* [LARTC] r2q
@ 2002-05-19 8:09 john huttley
2002-05-19 9:09 ` Martin Devera
` (13 more replies)
0 siblings, 14 replies; 15+ messages in thread
From: john huttley @ 2002-05-19 8:09 UTC (permalink / raw)
To: lartc
Hello,
I've installed htb3 on my simple config
(htb + sfq) . Something is now complaining that quantum is too small
and suggest that r2q be changed.
There seems to be no documentation on this.
Ideas anyone?
Regards
John
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
@ 2002-05-19 9:09 ` Martin Devera
2004-01-26 23:04 ` [LARTC] R2Q Mihai Vlad
` (12 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Martin Devera @ 2002-05-19 9:09 UTC (permalink / raw)
To: lartc
It is documented in HTB manual. The complaint was not displayed
before and sharing was silently wrong.
Quantum of leaf should be between 1500 and 60000 and it is computed
as rate/r2q (or can be supplied independly).
So that if your maximal leaf rate is 1Mbit (120000Bps) then r2q
should be 3. If max rate is 10kbit (1200Bps) then r2q should be 1.
It is warning only and will not affect functionality, only precision.
devik
On 19 May 2002, john huttley wrote:
>
> Hello,
> I've installed htb3 on my simple config
> (htb + sfq) . Something is now complaining that quantum is too small
> and suggest that r2q be changed.
>
> There seems to be no documentation on this.
>
> Ideas anyone?
>
> Regards
>
> John
>
>
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
>
>
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* [LARTC] R2Q
2002-05-19 8:09 [LARTC] r2q john huttley
2002-05-19 9:09 ` Martin Devera
@ 2004-01-26 23:04 ` Mihai Vlad
2004-01-27 18:23 ` Stef Coene
` (11 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Mihai Vlad @ 2004-01-26 23:04 UTC (permalink / raw)
To: lartc
Hello again,
I need to change the R2Q for my script, as setting the quantum manually for
each class is painful. Can you tell me exactly where to set R2Q = x?
I get syntax errors all the time. Should it be specified for each class? I
do not know where to place this setting...
Thanks in advance,
Vlad Mihai
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] R2Q
2002-05-19 8:09 [LARTC] r2q john huttley
2002-05-19 9:09 ` Martin Devera
2004-01-26 23:04 ` [LARTC] R2Q Mihai Vlad
@ 2004-01-27 18:23 ` Stef Coene
2004-01-27 23:18 ` Mihai Vlad
` (10 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Stef Coene @ 2004-01-27 18:23 UTC (permalink / raw)
To: lartc
On Tuesday 27 January 2004 00:04, Mihai Vlad wrote:
> Hello again,
>
> I need to change the R2Q for my script, as setting the quantum manually for
> each class is painful. Can you tell me exactly where to set R2Q = x?
If you add the htb qdisc.
Stef
--
stef.coene@docum.org
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.openprojects.net
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* [LARTC] R2Q
2002-05-19 8:09 [LARTC] r2q john huttley
` (2 preceding siblings ...)
2004-01-27 18:23 ` Stef Coene
@ 2004-01-27 23:18 ` Mihai Vlad
2004-01-29 17:58 ` Stef Coene
` (9 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Mihai Vlad @ 2004-01-27 23:18 UTC (permalink / raw)
To: lartc
Here is a quote from docum.org:
-------------------------------------------------------------------------
Counting packets with quantum can be strange. If we have a low rate class
(rate = 5kbit), default quantum = 5000 / 10 = 500 bytes. But most packets
are more then 500 bytes. Htb version 1 and 2 uses DRR, so a packet larger
then 1000 bytes will be sent and it will remember how much it sent and wait
until the packet is paid back before another packet is send. So if you send
1000 byte, next time the class is polled, you will not be allowed to send.
Htb3 uses the WRR scheduler. When a packet with size > quantum is sent, it
will be sent and an error that the quantum is too small will be logged. But
there is no pay back. The WRR scheduler is faster then the DRR scheduler. So
make sure quantum is bigger then the default packet size. For 15 kbyte/s and
default r2q, quantum is 1500 and this is exactly the maximum packet size. If
you want to tune htb for rates smaller then 15 kbyte/s, you can manually set
the r2q and/or quantum.
----------------------------------------------------------------------------
Assuming the 5kbit example (kbit not kbytes) and that the R2Q is 10, we can
compute the quantum like this:
5 kbit = 5000 bit
5000 bit / 10 = 500 byte
Is it bytes or bits?
I guess the first term (the rate) is measured in bits and the quantum in
bytes.
Taking into account the second example (15 kbyte), we compute the quantum
like this:
15 kbyte = 15000 byte
15000 byte / 10 = 1500 byte
Is it bytes or bits?
So, in order to have a fully functional HTB 3 script I need to have each of
my class rates bigger than 15 kbyte? This is about 120 kbit.
What happens if I need lower rates like 8 kbit?
Do I need to set up the quantum manually?
Please don't laugh if I am talking nonsense, but I cannot figure it out...
Thanks,
Mihai
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] R2Q
2002-05-19 8:09 [LARTC] r2q john huttley
` (3 preceding siblings ...)
2004-01-27 23:18 ` Mihai Vlad
@ 2004-01-29 17:58 ` Stef Coene
2004-03-19 21:20 ` [LARTC] r2q ThE LinuX_KiD
` (8 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Stef Coene @ 2004-01-29 17:58 UTC (permalink / raw)
To: lartc
On Wednesday 28 January 2004 00:18, Mihai Vlad wrote:
> Here is a quote from docum.org:
>
> -------------------------------------------------------------------------
> Counting packets with quantum can be strange. If we have a low rate class
> (rate = 5kbit), default quantum = 5000 / 10 = 500 bytes. But most packets
> are more then 500 bytes. Htb version 1 and 2 uses DRR, so a packet larger
> then 1000 bytes will be sent and it will remember how much it sent and wait
> until the packet is paid back before another packet is send. So if you send
> 1000 byte, next time the class is polled, you will not be allowed to send.
>
> Htb3 uses the WRR scheduler. When a packet with size > quantum is sent, it
> will be sent and an error that the quantum is too small will be logged. But
> there is no pay back. The WRR scheduler is faster then the DRR scheduler.
> So make sure quantum is bigger then the default packet size. For 15 kbyte/s
> and default r2q, quantum is 1500 and this is exactly the maximum packet
> size. If you want to tune htb for rates smaller then 15 kbyte/s, you can
> manually set the r2q and/or quantum.
> ---------------------------------------------------------------------------
>
> Assuming the 5kbit example (kbit not kbytes) and that the R2Q is 10, we can
> compute the quantum like this:
> 5 kbit = 5000 bit
> 5000 bit / 10 = 500 byte
>
> Is it bytes or bits?
bytes
> I guess the first term (the rate) is measured in bits and the quantum in
> bytes.
Indeed.
> Taking into account the second example (15 kbyte), we compute the quantum
> like this:
> 15 kbyte = 15000 byte
> 15000 byte / 10 = 1500 byte
>
> Is it bytes or bits?
bytes
> So, in order to have a fully functional HTB 3 script I need to have each of
> my class rates bigger than 15 kbyte? This is about 120 kbit.
Indeed.
> What happens if I need lower rates like 8 kbit?
You can specify a lower r2q if you add the htb qdisc. And/or you can specify
quantum when you add a class.
> Do I need to set up the quantum manually?
If you don't find a good r2q, yes.
Stef
--
stef.coene@docum.org
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.openprojects.net
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (4 preceding siblings ...)
2004-01-29 17:58 ` Stef Coene
@ 2004-03-19 21:20 ` ThE LinuX_KiD
2004-03-19 21:53 ` Roy
` (7 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: ThE LinuX_KiD @ 2004-03-19 21:20 UTC (permalink / raw)
To: lartc
Hi...
I'm working with HTB, and have a question...
what happen if I create a class of "8kbit" ??
I get a log :
"HTB: quantum of class .... is small. Consider r2q change"
but, my r2q=1 I can't reduce that variable...
how must I procced ?
Thank you!
mac
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (5 preceding siblings ...)
2004-03-19 21:20 ` [LARTC] r2q ThE LinuX_KiD
@ 2004-03-19 21:53 ` Roy
2004-03-20 20:17 ` ThE LinuX_KiD
` (6 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Roy @ 2004-03-19 21:53 UTC (permalink / raw)
To: lartc
Simply set quantum manualy,
now quantum = rate / r2q
set quantum to one mtu ( 1500 bytes)
Remember that quantum affects priority
bigger quantum means higer priority.
----- Original Message -----
From: "ThE LinuX_KiD" <gregoriandres@yahoo.com.ar>
To: "lartc" <lartc@mailman.ds9a.nl>
Sent: Friday, March 19, 2004 11:20 PM
Subject: [LARTC] r2q
> Hi...
>
> I'm working with HTB, and have a question...
>
> what happen if I create a class of '"'8kbit'"' ??
>
> I get a log :
>
> '"'HTB: quantum of class .... is small. Consider r2q change'"'
>
> but, my r2q=1 I can't reduce that variable...
>
> how must I procced ?
>
> Thank you!
> mac
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
>
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* RE: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (6 preceding siblings ...)
2004-03-19 21:53 ` Roy
@ 2004-03-20 20:17 ` ThE LinuX_KiD
2004-03-21 7:20 ` rubens
` (5 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: ThE LinuX_KiD @ 2004-03-20 20:17 UTC (permalink / raw)
To: lartc
quantum = rate (bytes) / r2q
Rate=8 Kbits/s = 1 Kbyte/s = 1024 Bytes
r2q=1
quantum = 1024/1.
Then Quantum = 1024
But MTU\x1500 => MTU > Quantum
Must I set MTU to 1024 ????
Thank you!
Andres.
-> Simply set quantum manualy,
-> now quantum = rate / r2q
-> set quantum to one mtu ( 1500 bytes)
->
-> Remember that quantum affects priority
-> bigger quantum means higer priority.
->
->
-> ----- Original Message -----
-> From: "ThE LinuX_KiD" <gregoriandres@yahoo.com.ar>
-> To: "lartc" <lartc@mailman.ds9a.nl>
-> Sent: Friday, March 19, 2004 11:20 PM
-> Subject: [LARTC] r2q
->
->
-> > Hi...
-> >
-> > I'm working with HTB, and have a question...
-> >
-> > what happen if I create a class of '"'8kbit'"' ??
-> >
-> > I get a log :
-> >
-> > '"'HTB: quantum of class .... is small. Consider r2q change'"'
-> >
-> > but, my r2q=1 I can't reduce that variable...
-> >
-> > how must I procced ?
-> >
-> > Thank you!
-> > mac
-> > _______________________________________________
-> > LARTC mailing list / LARTC@mailman.ds9a.nl
-> > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
-> >
-> _______________________________________________
-> LARTC mailing list / LARTC@mailman.ds9a.nl
-> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
->
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (7 preceding siblings ...)
2004-03-20 20:17 ` ThE LinuX_KiD
@ 2004-03-21 7:20 ` rubens
2004-03-21 8:21 ` Bikrant Neupane
` (4 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: rubens @ 2004-03-21 7:20 UTC (permalink / raw)
To: lartc
Specify a quantum value(MTU suggested) on the class with 8 kbit.
Rubens
On Fri, 19 Mar 2004, ThE LinuX_KiD wrote:
> Hi...
>
> I'm working with HTB, and have a question...
>
> what happen if I create a class of "8kbit" ??
>
> I get a log :
>
> "HTB: quantum of class .... is small. Consider r2q change"
>
> but, my r2q=1 I can't reduce that variable...
>
> how must I procced ?
>
> Thank you!
> mac
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
>
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (8 preceding siblings ...)
2004-03-21 7:20 ` rubens
@ 2004-03-21 8:21 ` Bikrant Neupane
2004-03-22 18:50 ` ThE LinuX_KiD
` (3 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Bikrant Neupane @ 2004-03-21 8:21 UTC (permalink / raw)
To: lartc
I'm still confused about choosing r2q and quantum values.
I've got different classes. Some have rate as low as 16Kbit and some have
2048Kbit. I'm using default r2q value ie 10 I guess. Also I'm using same
quantum value for all the classes (ie quantum 1500) whatever their rate is.
Is this a good practice? I don't know. For r2q\x10 I think the value that i'm
using for quantuim (1500) is very high! In this case Should I lower the
quantum value or the r2q value.
I think I have two options:
1. r2q=1 and qunatum\x1500
2. r2q\x10 and quantum\x150
which one is better?
with regards,
Bikrant
On Saturday 20 March 2004 03:38, Roy wrote:
> Simply set quantum manualy,
> now quantum = rate / r2q
> set quantum to one mtu ( 1500 bytes)
>
> Remember that quantum affects priority
> bigger quantum means higer priority.
>
>
> ----- Original Message -----
> From: "ThE LinuX_KiD" <gregoriandres@yahoo.com.ar>
> To: "lartc" <lartc@mailman.ds9a.nl>
> Sent: Friday, March 19, 2004 11:20 PM
> Subject: [LARTC] r2q
>
> > Hi...
> >
> > I'm working with HTB, and have a question...
> >
> > what happen if I create a class of '"'8kbit'"' ??
> >
> > I get a log :
> >
> > '"'HTB: quantum of class .... is small. Consider r2q change'"'
> >
> > but, my r2q=1 I can't reduce that variable...
> >
> > how must I procced ?
> >
> > Thank you!
> > mac
> > _______________________________________________
> > LARTC mailing list / LARTC@mailman.ds9a.nl
> > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
>
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* RE: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (9 preceding siblings ...)
2004-03-21 8:21 ` Bikrant Neupane
@ 2004-03-22 18:50 ` ThE LinuX_KiD
2004-03-23 15:13 ` Roy
` (2 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: ThE LinuX_KiD @ 2004-03-22 18:50 UTC (permalink / raw)
To: lartc
Thank you, Roy !
I've setted quantum to 1500 when:
quantum = rate / r2q < 1500
and error logs are stopped.
In other hand, I get a new warning message:
"HTB: quantum of class .... is BIG. Consider r2q change"
because, some rates are about 1500kbits, then with
above formula quantum is about 192000 > 60000
But, I can't change r2q, then I put "quantum 55000"
on this rules, and error has been fixed.
Is correct the value "quantum 55000" ?
Or I must put mtu (1500) value for quantum ?
Thank you!
Andres.
->
->
-> No,
-> leave mtu unchanged, or you risk to have problems with some dumb servers.
-> just set qantum to 1500, for each rule when it is lower than mtu
-> bu default
->
-> (quantum = rate (bytes) / r2q) this formula is used by tc to calculate
-> default quantum
-> you do not need to use it.
->
->
-> ----- Original Message -----
-> From: "ThE LinuX_KiD" <gregoriandres@yahoo.com.ar>
-> To: "lartc" <lartc@mailman.ds9a.nl>
-> Sent: Saturday, March 20, 2004 10:17 PM
-> Subject: RE: [LARTC] r2q
->
->
-> >
-> >
-> >
-> >
-> > Rate=8 Kbits/s = 1 Kbyte/s = 1024 Bytes
-> > r2q=1
-> >
-> > quantum = 1024/1.
-> >
-> > Then Quantum = 1024
-> >
-> > But MTU\x1500 => MTU > Quantum
-> >
-> > Must I set MTU to 1024 ????
-> >
-> > Thank you!
-> >
-> > Andres.
-> >
-> >
-> >
-> > -> Simply set quantum manualy,
-> > -> now quantum = rate / r2q
-> > -> set quantum to one mtu ( 1500 bytes)
-> > ->
-> > -> Remember that quantum affects priority
-> > -> bigger quantum means higer priority.
-> > ->
-> > ->
-> > -> ----- Original Message -----
-> > -> From: '"'ThE LinuX_KiD'"' <gregoriandres@yahoo.com.ar>
-> > -> To: '"'lartc'"' <lartc@mailman.ds9a.nl>
-> > -> Sent: Friday, March 19, 2004 11:20 PM
-> > -> Subject: [LARTC] r2q
-> > ->
-> > ->
-> > -> > Hi...
-> > -> >
-> > -> > I'm working with HTB, and have a question...
-> > -> >
-> > -> > what happen if I create a class of ''"''8kbit''"'' ??
-> > -> >
-> > -> > I get a log :
-> > -> >
-> > -> > ''"''HTB: quantum of class .... is small. Consider r2q change''"''
-> > -> >
-> > -> > but, my r2q=1 I can't reduce that variable...
-> > -> >
-> > -> > how must I procced ?
-> > -> >
-> > -> > Thank you!
-> > -> > mac
-> > -> > _______________________________________________
-> > -> > LARTC mailing list / LARTC@mailman.ds9a.nl
-> > -> > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
-> http://lartc.org/
-> > -> >
-> > -> _______________________________________________
-> > -> LARTC mailing list / LARTC@mailman.ds9a.nl
-> > -> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
http://lartc.org/
> ->
> _______________________________________________
> LARTC mailing list / LARTC@mailman.ds9a.nl
> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
>
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (10 preceding siblings ...)
2004-03-22 18:50 ` ThE LinuX_KiD
@ 2004-03-23 15:13 ` Roy
2004-03-23 15:38 ` ThE LinuX_KiD
2004-03-28 17:59 ` Stef Coene
13 siblings, 0 replies; 15+ messages in thread
From: Roy @ 2004-03-23 15:13 UTC (permalink / raw)
To: lartc
I newer saw warning mesage about quantum too big.
but probably it is ok to have ANY quantum which is greater than mtu.
probabaly you shpuld not set quantum to 1500 if it is 50000 by defaul,
unless you want to lower that class priority very much.
but so high quantums can make trafic very bursty.
so you may set that quantum to 15000 for better result.
this will make these classes priority proportion as 1 to 10 ( 1 packet of
spare trafic for slow class to 10 packets for fast class)
----- Original Message -----
From: "ThE LinuX_KiD" <gregoriandres@yahoo.com.ar>
To: "lartc" <lartc@mailman.ds9a.nl>
Cc: "Roy" <roy@xxx.lt>
Sent: Monday, March 22, 2004 8:50 PM
Subject: RE: [LARTC] r2q
>
> Thank you, Roy !
>
> I've setted quantum to 1500 when:
>
> quantum = rate / r2q < 1500
>
> and error logs are stopped.
>
> In other hand, I get a new warning message:
>
> '"'HTB: quantum of class .... is BIG. Consider r2q change'"'
>
> because, some rates are about 1500kbits, then with
> above formula quantum is about 192000 > 60000
>
> But, I can't change r2q, then I put '"'quantum 55000'"'
> on this rules, and error has been fixed.
>
> Is correct the value '"'quantum 55000'"' ?
> Or I must put mtu (1500) value for quantum ?
>
> Thank you!
>
> Andres.
>
>
>
>
>
>
>
>
> ->
> ->
> -> No,
> -> leave mtu unchanged, or you risk to have problems with some dumb
servers.
> -> just set qantum to 1500, for each rule when it is lower than mtu
> -> bu default
> ->
> -> (quantum = rate (bytes) / r2q) this formula is used by tc to calculate
> -> default quantum
> -> you do not need to use it.
> ->
> ->
> -> ----- Original Message -----
> -> From: '"'ThE LinuX_KiD'"' <gregoriandres@yahoo.com.ar>
> -> To: '"'lartc'"' <lartc@mailman.ds9a.nl>
> -> Sent: Saturday, March 20, 2004 10:17 PM
> -> Subject: RE: [LARTC] r2q
> ->
> ->
> -> >
> -> >
> -> >
> -> >
> -> > Rate=8 Kbits/s = 1 Kbyte/s = 1024 Bytes
> -> > r2q=1
> -> >
> -> > quantum = 1024/1.
> -> >
> -> > Then Quantum = 1024
> -> >
> -> > But MTU\x1500 => MTU > Quantum
> -> >
> -> > Must I set MTU to 1024 ????
> -> >
> -> > Thank you!
> -> >
> -> > Andres.
> -> >
> -> >
> -> >
> -> > -> Simply set quantum manualy,
> -> > -> now quantum = rate / r2q
> -> > -> set quantum to one mtu ( 1500 bytes)
> -> > ->
> -> > -> Remember that quantum affects priority
> -> > -> bigger quantum means higer priority.
> -> > ->
> -> > ->
> -> > -> ----- Original Message -----
> -> > -> From: ''"''ThE LinuX_KiD''"'' <gregoriandres@yahoo.com.ar>
> -> > -> To: ''"''lartc''"'' <lartc@mailman.ds9a.nl>
> -> > -> Sent: Friday, March 19, 2004 11:20 PM
> -> > -> Subject: [LARTC] r2q
> -> > ->
> -> > ->
> -> > -> > Hi...
> -> > -> >
> -> > -> > I'm working with HTB, and have a question...
> -> > -> >
> -> > -> > what happen if I create a class of '''"'''8kbit'''"''' ??
> -> > -> >
> -> > -> > I get a log :
> -> > -> >
> -> > -> > '''"'''HTB: quantum of class .... is small. Consider r2q
> change'''"'''
> -> > -> >
> -> > -> > but, my r2q=1 I can't reduce that variable...
> -> > -> >
> -> > -> > how must I procced ?
> -> > -> >
> -> > -> > Thank you!
> -> > -> > mac
> -> > -> > _______________________________________________
> -> > -> > LARTC mailing list / LARTC@mailman.ds9a.nl
> -> > -> > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
> -> http://lartc.org/
> -> > -> >
> -> > -> _______________________________________________
> -> > -> LARTC mailing list / LARTC@mailman.ds9a.nl
> -> > -> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
> http://lartc.org/
> > ->
> > _______________________________________________
> > LARTC mailing list / LARTC@mailman.ds9a.nl
> > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
> >
>
>
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* RE: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (11 preceding siblings ...)
2004-03-23 15:13 ` Roy
@ 2004-03-23 15:38 ` ThE LinuX_KiD
2004-03-28 17:59 ` Stef Coene
13 siblings, 0 replies; 15+ messages in thread
From: ThE LinuX_KiD @ 2004-03-23 15:38 UTC (permalink / raw)
To: lartc
Ok, in order to conclude: is OK next table ?
QUANTUM values for HTB:
RATE --- QUANTUM
----------------------------------------------
8k --> set QUANTUM = 1500
16k --> Quantum is automatic by HTB algorithm
32k --> Quantum is automatic by HTB algorithm
64k --> Quantum is automatic by HTB algorithm
256k --> Quantum is automatic by HTB algorithm
512k --> set QUANTUM = 15000
----------------------------------------------
each with r2q=1
Also, I set r2q=1 on root class, in order to
support diferentes rates.
Thank you!
andres.
-> -----Mensaje original-----
-> De: lartc-admin@mailman.ds9a.nl [mailto:lartc-admin@mailman.ds9a.nl]En
-> nombre de Roy
-> Enviado el: Martes, 23 de Marzo de 2004 12:13 p.m.
-> Para: lartc
-> Asunto: Re: [LARTC] r2q
->
->
-> I newer saw warning mesage about quantum too big.
-> but probably it is ok to have ANY quantum which is greater than mtu.
->
-> probabaly you shpuld not set quantum to 1500 if it is 50000 by defaul,
-> unless you want to lower that class priority very much.
->
-> but so high quantums can make trafic very bursty.
-> so you may set that quantum to 15000 for better result.
->
-> this will make these classes priority proportion as 1 to 10 ( 1 packet of
-> spare trafic for slow class to 10 packets for fast class)
->
-> ----- Original Message -----
-> From: "ThE LinuX_KiD" <gregoriandres@yahoo.com.ar>
-> To: "lartc" <lartc@mailman.ds9a.nl>
-> Cc: "Roy" <roy@xxx.lt>
-> Sent: Monday, March 22, 2004 8:50 PM
-> Subject: RE: [LARTC] r2q
->
->
-> >
-> > Thank you, Roy !
-> >
-> > I've setted quantum to 1500 when:
-> >
-> > quantum = rate / r2q < 1500
-> >
-> > and error logs are stopped.
-> >
-> > In other hand, I get a new warning message:
-> >
-> > '"'HTB: quantum of class .... is BIG. Consider r2q change'"'
-> >
-> > because, some rates are about 1500kbits, then with
-> > above formula quantum is about 192000 > 60000
-> >
-> > But, I can't change r2q, then I put '"'quantum 55000'"'
-> > on this rules, and error has been fixed.
-> >
-> > Is correct the value '"'quantum 55000'"' ?
-> > Or I must put mtu (1500) value for quantum ?
-> >
-> > Thank you!
-> >
-> > Andres.
-> >
-> >
-> >
-> >
-> >
-> >
-> >
-> >
-> > ->
-> > ->
-> > -> No,
-> > -> leave mtu unchanged, or you risk to have problems with some dumb
-> servers.
-> > -> just set qantum to 1500, for each rule when it is lower than mtu
-> > -> bu default
-> > ->
-> > -> (quantum = rate (bytes) / r2q) this formula is used by tc
-> to calculate
-> > -> default quantum
-> > -> you do not need to use it.
-> > ->
-> > ->
-> > -> ----- Original Message -----
-> > -> From: '"'ThE LinuX_KiD'"' <gregoriandres@yahoo.com.ar>
-> > -> To: '"'lartc'"' <lartc@mailman.ds9a.nl>
-> > -> Sent: Saturday, March 20, 2004 10:17 PM
-> > -> Subject: RE: [LARTC] r2q
-> > ->
-> > ->
-> > -> >
-> > -> >
-> > -> >
-> > -> >
-> > -> > Rate=8 Kbits/s = 1 Kbyte/s = 1024 Bytes
-> > -> > r2q=1
-> > -> >
-> > -> > quantum = 1024/1.
-> > -> >
-> > -> > Then Quantum = 1024
-> > -> >
-> > -> > But MTU\x1500 => MTU > Quantum
-> > -> >
-> > -> > Must I set MTU to 1024 ????
-> > -> >
-> > -> > Thank you!
-> > -> >
-> > -> > Andres.
-> > -> >
-> > -> >
-> > -> >
-> > -> > -> Simply set quantum manualy,
-> > -> > -> now quantum = rate / r2q
-> > -> > -> set quantum to one mtu ( 1500 bytes)
-> > -> > ->
-> > -> > -> Remember that quantum affects priority
-> > -> > -> bigger quantum means higer priority.
-> > -> > ->
-> > -> > ->
-> > -> > -> ----- Original Message -----
-> > -> > -> From: ''"''ThE LinuX_KiD''"'' <gregoriandres@yahoo.com.ar>
-> > -> > -> To: ''"''lartc''"'' <lartc@mailman.ds9a.nl>
-> > -> > -> Sent: Friday, March 19, 2004 11:20 PM
-> > -> > -> Subject: [LARTC] r2q
-> > -> > ->
-> > -> > ->
-> > -> > -> > Hi...
-> > -> > -> >
-> > -> > -> > I'm working with HTB, and have a question...
-> > -> > -> >
-> > -> > -> > what happen if I create a class of '''"'''8kbit'''"''' ??
-> > -> > -> >
-> > -> > -> > I get a log :
-> > -> > -> >
-> > -> > -> > '''"'''HTB: quantum of class .... is small. Consider r2q
-> > change'''"'''
-> > -> > -> >
-> > -> > -> > but, my r2q=1 I can't reduce that variable...
-> > -> > -> >
-> > -> > -> > how must I procced ?
-> > -> > -> >
-> > -> > -> > Thank you!
-> > -> > -> > mac
-> > -> > -> > _______________________________________________
-> > -> > -> > LARTC mailing list / LARTC@mailman.ds9a.nl
-> > -> > -> > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
-> > -> http://lartc.org/
-> > -> > -> >
-> > -> > -> _______________________________________________
-> > -> > -> LARTC mailing list / LARTC@mailman.ds9a.nl
-> > -> > -> http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
-> > http://lartc.org/
-> > > ->
-> > > _______________________________________________
-> > > LARTC mailing list / LARTC@mailman.ds9a.nl
-> > > http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO:
http://lartc.org/
> >
>
>
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [LARTC] r2q
2002-05-19 8:09 [LARTC] r2q john huttley
` (12 preceding siblings ...)
2004-03-23 15:38 ` ThE LinuX_KiD
@ 2004-03-28 17:59 ` Stef Coene
13 siblings, 0 replies; 15+ messages in thread
From: Stef Coene @ 2004-03-28 17:59 UTC (permalink / raw)
To: lartc
On Tuesday 23 March 2004 16:13, Roy wrote:
> I newer saw warning mesage about quantum too big.
> but probably it is ok to have ANY quantum which is greater than mtu.
There is a maximum of 60000 for quantum to prevent class starvation.
Stef
--
stef.coene@docum.org
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.openprojects.net
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2004-03-28 17:59 UTC | newest]
Thread overview: 15+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-05-19 8:09 [LARTC] r2q john huttley
2002-05-19 9:09 ` Martin Devera
2004-01-26 23:04 ` [LARTC] R2Q Mihai Vlad
2004-01-27 18:23 ` Stef Coene
2004-01-27 23:18 ` Mihai Vlad
2004-01-29 17:58 ` Stef Coene
2004-03-19 21:20 ` [LARTC] r2q ThE LinuX_KiD
2004-03-19 21:53 ` Roy
2004-03-20 20:17 ` ThE LinuX_KiD
2004-03-21 7:20 ` rubens
2004-03-21 8:21 ` Bikrant Neupane
2004-03-22 18:50 ` ThE LinuX_KiD
2004-03-23 15:13 ` Roy
2004-03-23 15:38 ` ThE LinuX_KiD
2004-03-28 17:59 ` Stef Coene
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.