Hello,
In lib.core the MTU value that is queried from the ip utility
finds itself added a value of 100:
get_device_mtu1() # $1 = device
{
local output
output="$($IP link list dev $1 2> /dev/null)" # quotes required
for /bin/ash
local mtu
if [ -n "$output" ]; then
mtu=$(find_mtu $output)
if [ -n "$mtu" ]; then
[ $mtu = 1500 ] || echo mtu $(($mtu + 100))
fi
fi
}
What would be the reason as to why a value of 100 was chosen.
Many thanks.
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today''s security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
On 05/17/2012 08:05 AM, Fred Maillou wrote:> Hello, > > In lib.core the MTU value that is queried from the ip utility > finds itself added a value of 100: > > get_device_mtu1() # $1 = device > { > local output > output="$($IP link list dev $1 2> /dev/null)" # quotes required for /bin/ash > local mtu > > if [ -n "$output" ]; then > mtu=$(find_mtu $output) > if [ -n "$mtu" ]; then > [ $mtu = 1500 ] || echo mtu $(($mtu + 100)) > fi > fi > > } > > What would be the reason as to why a value of 100 was chosen.HTB assumes a value of 1600 (1500+100) if the HTB is not specified. So if the actual MTU is not 1500, I''m also adding 100 to the actual value. -Tom -- Tom Eastep \ When I die, I want to go like my Grandfather who Shoreline, \ died peacefully in his sleep. Not screaming like Washington, USA \ all of the passengers in his car http://shorewall.net \________________________________________________ ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today''s security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> HTB assumes a value of 1600 (1500+100) if the HTB is not > specified. So if the actual MTU is not 1500, I''m also adding > 100 to the actual value.Hmmmm... If not sure I get this ''if the HTB is not specified''. The HTB default is 1600. So if MTU 1500 is read then 100 is added to match, that seems OK. But of the MTU is 100, is there any need to add 100 ? Thanks. ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today''s security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
On 05/17/2012 11:13 AM, Fred Maillou wrote:>> HTB assumes a value of 1600 (1500+100) if the HTB is not >> specified. So if the actual MTU is not 1500, I''m also adding >> 100 to the actual value. > > Hmmmm... If not sure I get this ''if the HTB is not specified''. > The HTB default is 1600. So if MTU 1500 is read then 100 is > added to match, that seems OK. But of the MTU is 100, is there > any need to add 100 ?While the minimum *path* MTU on the Internet is 68 bytes, Internet hosts are required to handle 576-byte packets. So a practical lower bound on the MTU is 576. The ethernet v2 max MTU is 1500. I actually had an Internet uplink that forced the MTU on the interface to 576 and Shorewall''s HTB traffic shaping worked well on it. So I didn''t see any need to scale the delta based on the actual MTU setting. -Tom -- Tom Eastep \ When I die, I want to go like my Grandfather who Shoreline, \ died peacefully in his sleep. Not screaming like Washington, USA \ all of the passengers in his car http://shorewall.net \________________________________________________ ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today''s security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/