Skip to content

Latest commit

 

History

History
105 lines (83 loc) · 7.73 KB

traceroute-网络路由跟踪.md

File metadata and controls

105 lines (83 loc) · 7.73 KB

Traceroute command

The traceroute command attempts to trace the route of IP packets to a particular Internet host by initiating probing packets with a small time-to-live (ttl) and listening for ICMP timeout replies from gateways. It starts with a ttl of 1 and increments it until it either receives an ICMP port unreachable or TCP reset, which indicates that the host has been reached or the maximum number of hops (default of 30 hops) has been reached. It sends three probes (by default) at each ttl and prints a line showing the ttl, gateway address, and round-trip time for each probe. Additional information can be provided after the address in the request. If responses come from different gateways, it prints the address of each responding system. If there is no response within 5.0 seconds (default value), it prints a * for that probe.

Syntax

traceroute [-46dFITUnreAV] [-f first_ttl] [-g gate,...] [-i device] 
           [-m max_ttl] [-p port] [-s src_addr] [-q nqueries] 
           [-N squeries] [-t tos] [-l flow_label] [-w waittime] 
           [-z sendwait] [-UL] [-D] [-P proto] [--sport=port] [-M method] 
           [-O mod_options] [--mtu] [--back] host [packet_len]

Parameters

  • -4, -6: Explicitly force IPv4 or IPv6 traceroute, by default, the program will attempt to resolve the given name and automatically select the appropriate protocol, if resolving the hostname returns both IPv4 and IPv6 addresses, traceroute will use IPv4.
  • -I, --icmp: Use ICMP ECHO for probing.
  • -T, --tcp: Use TCP SYN for probing.
  • -d, --debug: Enable socket-level debugging, if supported by the kernel.
  • -F, --dont-fragment: Do not fragment the probing packets, for IPv4 it also sets the DF bit, which instructs intermediate routers not to perform remote fragmentation. Change the size of the probing packets via the packet_len command parameter to manually obtain information about the MTU of individual network hops. Starting from Linux kernel 2.6.22, the non-fragmentation feature (e.g., -F or --mtu) can function properly. Before this version, IPv6 was always fragmented and IPv4 could only use the final mtu discovered once (from the route cache), which may be smaller than the actual mtu of the device.
  • -f first_ttl, --first=first_ttl: Specify the starting TTL, defaults to 1.
  • -g gate,..., --gateway=gate,...: Instruct traceroute to add the IP source route option to outgoing packets, which tells the network to route packets through the specified gateway (for security reasons, most routers have disabled source routing), multiple gateways (comma-separated list) are usually allowed. For IPv6, it is possible to use the form num, addr, addr, ..., where num is the route header type (default is type 2), note that according to rfc 5095, the use of route type 0 is now deprecated.
  • -i device, --interface=device: Specify the interface through which traceroute should send its packets, by default, the interface is chosen based on the routing table.
  • -m max_ttl, --max-hops=max_ttl: Specify the maximum number of hops traceroute should probe (maximum time to live value), default is 30.
  • -N squeries, --sim-queries=squeries: Specify the number of probe messages to be sent simultaneously, sending multiple probes can greatly speed up route tracing, the default value is 16, note that some routers and hosts may employ ICMP rate limiting, in which case specifying a large number may result in some responses being lost.
  • -n: Do not attempt to map the IP addresses to host names when displaying them.
  • -p port, --port=port: For UDP tracing, specify the destination port number traceroute will use, the target port number will increment with each probe, for ICMP tracing, specify the initial ICMP sequence value (also incremented per probe), for TCP and other protocols, only specify the (constant) target port to connect to, when using the tcptraceroute wrapper program, the -p specifies the source port.
  • -t tos, --tos=tos: For IPv4, set the type of service TOS and priority value, useful values are 16 for low delay and 8 for high throughput, to use certain TOS priority values, superuser permission is required, for IPv6, set the traffic class value.
  • -l flow_label, --flowlabel=flow_label: Use the specified flow_label for IPv6 packets.
  • -w MAX,HERE,NEAR, --wait=MAX,HERE,NEAR: Set the time to wait for probe responses, in seconds, default is 5.0.
  • -q nqueries, --queries=nqueries: Set the number of probe packets per hop, default is 3.
  • -r: Bypass the regular routing table and send directly to a host on the connected network, return an error if the host is not on the directly connected network, this option can be used to perform a ping operation on the local host through an interface without a route.
  • -s src_addr, --source=src_addr: Select an alternate source address, note that an address of an interface must be selected, by default, the address of the outgoing interface is used.
  • -z sendwait, --sendwait=sendwait: Minimum time interval between probes, default is 0, if the value is greater than 10, specify a number in milliseconds, otherwise it is in seconds, floating-point values are also allowed, very useful when some routers employ rate limiting for ICMP messages.
  • -e, --extensions: Display ICMP extension names, the universal format is CLASS / TYPE followed by a hexadecimal dump, indicated MPLS multi-protocol label switching data is parsed, in the format MPLS:L=label,E=exp_use,S=stack_bottom,T=TTL (with any further objects separated by a slash ("/")).
  • -A, --as-path-lookups : Perform AS path lookups in the routing registry and print the result directly after the corresponding address.
  • -M name --module=name: Use the specified module (built-in or external) for traceroute operations, most methods have their shortcuts, for example, -I represents -M icmp, etc.
  • -O OPTS,..., --options=OPTS,...: Use specific module options OPTS for the traceroute module, allow several OPTS, separated by commas, for example, if OPTS is help, print the help information.
  • --sport=num: Use the source port number for outgoing packets, indicated as -N 1.
  • --fwmark=num: Set the firewall mark for outgoing packets.
  • -U --udp: Use UDP for routing to a specific port, instead of increasing the port for each probe, the default port is 53.
  • -UL: Use UDP LITE for routing, default target port is 53.
  • -D --dccp: Use DCCP requests for routing, default port is 33434.
  • -P prot --protocol=prot: Use protocol-protected raw packets for routing.
  • --mtu: Discover MTU along the traced path, indicated as -F-N 1.
  • --back: Infer hops on the backward path, print if different.
  • -V: Output version information.
  • --help: Output help information.

Example

Use traceroute to view routing information.

traceroute www.google.com

Specify IPv4 to view routing information.

traceroute -4 www.google.com

Specify the TTL to start with, default is 1.

traceroute -f 3 www.google.com

Do not resolve IP addresses to their domain names.

traceroute -n www.google.com

Set the number of probes per hop, default is 3.

traceroute -q 1 www.google.com

Specify the complete packet length, default is a 60 byte packet.

traceroute www.google.com 100

Set the target port to be used, default is 33434.

traceroute -p 20292 www.google.com

Daily Question

https://github.com/WindrunnerMax/EveryDay

References

https://www.computerhope.com/unix/utracero.htm
https://www.runoob.com/linux/linux-comm-traceroute.html
https://www.geeksforgeeks.org/traceroute-command-in-linux-with-examples/