 because the pattern that is of interest is at the data link level, and
216 the relationship between what you type and what the controllers trans‐
217 mit can be complicated.
218 This means that if you have a data-dependent problem you will probably
220 have to do a lot of testing to find it. If you are lucky, you may man‐
221 age to find a file that either can't be sent across your network or
222 that takes much longer to transfer than other similar length files.
223 You can then examine this file for repeated patterns that you can test
224 using the -p option of ping.
225
227 The TTL value of an IP packet represents the maximum number of IP
228 routers that the packet can go through before being thrown away. In
229 current practice you can expect each router in the Internet to decre‐
230 ment the TTL field by exactly one.
231 The TCP/IP specification states that the TTL field for TCP packets
233 should be set to 60, but many systems use smaller values (4.3 BSD uses
234 30, 4.2 used 15).
235 The maximum possible value of this field is 255, and most Unix systems
237 set the TTL field of ICMP ECHO_REQUEST packets to 255. This is why you
238 will find you can ``ping'' some hosts, but not reach them with tel‐
239 net(1) or ftp(1).
240 In normal operation ping prints the ttl value from the packet it
242 receives. When a remote system receives a ping packet, it can do one
243 of three things with the TTL field in its response:
244 · Not change it; this is what Berkeley Unix systems did before the
246 4.3BSD Tahoe release. In this case the TTL value in the received
247 packet will be 255 minus the number of routers in the round-trip
248 path.
249 · Set it to 255; this is what current Berkeley Unix systems do. In
251 this case the TTL value in the received packet will be 255 minus the
252 number of routers in the path from the remote system to the pinging
253 host.
254 · Set it to some other value. Some machines use the same value for ICMP
256 packets that they use for TCP packets, for example either 30 or 60.
257 Others may use completely wild values.
258
260 · Many Hosts and Gateways ignore the RECORD_ROUTE option.
261 · The maximum IP header length is too small for options like
263 RECORD_ROUTE to be completely useful. There's not much that that can
264 be done about this, however.
265 · Flood pinging is not recommended in general, and flood pinging the
267 broadcast address should only be done under very controlled condi‐
268 tions.
269
271 netstat(1), ifconfig(8).
272
274 The ping command appeared in 4.3BSD.
275 The version described here is its descendant specific to Linux.
277
279 ping requires CAP_NET_RAWIO capability to be executed. It may be used
280 as set-uid root.
281
283 ping is part of iputils package and the latest versions are available
284 in source form at http://www.skbuff.net/iputils/iputils-cur‐
285 rent.tar.bz2.
286iputils-070202 06 April 2007 PING(8)