Dear sir/madam, I discover that bandwidth usage which is limited using htb is about 5 to 10% higher than the rate/ceil we set, if we monitor it using iptraf. I become more confused since I saw, the grafic in HTB homepage are also show more bandwidth than the setting. My question is: * what is the exac value of this over bandwidth? * how to make the bandwidth are limited exacly at "x" kbps? -- Regards, Adi Nugroho
On Monday 16 December 2002 03:25, Adi Nugroho wrote: > Dear sir/madam, > > I discover that bandwidth usage which is limited using htb is about 5 t= o > 10% higher than the rate/ceil we set, if we monitor it using iptraf. > > I become more confused since I saw, the grafic in HTB homepage are also > show more bandwidth than the setting. > > My question is: > * what is the exac value of this over bandwidth? > * how to make the bandwidth are limited exacly at "x" kbps? I did some "stream" tests with htb. I generated a lot of tcp traffic wit= h tcp=20 and recorded the shaped bandwidth. The result is allmost perfect. (resul= ts=20 can be found on www.docum.org on the test pages). But in a real world scenario, you also have bursts in the traffic. Maybe= you=20 allow bursts (you always have some minimal bursts) so the average bandwid= th=20 may be higher then the configured bandwith. =20 Stef --=20 stef.coene@docum.org "Using Linux as bandwidth manager" http://www.docum.org/ #lartc @ irc.oftc.net
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Adi Nugroho wrote: | Dear sir/madam, | | I discover that bandwidth usage which is limited using htb is about 5 to 10% | higher than the rate/ceil we set, if we monitor it using iptraf. | | I become more confused since I saw, the grafic in HTB homepage are also show | more bandwidth than the setting. | | My question is: | * what is the exac value of this over bandwidth? It depends of burst and cburst values that you (or tc) set. In Devik's user guide says: "Latest tc tool will compute and set the smallest possible burst when it is not specified". This means: minimal_burst=max_rate*timer_resolution So, for 2Mbit bandwidth on i386, minimal_burst=2Mbit*10ms=20kbit. This says the theory, but in real world I saw that this minimal_burst calculated by tc could be up to 64kbit. Am I missing something? | * how to make the bandwidth are limited exacly at "x" kbps? Set a lower value for ceil and a value for cburst thus: ceil+cburst=limited_bandwidth - -- Catalin Bucur mailto:cata@geniusnet.ro NOC @ Genius Network SRL - Galati - Romania -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9/ggIpDe20wwI9oIRApcqAJ94Z53bvrGpAvCUVPO8mmI3eHIV2gCfcwXK pALs5TiT9dbZf42GNtVBHOc= =JRkQ -----END PGP SIGNATURE-----
Pada hari Selasa, 17 Desember 2002 01:06, Catalin Bucur menulis: > | * how to make the bandwidth are limited exacly at "x" kbps? > > Set a lower value for ceil and a value for cburst thus: > ceil+cburst=limited_bandwidth Thank you very much. This calculation is very helpfull. -- Salam, Adi Nugroho