SUMMARY: Value for TTL

From: Ricardo del Cueto <ricardo_at_r.iie.org.mx>
Date: Tue, 16 Jan 1996 13:44:34 -0600

Hi group:

I want to thank you all for the quick response.

The QUESTION was:

Is there a way I can find what the value of TTL is, particullary for telnet ?
Is there a dafault value for it ?. Is it possible to change it ? I am running
OSF/1 3.0.

Many thanks to:

Arnaud Valeix fnet_at_ifh.sncf.Fr
Pascal Pederiva pascal_at_zuo.dec.com
Arrigo Triulzi arrigo_at_lpac.ac.uk
Matt Thomas thomas_at_lkg.dec.com
Michael R. Kline mike_at_lib.utexas.edu

Thanks to this people I found that the default values for TCP TTL is 60
and for UDP is 30 (I am running OSF/1 3.0).

There are several ways to inspect the value for TTL (time-to-live):

1) In /usr/sys/include/netinet/inet_config.h there are two lines with the
   dafault values for TTL:

        int tcpttl; /* Default time to live (60) */
        int udpttl; /* Default time to live (30) */

2) Using the Netconfig utility which you can get in
   ftp://cs.ucsd.edu/pub/csl/Netconfig/netconfig2.3.tar.gz

   Once you build it, you type the command with no arguments and
   you get the following:

Sockets:
    Max socket buffer size (_sb_max): 131072
    Large mbuf use threshold (_mbufthreshold): 1024
    Max listen queue len (_somaxconn): 8
IP:
    Routing (_ipforwarding): 0
    Ipgateway (_ipgateway): off
ICMP:
    Redirects (_ipsendredirects): on
TCP:
    Output Buffer Size (_tcp_sendspace): 32768
    Input Buffer Size (_tcp_recvspace): 32768
    Retransmit Threshold (_tcprexmtthresh): 3
    Default Max Seg Size (_tcp_mssdflt): 536
    Keepalive Idle Time (_tcp_keepidle): 14400
    Keepalive Interval (_tcp_keepintvl): 150
    Default TTL (_tcp_ttl): 60
    Don't scale windows (_tcp_dont_winscale): off
    Subnets Are Local (_subnetsarelocal): on
UDP:
    Checksum (_udpcksum): on
    Max Send Msg Size (_udp_sendspace): 9216
    Max Rcv Buffer (_udp_recvspace): 41600
    Default TTL (_udp_ttl): 30
TCP/IP Misc.:
    Internet checksum routine active (_nocksum):off



3) The other way to examine tha values for TTL is by using the dbg.
   Type as root:

dbx -k /vmunix
(dbx) print tcp_ttl

 and you can also change the values by using something like:

(dbx) assign tcp_ttl = 128
(dbx) patch tcp_ttl = 130




Finally, I got this interesting comment about TTL:

>
>As the worldwide Internet grows the number of routers keeps
>increasing, which increases the total hop count (number of router
>"hops") a packet must transit to get to its destination.
>
>This can be a problem since the Time To Live (TTL) value in the packet
>is decremented at each router hop. If the TTL value is 32 and there
>are 40 routers in the path to the destination, then the packet will be
>discarded when the TTL reaches zero and won't make it to the
>destination.
>
>The following web page lists common TTL values in various operating
>systems along with tools that can be used to increase the size of the
>TTL value.
>
>http://www.switch.ch/switch/docs/ttl_default.html
>
>-Charles
>
>-
>Charles Spurgeon
>c.spurgeon_at_mail.utexas.edu
>512.475.9265
>
>

Anyone with interest about TTL values should take a look to the document:

http://www.switch.ch/switch/docs/ttl_default.html

This document has a comment about the default values for TTL in OSF/1, it says:

OSF/1 3.2A

  TCP TTL is 60 but UDP has too low TTL of 30. Modification of TCP/IP stack
  parameters can be made with netconfig2.3

-----
Received on Tue Jan 16 1996 - 21:11:48 NZDT

This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:46 NZDT