On Thu, 1 Apr 2004, Claudiu Cismaru wrote: > =2D----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > > > LOL, sure. So 100mbit inseamna ~12mbytes/s, so 125000 quantum (si > > acum am gresit?) > > Nu as risca sa-l pun asa de mare... cf. docs: > > It is important to know that for precise operation quantums need to be=20 > as small as possible and larger than MTU.=20 > Normaly you don't need to specify quantums manualy as HTB chooses=20 > precomputed values. It computes classe's quantum (when you add or=20 > change it) as its rate divided by r2q global parameter. Its default=20 > value is 10 and because typical MTU is 1500 the default is good for=20 > rates from 15 kBps (120 kbit). For smaller minimal rates specify r2q 1=20 > when creating qdisc - it is good from 12 kbit which should be enough.=20 > If you will need you can specify quantum manualy when adding or=20 > changing the class. You can avoid warnings in log if precomputed value=20 > would be bad. When you specify quantum on command line the r2q is=20 > ignored for that class.=20 > > Daca pui un r2q mic incat sa ai 125000 quantum o sa-ti dea warning...=20 > Daca-l specifici tu explicit atata... nu stiu cat o sa-ti functioneze=20 > de bine...
Problema nu e daca risti tu sau nu, ci daca pui un quantum prea mic pt garantari foarte mari (eu vorbeam de 100mbit) nu o sa atingi acele valori. r2q este practic doar daca ai garantari fara variatie mare (si de exemplu nu merge folosit daca ai garantari care variaza de la 16k la 10mbit, in acel caz ajungi la ce am zis eu, trebuie sa precizezi quantumul). > =2D --=20 > Claudiu Cismaru > GPG Key: http://maya.cnixs.com/~claudiu/claudiu.gpg -- Mihai RUSU Email: [EMAIL PROTECTED] GPG : http://dizzy.roedu.net/dizzy-gpg.txt WWW: http://dizzy.roedu.net "Linux is obsolete" -- AST --- Detalii despre listele noastre de mail: http://www.lug.ro/
