Re: keyboard latency from time to time

[Matthias Apitz]

El día Monday, April 09, 2012 a las 07:06:59PM +0200, Matthias Apitz escribió:

> Months later, in some other issue, I learned about the feature of KDE
> "slow keys" and what I have described is exactly the same behaviour and
> I can now even reproduce this with just pressing and holding down the
> Shift-key for around 8 secs; when it happens one must go to the KDE
> Control Center and activate 'slow keys' (yes, they are not shown as
> activated in this moment) and deactivate 'slow keys' again, and all is
> fine.

I was curious and digged deeper into this...

"slow keys" is part of the X11 XKB-protocol (details in XKBproto.pdf or
XkbGetSlowKeysDelay(3) man page; in the ports there is ports/x11/xkbset
utility and with this one can set the "Slowkeys acceptance delay" (time
in ms the X server awaits a key is hold down before sending out the
keypress event) or to switch it off again:

$ xkbset slowkeys 500  # to switch in on and set 500ms delay
$ xkbset -slowkeys # to switch it off again

HIH

matthias
-- 
Matthias Apitz
e  - w http://www.unixarea.de/
UNIX since V7 on PDP-11, UNIX on mainframe since ESER 1055 (IBM /370)
UNIX on x86 since SVR4.2 UnixWare 2.1.2, FreeBSD since 2.2.5
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: keyboard latency from time to time

[Matthias Apitz]

El día Monday, July 25, 2011 a las 05:10:47PM +0200, Matthias Apitz escribió:

> 
> Hello,
> 
> I run a 9-CURRENT from end of October on an Acer D250 laptop; which in
> general runs very fine; from time to time (say once a month) I encounter
> the following situation within KDE3 or X11:
> 
> from a moment to another (can't say what action triggers this) the
> keyboard stops working; there are just no keyevents delivered on short
> press to any window; I've checked it with xev(1); mouse is working fine and I 
> can
> close the windows or the whole session of KDE, of course loosing my
> connections or the content of files I'm editing in that moment;
> 
> the keyevent is only delivered when you press the key for around half
> second, or so, and after this also the normal key re-iteration is
> produced; its even hard to catch only one keyevent and not twice or more;
> this is true for all keys, including Ctrl and Backspace, ...
> 
> after restarting X11 and KDE it is fine again;

Hello,

Months later, in some other issue, I learned about the feature of KDE
"slow keys" and what I have described is exactly the same behaviour and
I can now even reproduce this with just pressing and holding down the
Shift-key for around 8 secs; when it happens one must go to the KDE
Control Center and activate 'slow keys' (yes, they are not shown as
activated in this moment) and deactivate 'slow keys' again, and all is
fine.

matthias
-- 
Matthias Apitz
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e  - w http://www.unixarea.de/
UNIX since V7 on PDP-11 | UNIX on mainframe since ESER 1055 (IBM /370)
UNIX on x86 since SVR4.2 UnixWare 2.1.2 | FreeBSD since 2.2.5
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

keyboard latency from time to time

[Matthias Apitz]

Hello,

I run a 9-CURRENT from end of October on an Acer D250 laptop; which in
general runs very fine; from time to time (say once a month) I encounter
the following situation within KDE3 or X11:

from a moment to another (can't say what action triggers this) the
keyboard stops working; there are just no keyevents delivered on short
press to any window; I've checked it with xev(1); mouse is working fine and I 
can
close the windows or the whole session of KDE, of course loosing my
connections or the content of files I'm editing in that moment;

the keyevent is only delivered when you press the key for around half
second, or so, and after this also the normal key re-iteration is
produced; its even hard to catch only one keyevent and not twice or more;
this is true for all keys, including Ctrl and Backspace, ...

after restarting X11 and KDE it is fine again;

What could this causing?

Thanks

matthias
-- 
Matthias Apitz
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e  - w http://www.unixarea.de/
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Online gaming and file downloads - latency hell! (fwd)

[Ian Smith]

Hi .. as suggested, posting this discussion to ipfw@ too .. thanks, Ian

-- Forwarded message --
Date: Mon, 21 Jun 2010 13:00:14 +0200
From: Luigi Rizzo 
To: Ian Smith 
Subject: Re: Online gaming and file downloads - latency hell! (fwd)

On Mon, Jun 21, 2010 at 10:25 AM, Ian Smith  wrote:
> Hi Luigi,
>
> thought you might be interested, seeing you and ipfw+dummynet rate a
> mention .. maybe I'm ill-informed about the gaming latency aspect?
>
> cheers, Ian

for responsive flows (e.g. tcp), it is actually effective to shape
incoming traffic for the flows that you want slow down.
In fact, even though it is true that packets have already traversed
the bottleneck,
you are slowing down their arrival to the receiving TCP, which has the side
effect of slowing down outgoing acks and thus slowing down the source.
Queues will build up at the remote end while the system settles, but
after that (and if you make sure that the total rate is lower than link
capacity) it will be the sources that do not generate traffic that
saturates the link.

Pretty trivial to implement, too: you just need a set of rules that match
the incoming traffic that you want to slow down, pass it to one pipe, and
let the other traffic go through without filtering.

cheers
luigi

>
> -- Forwarded message --
> Date: Mon, 21 Jun 2010 15:50:05 +1000 (EST)
> From: Ian Smith 
> To: Morgan Wesstr?m 
> Cc: freebsd-questions@freebsd.org, Modulok 
> Subject: Re: Online gaming and file downloads - latency hell!
>
> In freebsd-questions Digest, Vol 315, Issue 11, Message: 9
> On Fri, 18 Jun 2010 12:11:48 +0200
> Morgan Wesstr?m  wrote:
>  > On 2010-06-16 02:51, Modulok wrote:
>  > > Yo,
>  > >
>  > > I have a FreeBSD box acting as a router between me and the Internet.
>  > > Whenever someone on the local network downloads something, the other
>  > > connections have a really high latency. A second or more. For people
>  > > who like to download large files and play online games, it's not good.
>  > >
>  > > I tried traffic shaping with PF, which works - almost: I tried the
>  > > home example in the PF book, but it doesn't work out so well. I can
>  > > throttle users with no trouble, but even so that doesn't seem to help
>  > > the latency issue unless I choke the 'big file download' users almost
>  > > completely off. It's like nothing helps. I tried a priority based
>  > > queue where all traffic on the gaming ports was placed in front of all
>  > > other traffic, and while I saw a very mild improvement, latency was
>  > > still really pitiful.
>  > >
>  > > Is there anything else I can do? Anyone have a similar setup and wish
>  > > to share config files? Are there some sysctl's that would help this
>  > > out or something. I'm almost ready to just buy a 'gaming' *gag* router
>  > > which implements their own brand of QoS, but don't want to sink to
>  > > that level if I don't have to.
>  > >
>  > > Help!
>  > > -Modulok-
>  >
>  > Traffic shaping on your side when downloading unfortunately doesn't help
>  > you. The data has already been transferred across your cable or DSL
>  > connection by then and reordering any packets on your side will not
>  > change the latency. Traffic shaping your download has to be performed at
>  > your upstream router which you probably don't control. PF can help you
>  > traffic shape your outgoing traffic. I have used it for this for the
>  > past 6 years to help me maintain a low and stable ping while I play
>  > online, even if I upload simultaneously. I've read about people trying
>  > to throttle outgoing ACKs to slow down their download but that still
>  > wouldn't rearrange any incoming data packets so I don't see how that
>  > would help. I haven't tried it myself though but neither have I read
>  > about anyone successfully accomplishing this.
>  >
>  > Regards
>  > Morgan
>
> A short story:
>
> About 15 months ago, before becoming aware that Luigi and colleagues had
> been busy porting ipfw and dummynet to Linux, I was asked to implement a
> shaping solution for a very limited (512/512kbps) ADSL connection for a
> community radio station using a Linux firewall-in-a-box called IPCop as
> router, whose shaping was based on Bert Hubert's WonderShaper script,
> using Linux' tc module to prioritise and shape only outbound traffic.
>
> Having used ipfw+dummynet successfully for some years to shape traffic
> for a local voluntary organisation 'Community Technology Centre'

Re: Online gaming and file downloads - latency hell!

[Ian Smith]

On Mon, 21 Jun 2010, Olivier Nicole wrote:
 > Hi,
 > 
 >  > I've read about people trying
 >  > to throttle outgoing ACKs to slow down their download but that still
 >  > wouldn't rearrange any incoming data packets so I don't see how that
 >  > would help. I haven't tried it myself though but neither have I read
 >  > about anyone successfully accomplishing this.
 > 
 > TCP uses a window: the maximum number of packects that you can receive
 > before you send an ACK. As long as ACK come flowing, the window size
 > increases.
 > 
 > Limit the ACK, you limit/reduce the size of the window, so you
 > limit/reduce the incoming trafic.

Indeed.  If you've an in-house router queueing through traffic against 
some bandwidth limit imposed on an inside clients' download, dropping 
any excess TCP packets arriving on top of a full queue or pipe (eg with 
ipfw/dummynet), there'll be a few packets requiring retransmission to 
continue the transfer, now and again, without need to throttle ACKs; in 
fact we're expediting ACKs uphill, after streaming, ssh and ICMP.

I've been surprised by how few packets get dropped and so resent, way 
less than 1%, even when pulling large files from fast providers through 
a slow link (512/512 ADSL as mentioned) then further limited to clients.

Which are mostly 'doze, a mac or two, a couple of linux boxes; all seem 
to use SACK but I haven't looked into negotiated window sizes.  I don't 
know TCP in any depth but watch with awe as people enhance and tune the 
stack; all I can say is 'it seems to mostly work pretty well here' ..

How UDP-based services cope with dropped packets is another matter; 
perhaps that's a big issue for some games that may need expediting?

 > I beleive there could even be some nasty rewritting that would
 > artifically change the window size so the TCP stream is slowed down.

Quite a job, intervening and rewriting packets, and maintaining state on 
whole streams; I gather TCP is resistant to Man in the Middle attack ..

Anyway, a lot harder than configuring a few dummynet pipes and queues :)

cheers, Ian
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Online gaming and file downloads - latency hell!

[Morgan Wesström]

On 2010-06-21 07:50, Ian Smith wrote:
> In freebsd-questions Digest, Vol 315, Issue 11, Message: 9
> On Fri, 18 Jun 2010 12:11:48 +0200
> Morgan Wesstr?m  wrote:
>  > On 2010-06-16 02:51, Modulok wrote:
>  > > Yo,
>  > > 
>  > > I have a FreeBSD box acting as a router between me and the Internet.
>  > > Whenever someone on the local network downloads something, the other
>  > > connections have a really high latency. A second or more. For people
>  > > who like to download large files and play online games, it's not good.
>  > > 
>  > > I tried traffic shaping with PF, which works - almost: I tried the
>  > > home example in the PF book, but it doesn't work out so well. I can
>  > > throttle users with no trouble, but even so that doesn't seem to help
>  > > the latency issue unless I choke the 'big file download' users almost
>  > > completely off. It's like nothing helps. I tried a priority based
>  > > queue where all traffic on the gaming ports was placed in front of all
>  > > other traffic, and while I saw a very mild improvement, latency was
>  > > still really pitiful.
>  > > 
>  > > Is there anything else I can do? Anyone have a similar setup and wish
>  > > to share config files? Are there some sysctl's that would help this
>  > > out or something. I'm almost ready to just buy a 'gaming' *gag* router
>  > > which implements their own brand of QoS, but don't want to sink to
>  > > that level if I don't have to.
>  > > 
>  > > Help!
>  > > -Modulok-
>  > 
>  > Traffic shaping on your side when downloading unfortunately doesn't help
>  > you. The data has already been transferred across your cable or DSL
>  > connection by then and reordering any packets on your side will not
>  > change the latency. Traffic shaping your download has to be performed at
>  > your upstream router which you probably don't control. PF can help you
>  > traffic shape your outgoing traffic. I have used it for this for the
>  > past 6 years to help me maintain a low and stable ping while I play
>  > online, even if I upload simultaneously. I've read about people trying
>  > to throttle outgoing ACKs to slow down their download but that still
>  > wouldn't rearrange any incoming data packets so I don't see how that
>  > would help. I haven't tried it myself though but neither have I read
>  > about anyone successfully accomplishing this.
>  > 
>  > Regards
>  > Morgan
> 
> A short story:
> 
> About 15 months ago, before becoming aware that Luigi and colleagues had 
> been busy porting ipfw and dummynet to Linux, I was asked to implement a 
> shaping solution for a very limited (512/512kbps) ADSL connection for a 
> community radio station using a Linux firewall-in-a-box called IPCop as 
> router, whose shaping was based on Bert Hubert's WonderShaper script, 
> using Linux' tc module to prioritise and shape only outbound traffic.
> 
> Having used ipfw+dummynet successfully for some years to shape traffic 
> for a local voluntary organisation 'Community Technology Centre', I was 
> staggered to find that all of the collective Linux wisdom on the subject 
> chanted that same mantra .. that you can't prioritise download traffic, 
> as the ISP will have 'gigantic queues' of TCP traffic that you can't 
> control, and that prioritising ACKs, QoS and ICMP traffic and such is 
> the best you can do.  By this philosophy, tc only implements limiting 
> total bandwidth of inbound traffic, shaping outbound by QoS and classes.
> 
> To disprove this pervasive myth I had to implement inbound shaping by 
> using tc to control the _outbound_ traffic to the _inside_ interface, 
> where all sorts of random clients are doing big downloads, yootoobing 
> and such plus some big uploads, while guaranteeing that the station's 
> outbound audio stream had fully half the outbound-to-net bandwidth free 
> without undue pressure and that remote ssh sessions etc remained snappy.  
> 
> This involves queuing inbound (mostly TCP) traffic on the local router, 
> dropping any excess, which works most effectively to maintain a hard 
> limit to downloads (at around 85% of 512kbps) while keeping the outbound 
> (to-net) channel lightly loaded after streaming, ACKs, and uploads.
> 
> I don't know how pf works (or can be made to work) in this regard, nor 
> can I speculate about gaming latency particularly, but hope to find out 
> soon by either replacing the old IPCop box with pfSense, or trying ipfw 

Re: Online gaming and file downloads - latency hell!

[Olivier Nicole]

Hi,

 > I've read about people trying
 > to throttle outgoing ACKs to slow down their download but that still
 > wouldn't rearrange any incoming data packets so I don't see how that
 > would help. I haven't tried it myself though but neither have I read
 > about anyone successfully accomplishing this.

TCP uses a window: the maximum number of packects that you can receive
before you send an ACK. As long as ACK come flowing, the window size
increases.

Limit the ACK, you limit/reduce the size of the window, so you
limit/reduce the incoming trafic.

I beleive there could even be some nasty rewritting that would
artifically change the window size so the TCP stream is slowed down.

Bests,

olivier
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Online gaming and file downloads - latency hell!

[Ian Smith]

In freebsd-questions Digest, Vol 315, Issue 11, Message: 9
On Fri, 18 Jun 2010 12:11:48 +0200
Morgan Wesstr?m  wrote:
 > On 2010-06-16 02:51, Modulok wrote:
 > > Yo,
 > > 
 > > I have a FreeBSD box acting as a router between me and the Internet.
 > > Whenever someone on the local network downloads something, the other
 > > connections have a really high latency. A second or more. For people
 > > who like to download large files and play online games, it's not good.
 > > 
 > > I tried traffic shaping with PF, which works - almost: I tried the
 > > home example in the PF book, but it doesn't work out so well. I can
 > > throttle users with no trouble, but even so that doesn't seem to help
 > > the latency issue unless I choke the 'big file download' users almost
 > > completely off. It's like nothing helps. I tried a priority based
 > > queue where all traffic on the gaming ports was placed in front of all
 > > other traffic, and while I saw a very mild improvement, latency was
 > > still really pitiful.
 > > 
 > > Is there anything else I can do? Anyone have a similar setup and wish
 > > to share config files? Are there some sysctl's that would help this
 > > out or something. I'm almost ready to just buy a 'gaming' *gag* router
 > > which implements their own brand of QoS, but don't want to sink to
 > > that level if I don't have to.
 > > 
 > > Help!
 > > -Modulok-
 > 
 > Traffic shaping on your side when downloading unfortunately doesn't help
 > you. The data has already been transferred across your cable or DSL
 > connection by then and reordering any packets on your side will not
 > change the latency. Traffic shaping your download has to be performed at
 > your upstream router which you probably don't control. PF can help you
 > traffic shape your outgoing traffic. I have used it for this for the
 > past 6 years to help me maintain a low and stable ping while I play
 > online, even if I upload simultaneously. I've read about people trying
 > to throttle outgoing ACKs to slow down their download but that still
 > wouldn't rearrange any incoming data packets so I don't see how that
 > would help. I haven't tried it myself though but neither have I read
 > about anyone successfully accomplishing this.
 > 
 > Regards
 > Morgan

A short story:

About 15 months ago, before becoming aware that Luigi and colleagues had 
been busy porting ipfw and dummynet to Linux, I was asked to implement a 
shaping solution for a very limited (512/512kbps) ADSL connection for a 
community radio station using a Linux firewall-in-a-box called IPCop as 
router, whose shaping was based on Bert Hubert's WonderShaper script, 
using Linux' tc module to prioritise and shape only outbound traffic.

Having used ipfw+dummynet successfully for some years to shape traffic 
for a local voluntary organisation 'Community Technology Centre', I was 
staggered to find that all of the collective Linux wisdom on the subject 
chanted that same mantra .. that you can't prioritise download traffic, 
as the ISP will have 'gigantic queues' of TCP traffic that you can't 
control, and that prioritising ACKs, QoS and ICMP traffic and such is 
the best you can do.  By this philosophy, tc only implements limiting 
total bandwidth of inbound traffic, shaping outbound by QoS and classes.

To disprove this pervasive myth I had to implement inbound shaping by 
using tc to control the _outbound_ traffic to the _inside_ interface, 
where all sorts of random clients are doing big downloads, yootoobing 
and such plus some big uploads, while guaranteeing that the station's 
outbound audio stream had fully half the outbound-to-net bandwidth free 
without undue pressure and that remote ssh sessions etc remained snappy.  

This involves queuing inbound (mostly TCP) traffic on the local router, 
dropping any excess, which works most effectively to maintain a hard 
limit to downloads (at around 85% of 512kbps) while keeping the outbound 
(to-net) channel lightly loaded after streaming, ACKs, and uploads.

I don't know how pf works (or can be made to work) in this regard, nor 
can I speculate about gaming latency particularly, but hope to find out 
soon by either replacing the old IPCop box with pfSense, or trying ipfw 
and dummynet on Linux .. I know, but they're still reluctant to shop 
other than Linux, and the idea of implementing a FreeBSD-derived 
firewall and shaping solution on Linux has a good deal of appeal ..

HTH (or at least, doesn't hurt :)

cheers, Ian
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Online gaming and file downloads - latency hell!

[RW]

On Fri, 18 Jun 2010 12:11:48 +0200
Morgan Wesström  wrote:

> On 2010-06-16 02:51, Modulok wrote:
> > Yo,
> > 
> > I have a FreeBSD box acting as a router between me and the Internet.
> > Whenever someone on the local network downloads something, the other
> > connections have a really high latency. A second or more. For people
> > who like to download large files and play online games, it's not
> > good.

> Traffic shaping on your side when downloading unfortunately doesn't
> help you. The data has already been transferred across your cable or
> DSL connection by then and reordering any packets on your side will
> not change the latency. Traffic shaping your download has to be
> performed at your upstream router which you probably don't control.

If the downloads are ordinary http and ftp, rather than P2P, you can
use squid to throttle at the TCP level. It needs to be built with delay
pools.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: Online gaming and file downloads - latency hell!

[Morgan Wesström]

On 2010-06-16 02:51, Modulok wrote:
> Yo,
> 
> I have a FreeBSD box acting as a router between me and the Internet.
> Whenever someone on the local network downloads something, the other
> connections have a really high latency. A second or more. For people
> who like to download large files and play online games, it's not good.
> 
> I tried traffic shaping with PF, which works - almost: I tried the
> home example in the PF book, but it doesn't work out so well. I can
> throttle users with no trouble, but even so that doesn't seem to help
> the latency issue unless I choke the 'big file download' users almost
> completely off. It's like nothing helps. I tried a priority based
> queue where all traffic on the gaming ports was placed in front of all
> other traffic, and while I saw a very mild improvement, latency was
> still really pitiful.
> 
> Is there anything else I can do? Anyone have a similar setup and wish
> to share config files? Are there some sysctl's that would help this
> out or something. I'm almost ready to just buy a 'gaming' *gag* router
> which implements their own brand of QoS, but don't want to sink to
> that level if I don't have to.
> 
> Help!
> -Modulok-

Traffic shaping on your side when downloading unfortunately doesn't help
you. The data has already been transferred across your cable or DSL
connection by then and reordering any packets on your side will not
change the latency. Traffic shaping your download has to be performed at
your upstream router which you probably don't control. PF can help you
traffic shape your outgoing traffic. I have used it for this for the
past 6 years to help me maintain a low and stable ping while I play
online, even if I upload simultaneously. I've read about people trying
to throttle outgoing ACKs to slow down their download but that still
wouldn't rearrange any incoming data packets so I don't see how that
would help. I haven't tried it myself though but neither have I read
about anyone successfully accomplishing this.

Regards
Morgan
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Online gaming and file downloads - latency hell!

[Modulok]

Yo,

I have a FreeBSD box acting as a router between me and the Internet.
Whenever someone on the local network downloads something, the other
connections have a really high latency. A second or more. For people
who like to download large files and play online games, it's not good.

I tried traffic shaping with PF, which works - almost: I tried the
home example in the PF book, but it doesn't work out so well. I can
throttle users with no trouble, but even so that doesn't seem to help
the latency issue unless I choke the 'big file download' users almost
completely off. It's like nothing helps. I tried a priority based
queue where all traffic on the gaming ports was placed in front of all
other traffic, and while I saw a very mild improvement, latency was
still really pitiful.

Is there anything else I can do? Anyone have a similar setup and wish
to share config files? Are there some sysctl's that would help this
out or something. I'm almost ready to just buy a 'gaming' *gag* router
which implements their own brand of QoS, but don't want to sink to
that level if I don't have to.

Help!
-Modulok-
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Tune USB latency...is that the right word?

[Modulok]

Aside from hacking ulpt.c, is there any way to tune the USB latency? More...

I have a USB printer, which works, but it takes about 2 minutes before
FreeBSD 7.2 will send it ANY data...during that 2 minutes nothing is
hapenning. It's not a printer warmup issue either, or a cable or
anything like that, as it worked on FreeBSD 6.1 (after applying a
patch to ulpt.c as submitted by someone on this list. (You know who
you are! Thanks!) For lack of a better term is there a, "don't wait 2
minutes before you decide the printer is probably not up to USB spec
and start talking to it anyway" sysctl or something?

-Modulok-
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Re: FreeBSD on Mini-ITX has web page latency

[Lowell Gilbert]

Don't top-post, please.

bobmc <[EMAIL PROTECTED]> writes:

> Well, it seems that the driver and the hardware works so there must be
> a problem in the generic part of the software.  There are plenty of tools
> for networking analysis but I am not a networking adept. So I will
> carefully repeat the install.  Thanks.  -Bob-

In that case, definitely try 6.2...
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD on Mini-ITX has web page latency

[bobmc]

Well, it seems that the driver and the hardware works so there must be
a problem in the generic part of the software.  There are plenty of tools
for networking analysis but I am not a networking adept. So I will
carefully repeat the install.  Thanks.  -Bob-

BTW, my other computer is a Biostar Ideq SFF with Via chipsets. FreeBSD
reports a 6102 ethernet but cannot map the interrupt. Instead it wants to
emulate ethernet on FireWire.  But that is for another day.  :-)

Lowell Gilbert wrote:
> Bob McIsaac <[EMAIL PROTECTED]> writes:
> FreeBSD using Konqueor or Lynx takes more than 10 seconds.  This is puzzling 
> since ftp transfers at 400kbs, pings of freebsd.org take 80ms, and top shows 
> CPU is 93% idle.
>   
>
> I take it, then, that FreeBSD is the web client, not the server?  
> Have you checked for whether the delays are being caused by name
> service, before the HTTP session is even started?
>   
>> This is for a EPIA-CN13 mini-itx with .5gb memory.
>> Note the message log sees a VT6102 LAN but the board has a 6103. Perhaps 
>> that explains the problem. Otherwise, it must be a protocol issue.
>> 
> A similarly identified interface works okay for me on my Via C3
> board.  Admittedly, they are lousy chips, but you should't notice for
> most purposes.
>
>   

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD on Mini-ITX has web page latency

[Lowell Gilbert]

Bob McIsaac <[EMAIL PROTECTED]> writes:

> Usability studies say that a person won't wait more than 4 seconds for a
> web page download.  FreeBSD using Konqueor or Lynx takes more than 10
> seconds.  This is puzzling since ftp transfers at 400kbs, pings of
> freebsd.org take 80ms, and top shows CPU is 93% idle.

I take it, then, that FreeBSD is the web client, not the server?  
Have you checked for whether the delays are being caused by name
service, before the HTTP session is even started?

> This is for a EPIA-CN13 mini-itx with .5gb memory.
> http://www.via.com.tw/en/products/mainboards/motherboards.jsp?motherboard_id=400
>
> Via says it works with Windows and Linux. It worked well for me using Mepis
> Linux.
>
> Note the message log sees a VT6102 LAN but the board has a 6103. Perhaps
> that explains the problem. Otherwise, it must be a protocol issue.

A similarly identified interface works okay for me on my Via C3
board.  Admittedly, they are lousy chips, but you should't notice for
most purposes.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

FreeBSD on Mini-ITX has web page latency

[Bob McIsaac]

Hi:

Usability studies say that a person won't wait more than 4 seconds for a
web page download.  FreeBSD using Konqueor or Lynx takes more than 10
seconds.  This is puzzling since ftp transfers at 400kbs, pings of
freebsd.org take 80ms, and top shows CPU is 93% idle.

This is for a EPIA-CN13 mini-itx with .5gb memory.
http://www.via.com.tw/en/products/mainboards/motherboards.jsp?motherboard_id=400

Via says it works with Windows and Linux. It worked well for me using Mepis
Linux.

Note the message log sees a VT6102 LAN but the board has a 6103. Perhaps
that explains the problem. Otherwise, it must be a protocol issue.

Any suggestions would be welcome

regards,
-Bob-


k: FreeBSD 6.1-RELEASE #0: Sun May  7 04:32:43 UTC 2006
k: [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC
k: Timecounter "i8254" frequency 1193182 Hz quality 0
k: CPU: VIA/IDT Unknown (998.51-MHz 686-class CPU)
k: Origin = "CentaurHauls"  Id = 0x6a9  Stepping = 9
k: 
Features=0xa7c9b8ff

k: Features2=0x181
k: real memory  = 469696512 (447 MB)
k: avail memory = 450207744 (429 MB)
k: kbd1 at kbdmux0
k: acpi0:  on motherboard
k: acpi0: Power Button (fixed)
k: Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000
k: acpi_timer0: <24-bit timer at 3.579545MHz> port 0x408-0x40b on acpi0
k: cpu0:  on acpi0
k: acpi_perf0:  on cpu0
k: acpi_throttle0:  on cpu0
k: acpi_button0:  on acpi0
k: acpi_button1:  on acpi0
k: pcib0:  port 0xcf8-0xcff on acpi0
k: pci_link2: BIOS IRQ 5 for 0.16.INTC is invalid
k: pci_link2: BIOS IRQ 5 for 0.17.INTC is invalid
k: pci0:  on pcib0
k: pcib1:  at device 1.0 on pci0
k: pci1:  on pcib1
k: pci1:  at device 0.0 (no driver attached)
k: atapci0:  port 
0xfc00-0xfc07,0xf800-0xf803,0xf400-0xf407,0xf000-0xf003,0xec00-0xec0f,0xe800-0xe8ff 
irq 11 at device 15.0 on pci0

k: ata2:  on atapci0
k: ata3:  on atapci0
k: atapci1:  port 
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xe400-0xe40f at device 15.1 on pci0

k: ata0:  on atapci1
k: ata1:  on atapci1
k: uhci0:  port 0xe000-0xe01f irq 10 at 
device 16.0 on pci0

k: uhci0: [GIANT-LOCKED]
k: usb0:  on uhci0
k: usb0: USB revision 1.0
k: uhub0: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
k: uhub0: 2 ports with 2 removable, self powered
k: uhci1:  port 0xdc00-0xdc1f irq 10 at 
device 16.1 on pci0

k: uhci1: [GIANT-LOCKED]
k: usb1:  on uhci1
k: usb1: USB revision 1.0
k: uhub1: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
k: uhub1: 2 ports with 2 removable, self powered
k: uhci2:  port 0xd800-0xd81f irq 11 at 
device 16.2 on pci0

k: uhci2: [GIANT-LOCKED]
k: usb2:  on uhci2
k: usb2: USB revision 1.0
k: uhub2: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
k: uhub2: 2 ports with 2 removable, self powered
k: uhci3:  port 0xd400-0xd41f irq 11 at 
device 16.3 on pci0

k: uhci3: [GIANT-LOCKED]
k: usb3:  on uhci3
k: usb3: USB revision 1.0
k: uhub3: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1
k: uhub3: 2 ports with 2 removable, self powered
k: ehci0:  mem 0xfdfff000-0xfdfff0ff irq 
9 at device 16.4 on pci0

k: ehci0: [GIANT-LOCKED]
k: usb4: EHCI version 1.0
k: usb4: companion controllers, 2 ports each: usb0 usb1 usb2 usb3
k: usb4:  on ehci0
k: usb4: USB revision 2.0
k: uhub4: VIA EHCI root hub, class 9/0, rev 2.00/1.00, addr 1
k: uhub4: 8 ports with 8 removable, self powered
k: isab0:  at device 17.0 on pci0
k: isa0:  on isab0
k: pcm0:  port 0xd000-0xd0ff irq 9 at device 17.5 on pci0
k: pcm0: 
k: pcm0: 
k: vr0:  port 0xc800-0xc8ff mem 
0xfdffe000-0xfdffe0ff irq 10 at device 18.0 on pci0

k: miibus0:  on vr0
k: ukphy0:  on miibus0
k: ukphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
k: vr0: Ethernet address: 00:40:63:e6:41:ba
k: sio0: <16550A-compatible COM port> port 0x3f8-0x3ff irq 4 flags 0x10 
on acpi0

k: sio0: type 16550A
k: ppc0:  port 0x378-0x37f irq 7 on acpi0
k: ppc0: Generic chipset (EPP/NIBBLE) in COMPATIBLE mode
k: ppbus0:  on ppc0
k: plip0:  on ppbus0
k: lpt0:  on ppbus0
k: lpt0: Interrupt-driven port
k: ppi0:  on ppbus0
k: atkbdc0:  port 0x60,0x64 irq 1 on acpi0
k: atkbd0:  irq 1 on atkbdc0
k: kbd0 at atkbd0
k: atkbd0: [GIANT-LOCKED]
k: psm0:  irq 12 on atkbdc0
k: psm0: [GIANT-LOCKED]
k: psm0: model IntelliMouse Explorer, device ID 4
k: pmtimer0 on isa0
k: sc0:  at flags 0x100 on isa0
k: sc0: VGA <16 virtual consoles, flags=0x300>
k: sio1: configured irq 3 not in bitmap of probed irqs 0
k: sio1: port may not be enabled
k: vga0:  at port 0x3c0-0x3df iomem 0xa-0xb on isa0
k: Timecounter "TSC" frequency 998508102 Hz quality 800
k: Timecounters tick every 1.000 msec
k: ad0: 19541MB  at ata0-master UDMA100
k: acd0: DVDR  at ata1-slave UDMA33
k: Trying to mount root from ufs:/dev/ad0s1a


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Determining disk latency

[Charles Swiger]

On Dec 14, 2005, at 4:56 PM, Robert Fitzpatrick wrote:

Dec 14 14:17:49 esmtp postfix/qmgr[29605]: 7BEE97E997:
from=<[EMAIL PROTECTED]>, size=11246, nrcpt=1 (queue active)
Dec 14 14:27:29 esmtp postfix/qmgr[29605]: 7BEE97E997:
to=<[EMAIL PROTECTED]>, relay=none, delay=592, status=deferred
(delivery temporarily suspended: connect to 127.0.0.1[127.0.0.1]:
Connection refused)


This implies that postfix tried to pass the message off to your local  
content filter and could not connect to it, so the message got queued  
up for the next queue run an hour or two later.


You ought to adjust the number in the second-to-last column in /usr/ 
local/etc/postfix/master.cf for your virus scanning task (often  
called scan), to correspond to the number of amavisd children you are  
running, and you should adjust amavisd to only run as many children  
as your hardware can comfortably handle without swapping.


--
-Chuck

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Determining disk latency

[Robert Fitzpatrick]

I have been trying to track down an issue where are mail messages are
delayed via postfix, amavis, clamav and SA. Seems the messages move
right through amavis now in just a second, like the message below. I'm
sure the queue active messages are related to me restarting postfix
after making changes. I have max_server for amavis set to 2 just like
the smtp-amavis transport in master.cf in postfix. If I try to increase
this, my CPU utilization spikes.

Messages come in as queue active and sit for, in this case 2 and half
hours, until amavis picks it up. Someone suggested on the amavis list to
look for disk latency issues, how do I do this?

esmtp# grep 7BEE97E997 /var/log/maillog
Dec 14 14:17:38 esmtp postfix/smtpd[32427]: 7BEE97E997:
client=mail186.e2ma.net[66.179.147.186]
Dec 14 14:17:38 esmtp postfix/cleanup[33547]: 7BEE97E997:
message-id=<[EMAIL PROTECTED]>
Dec 14 14:17:49 esmtp postfix/qmgr[29605]: 7BEE97E997:
from=<[EMAIL PROTECTED]>, size=11246, nrcpt=1 (queue active)
Dec 14 14:27:29 esmtp postfix/qmgr[29605]: 7BEE97E997:
to=<[EMAIL PROTECTED]>, relay=none, delay=592, status=deferred
(delivery temporarily suspended: connect to 127.0.0.1[127.0.0.1]:
Connection refused)
Dec 14 14:46:25 esmtp postfix/qmgr[35306]: 7BEE97E997:
from=<[EMAIL PROTECTED]>, size=11246, nrcpt=1 (queue active)
Dec 14 14:58:56 esmtp postfix/qmgr[36371]: 7BEE97E997:
from=<[EMAIL PROTECTED]>, size=11246, nrcpt=1 (queue active)
Dec 14 15:25:44 esmtp postfix/qmgr[37488]: 7BEE97E997:
from=<[EMAIL PROTECTED]>, size=11246, nrcpt=1 (queue active)
Dec 14 16:44:02 esmtp postfix/smtp[40014]: 7BEE97E997:
to=<[EMAIL PROTECTED]>, relay=127.0.0.1[127.0.0.1], delay=8785,
status=sent (250 2.6.0 Ok, id=40110-01-4, from MTA([127.0.0.1]:10025):
250 Ok: queued as 4EAA07F5EF)
Dec 14 16:44:02 esmtp postfix/qmgr[37488]: 7BEE97E997: removed
esmtp# grep 4EAA07F5EF /var/log/maillog
Dec 14 16:44:02 esmtp postfix/smtpd[40116]: 4EAA07F5EF:
client=localhost.webtent.net[127.0.0.1]
Dec 14 16:44:02 esmtp postfix/cleanup[39841]: 4EAA07F5EF:
message-id=<[EMAIL PROTECTED]>
Dec 14 16:44:02 esmtp postfix/qmgr[37488]: 4EAA07F5EF:
from=<[EMAIL PROTECTED]>, size=11737, nrcpt=1 (queue active)
Dec 14 16:44:02 esmtp amavis[40110]: (40110-01-4) FWD via SMTP:
<[EMAIL PROTECTED]> -> <[EMAIL PROTECTED]>, 250 2.6.0 Ok,
id=40110-01-4, from MTA([127.0.0.1]:10025): 250 Ok: queued as 4EAA07F5EF
Dec 14 16:44:02 esmtp postfix/smtp[40014]: 7BEE97E997:
to=<[EMAIL PROTECTED]>, relay=127.0.0.1[127.0.0.1], delay=8785,
status=sent (250 2.6.0 Ok, id=40110-01-4, from MTA([127.0.0.1]:10025):
250 Ok: queued as 4EAA07F5EF)
Dec 14 16:44:02 esmtp postfix/smtp[38824]: 4EAA07F5EF:
to=<[EMAIL PROTECTED]>, relay=208.38.145.35[208.38.145.35],
delay=0, status=sent (250 2.0.0 jBELhw220499 Message accepted for
delivery)
Dec 14 16:44:02 esmtp postfix/qmgr[37488]: 4EAA07F5EF: removed

--
Robert

___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

RE: increasing latency on idle box with ASUS P4B533VM board

[Arjan van der Oest]

Dimitry,

vrijdag 19 augustus 2005 16:19, Dmitry Mityugov
 wrote:

> Just a shot in the dark but is HT enabled in that P4?

It's an older MB but as far as I can see in the BIOS and the specs on
asus.com it doesn't support HT.

Thanks,

Arjan
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: increasing latency on idle box with ASUS P4B533VM board

[Dmitry Mityugov]

On 8/19/05, Arjan van der Oest <[EMAIL PROTECTED]> wrote:
> Hi,
> 
> I've build a new P4 box (1.8Ghz, 512MB mem) based upon a ASUS P4B533VM
> motherboard. This board has an Intel 845G chipset and onboard 10/100
> mbit/s LAN (fxp driver).
> 
> When connected directly to a gateway (in my case an Extreme BD10k) I
> observe the following behaviour: a ping to the gateway starts around
> 1ms and slowly increases in steps of .1 or .2ms to around 10ms. Then
> it suddenly drops back to 1ms and the process starts all over again.
> 
> I've tried a different NIC (3com, xl driver), new cabling, new
> switchport and all gives me the same result. My other labbox (a AMD
> 2000 XP+ with 3com PCI nic) connected to the same router is showing
> normal latency. This gives me the feeling it's some sort of timing
> issue on the P4's motherboard itself.
> 
> Has anyone seen this before? I've searched the archives and found
> little on this topic. Any tips/hints where I can seach for further
> clues?

Just a shot in the dark but is HT enabled in that P4?

-- 
Dmitry Mityugov, St. Petersburg, Russia
I ignore all messages with confidentiality statements

"We live less by imagination than despite it" - Rockwell Kent, "N by E"
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

increasing latency on idle box with ASUS P4B533VM board

[Arjan van der Oest]

Hi,

I've build a new P4 box (1.8Ghz, 512MB mem) based upon a ASUS P4B533VM
motherboard. This board has an Intel 845G chipset and onboard 10/100
mbit/s LAN (fxp driver).

When connected directly to a gateway (in my case an Extreme BD10k) I
observe the following behaviour: a ping to the gateway starts around
1ms and slowly increases in steps of .1 or .2ms to around 10ms. Then
it suddenly drops back to 1ms and the process starts all over again.

I've tried a different NIC (3com, xl driver), new cabling, new
switchport and all gives me the same result. My other labbox (a AMD
2000 XP+ with 3com PCI nic) connected to the same router is showing
normal latency. This gives me the feeling it's some sort of timing
issue on the P4's motherboard itself.

Has anyone seen this before? I've searched the archives and found
little on this topic. Any tips/hints where I can seach for further
clues?

Thanks,

Arjan
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

increasing latency on idle box with ASUS P4B533VM board

[Arjan van der Oest]

Hi,

I've build a new P4 box (1.8Ghz, 512MB mem) based upon a ASUS P4B533VM
motherboard. This board has an Intel 845G chipset and onboard 10/100
mbit/s LAN (fxp driver).

When connected directly to a gateway (in my case an Extreme BD10k) I
observe the following behaviour: a ping to the gateway starts around
1ms and slowly increases in steps of .1 or .2ms to around 10ms. Then
it suddenly drops back to 1ms and the process starts all over again.

I've tried a different NIC (3com, xl driver), new cabling, new
switchport and all gives me the same result. My other labbox (a AMD
2000 XP+ with 3com PCI nic) connected to the same router is showing
normal latency. This gives me the feeling it's some sort of timing
issue on the P4's motherboard itself.

Has anyone seen this before? I've searched the archives and found
little on this topic. Any tips/hints where I can seach for further
clues?

Thanks,

Arjan
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Audio latency

[J.E. Dooper]

Well, I guess it's a sound driver problem, because when i tried those 4front
OSS drivers ( http://www.opensound.com ) the latency was gone.

Why can 4front OSS function without latency and freebsd's drivers not?
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Audio latency

[Luyt]

On Thursday 03 March 2005 19:20, J.E. Dooper wrote:

> I think this might be the problem:
> http://lists.freebsd.org/mailman/htdig/freebsd-questions/2004-August/055314
>.html Though I don't understand much about the solution...

This might also be interesting, how sysctl hw.targetirqrate is used:

  http://fxr.watson.org/fxr/search?v=RELENG53;string=targetirqrate

I reduced the audio latency in KDE by using Control Center / Sound & 
Multimedia / Sound System / Skip prevention slider to 100 ms.

--
"The ability of the OSS process to collect and harness 
the collective IQ of thousands of individuals across 
the Internet is simply amazing." - Vinod Vallopillil
http://www.opensource.org/halloween/halloween4.php
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

RE: Audio latency

[Ted Mittelstaedt]

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] Behalf Of J.E. Dooper
> Sent: Thursday, March 03, 2005 10:20 AM
> To: freebsd-questions@freebsd.org
> Subject: Audio latency
> 
> 
> Hi, 
> 
> My sound works and when I use mplayer or xmms I don't experience 
> any (noticable!) audio latency. 
> In applications like doomlegacy and quakeforge I do.
> 
> I think this might be the problem:
> http://lists.freebsd.org/mailman/htdig/freebsd-questions/2004-A
> ugust/055314.html
> Though I don't understand much about the solution...
> 

Quite probably.  You should inform the developers of those
applications of the above.

> 
> My questions are:

> what can I do to fix this?

Inform the developers of the above so they can fix their apps.

Ted
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Audio latency

[J.E. Dooper]

Hi, 

My sound works and when I use mplayer or xmms I don't experience 
any (noticable!) audio latency. 
In applications like doomlegacy and quakeforge I do.

I think this might be the problem:
http://lists.freebsd.org/mailman/htdig/freebsd-questions/2004-August/055314.html
Though I don't understand much about the solution...

Some useful info:  

I'm using FreeBSD5.3-STABLE. Card: nForce2 onboard sound. And I compiled
my kernel with "device sound". And I load the snd_ich.ko module.

The output of `dmesg | grep pcm` :
pcm0:  port 0xd400-0xd47f,0xd000-0xd0ff mem 
0xe708-0xe7080fff irq 21 at device 6.0 on pci0
pcm0: [GIANT-LOCKED]
pcm0: 

The output of `cat /dev/sndstat` :
FreeBSD Audio Driver (newpcm)
Installed devices:
pcm0:  at io 0xd000, 0xd400 irq 21 bufsz 16384 kld snd_ich 
(1p/1r/0v channels duplex default)

The output of `sysctl -a | grep pcm` :
hw.snd.pcm0.buffersize: 16384
hw.snd.pcm0.vchans: 0
hw.snd.pcm0.ac97rate: 48000
dev.pcm.0.%desc: nVidia nForce2
dev.pcm.0.%driver: pcm
dev.pcm.0.%location: slot=6 function=0 handle=\_SB_.PCI0.MACI
dev.pcm.0.%pnpinfo: vendor=0x10de device=0x006a subvendor=0x1695 
subdevice=0x100 
  0 class=0x040100
dev.pcm.0.%parent: pci0

The output of `sysctl -a | grep snd`:
hw.snd.targetirqrate: 128
hw.snd.report_soft_formats: 1
hw.snd.verbose: 1
hw.snd.unit: 0
hw.snd.maxautovchans: 0
hw.snd.pcm0.buffersize: 16384
hw.snd.pcm0.vchans: 0
hw.snd.pcm0.ac97rate: 48000

My questions are:
What could be causing this latency, and what can I do to fix this?

Regards,
Jorma
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: high latency

[Skylar Thompson]

Peter Risdon wrote:
On Mon, 2005-02-21 at 11:52 +0200, Chris Knipe wrote:
 

Hi,
I have 4 FreeBSD Servers connected to a Cisco 2950 all doing inter-VLAN 
routing.  Everything is working right, but one server is getting absurdly 
high latency through the VLANs.
   

[...]
 

Now, the problematic box is running a RealTek card,
   

Ugh. You don't say which one, but I just cured a prob with nfs by
replacing a brand new Realtek card with an old digital NIC that was
lying around. Not for the first time, I vow never to use Realtek cards
again.
 

Not for naught is this comment in the source for the FreeBSD driver:
The RealTek 8139 PCI NIC redefines the meaning of 'low end.'
I've had nothing but bad luck from them.
--
-- Skylar Thompson ([EMAIL PROTECTED])
-- http://www.cs.earlham.edu/~skylar/


signature.asc
Description: OpenPGP digital signature

RE: high latency

[Peter Risdon]

On Mon, 2005-02-21 at 03:48 -0800, Ted Mittelstaedt wrote:
> > 
> > On Mon, 2005-02-21 at 11:52 +0200, Chris Knipe wrote:
> > > Hi,
> > > 
> > > I have 4 FreeBSD Servers connected to a Cisco 2950 all doing 
> > inter-VLAN 
> > > routing.  Everything is working right, but one server is 
> > getting absurdly 
> > > high latency through the VLANs.
> > 
> > [...]
> > 
> > > Now, the problematic box is running a RealTek card,
> > 
> > Ugh. You don't say which one, but I just cured a prob with nfs by
> > replacing a brand new Realtek card with an old digital NIC that was
> > lying around. Not for the first time, I vow never to use Realtek cards
> > again.
> > 
> 
> Realteks seem to have a problem detecting and setting up for the correct
> speed/duplex.  (The 10/100 versions, that is)  If you leave them in AUTO
> negotiation mode they only work right if connected to a 10/100 switch.
> Otherwise if you have problems they work fine if you hard code the
> speed/duplex in the /etc/rc.conf file.

I still feel an aversion to the rl cards, but thanks very much for the
tip.

Peter.


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

RE: high latency

[Ted Mittelstaedt]

> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] Behalf Of Peter Risdon
> Sent: Monday, February 21, 2005 3:31 AM
> To: Chris Knipe
> Cc: freebsd-questions@freebsd.org
> Subject: Re: high latency
> 
> 
> On Mon, 2005-02-21 at 11:52 +0200, Chris Knipe wrote:
> > Hi,
> > 
> > I have 4 FreeBSD Servers connected to a Cisco 2950 all doing 
> inter-VLAN 
> > routing.  Everything is working right, but one server is 
> getting absurdly 
> > high latency through the VLANs.
> 
> [...]
> 
> > Now, the problematic box is running a RealTek card,
> 
> Ugh. You don't say which one, but I just cured a prob with nfs by
> replacing a brand new Realtek card with an old digital NIC that was
> lying around. Not for the first time, I vow never to use Realtek cards
> again.
> 

Realteks seem to have a problem detecting and setting up for the correct
speed/duplex.  (The 10/100 versions, that is)  If you leave them in AUTO
negotiation mode they only work right if connected to a 10/100 switch.
Otherwise if you have problems they work fine if you hard code the
speed/duplex in the /etc/rc.conf file.

Ted
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Re: high latency

[Chris Knipe]

Hi,
I have 4 FreeBSD Servers connected to a Cisco 2950 all doing inter-VLAN
routing.  Everything is working right, but one server is getting absurdly
high latency through the VLANs.
[...]
Now, the problematic box is running a RealTek card,
Ugh. You don't say which one, but I just cured a prob with nfs by
replacing a brand new Realtek card with an old digital NIC that was
lying around. Not for the first time, I vow never to use Realtek cards
again.
Yup.  Switched the card and problem solved.  I use RealTek mainly because I
have heard that they have the best support for VLANs with large packets
inside those VLANs.
Oh well, we'll be running only cisco on the networking side pretty soon,
then there wont be a use for VLANs on the servers :)
Thanks
--
chris.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: high latency

[Peter Risdon]

On Mon, 2005-02-21 at 11:52 +0200, Chris Knipe wrote:
> Hi,
> 
> I have 4 FreeBSD Servers connected to a Cisco 2950 all doing inter-VLAN 
> routing.  Everything is working right, but one server is getting absurdly 
> high latency through the VLANs.

[...]

> Now, the problematic box is running a RealTek card,

Ugh. You don't say which one, but I just cured a prob with nfs by
replacing a brand new Realtek card with an old digital NIC that was
lying around. Not for the first time, I vow never to use Realtek cards
again.

Peter.


___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

high latency

[Chris Knipe]

Hi,
I have 4 FreeBSD Servers connected to a Cisco 2950 all doing inter-VLAN 
routing.  Everything is working right, but one server is getting absurdly 
high latency through the VLANs.

problem box:
[EMAIL PROTECTED]:/home/cknipe# ping 198.19.0.1
PING 198.19.0.1 (198.19.0.1): 56 data bytes
64 bytes from 198.19.0.1: icmp_seq=0 ttl=64 time=1020.571 ms
64 bytes from 198.19.0.1: icmp_seq=2 ttl=64 time=1114.468 ms
64 bytes from 198.19.0.1: icmp_seq=3 ttl=64 time=934.580 ms
64 bytes from 198.19.0.1: icmp_seq=4 ttl=64 time=814.296 ms
64 bytes from 198.19.0.1: icmp_seq=5 ttl=64 time=682.657 ms
64 bytes from 198.19.0.1: icmp_seq=6 ttl=64 time=1173.596 ms
64 bytes from 198.19.0.1: icmp_seq=7 ttl=64 time=1212.085 ms
64 bytes from 198.19.0.1: icmp_seq=8 ttl=64 time=1021.996 ms
64 bytes from 198.19.0.1: icmp_seq=9 ttl=64 time=826.783 ms
64 bytes from 198.19.0.1: icmp_seq=10 ttl=64 time=372.565 ms
^C
--- 198.19.0.1 ping statistics ---
12 packets transmitted, 10 packets received, 16% packet loss
round-trip min/avg/max/stddev = 372.565/917.360/1212.085/241.657 ms
second box:
[EMAIL PROTECTED]:~# ping 198.19.0.1
PING 198.19.0.1 (198.19.0.1): 56 data bytes
64 bytes from 198.19.0.1: icmp_seq=0 ttl=64 time=1.847 ms
64 bytes from 198.19.0.1: icmp_seq=1 ttl=64 time=1.484 ms
64 bytes from 198.19.0.1: icmp_seq=2 ttl=64 time=1.478 ms
64 bytes from 198.19.0.1: icmp_seq=3 ttl=64 time=1.564 ms
64 bytes from 198.19.0.1: icmp_seq=4 ttl=64 time=1.913 ms
64 bytes from 198.19.0.1: icmp_seq=5 ttl=64 time=3.057 ms
64 bytes from 198.19.0.1: icmp_seq=6 ttl=64 time=1.839 ms
64 bytes from 198.19.0.1: icmp_seq=7 ttl=64 time=1.526 ms
64 bytes from 198.19.0.1: icmp_seq=8 ttl=64 time=2.791 ms
64 bytes from 198.19.0.1: icmp_seq=9 ttl=64 time=1.522 ms
^C
--- 198.19.0.1 ping statistics ---
10 packets transmitted, 10 packets received, 0% packet loss
round-trip min/avg/max/stddev = 1.478/1.902/3.057/0.537 ms
The other's ping basically the same (1.4ms -> 4ms).
Now, the problematic box is running a RealTek card, netstat -bin reports no 
input / output errors.  The interface on the Cisco 2950 also doesn't report 
any problems or errors on the interface.  Does anyone have some 
recommendations?  I'm thinking of just switching the NIC out, but I'd rather 
want to make sure first that is actually the problem.

--
Chris.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD pcm(4) latency (From write() to audible output)

[Peter Wood]

Heya Dan,
I've found that there is about a 800ms delay between the output data
being written to /dev/dsp and being able to hear the output from the
soundcard on FreeBSD. I'm working to a 200ms deadline.
What's your write size?  At 44100khz*2 channels, a 65k write would take
around 800 ms to play.  Try writing smaller chunks, or maybe raise
hw.snd.targetirqrate.
Don't I feel like a dunce right now. I forgot the cardinal rule of 
reproducing my results with other software. I tried the same file with 
XMMS and it played instantly.

My problem was primarly due to the fragment sizes of buffers. I don't 
know what sizes are set as default but they're to big for my 
applpications. I've found with:

val = (2 << 16) | 9;
ioctl(audio_fd, SNDCTL_DSP_SETFRAGMENT, &val);
It kills the delay to next to nothing. As far as I understand this, and 
it's not well, the above means use a maximum of 2 fragments and each 
fragment should be 2^9 large (ie 512 bytes long).

Which I figure to be 1204 bytes total for buffer (or 5ms).
It still doesn't happen that quickly for me, but what I'd imagin is 
delaying it now (and is my problem) is the write() I'm doing. As I'm 
writing 20ms of Audio out at any time, so I'm waiting for that to complete.

Thanks for the suggestion Dan, helped a lot.
Andreas, this will be your problem to for your guitar problem, try using 
the above code (add error checks, it can fail on some hardware 
(statically set fragments)).

I also managed to find the guide to OSS at 
http://www.opensound.com/pguide/oss.pdf, I shall be looking there before 
asking here again, don't worry.

Off to knock the dust of my dunce cap off,
Pete.
--
Peter Wood BSc (Hons) :: <[EMAIL PROTECTED]> :: Tel +44 7974 799440
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD pcm(4) latency (From write() to audible output)

[Dan Nelson]

In the last episode (Aug 13), Peter Wood said:
> I'm currently in the middle of writing an automated radio playout
> system for a northern student radio station in the UK (post SBN
> liquidation).
> 
> I have the requirement of it running on a *nix system. I'm a large
> BSD advocate between our friends, so would prefer not to use Linux.
> 
> The system design consists of a daemon for each soundcard used in the
> studio. I'm currently in the process of writing this daemon.
> 
> I've found that there is about a 800ms delay between the output data
> being written to /dev/dsp and being able to hear the output from the
> soundcard on FreeBSD. I'm working to a 200ms deadline.
> 
> I'm opening the soundcard with:
> 
> audio_fd = open((char *) device, O_WRONLY | O_FSYNC | O_DIRECT);
> 
> I've used O_DIRECT and O_FSYNC to try and get rid of this delay, but
> alas that didn't work, device is cast as it's coming from a void *
> (thanks to pthread_create ;).

What's your write size?  At 44100khz*2 channels, a 65k write would take
around 800 ms to play.  Try writing smaller chunks, or maybe raise
hw.snd.targetirqrate.

-- 
Dan Nelson
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

FreeBSD pcm(4) latency (From write() to audible output)

[Peter Wood]

Moved from freebsd-chat:
Quick addition: Had a look through the questions archive, nothing jumped 
out and answered this, but it's possible I missed something.

Good Morning,
I'm currently in the middle of writing an automated radio playout system
for a northern student radio station in the UK (post SBN liquidation).
I have the requirement of it running on a *nix system. I'm a large BSD
advocate between our friends, so would prefer not to use Linux.
The system design consists of a daemon for each soundcard used in the
studio. I'm currently in the process of writing this daemon.
I've found that there is about a 800ms delay between the output data
being written to /dev/dsp and being able to hear the output from the
soundcard on FreeBSD. I'm working to a 200ms deadline.
I'm opening the soundcard with:
audio_fd = open((char *) device, O_WRONLY | O_FSYNC | O_DIRECT);
I've used O_DIRECT and O_FSYNC to try and get rid of this delay, but
alas that didn't work, device is cast as it's coming from a void *
(thanks to pthread_create ;).
The daemon outputs blocks of 0 when there is no actual audio to output
so the sound card is always fed.
While this isn't a problem for mpg123, it is for this application.  I've
tried to find where the delay is, however I haven't had any luck.
Could anyone knowledgable let me know if there's any hope for getting
rid of it (hacking the kernel is fine ;) or if I should "give up" and go
to linux?
I suppose Linux does have ALSA which claims 2.6ms latency but I was
hoping to just use OSS. Oh well.
For reference the test machine is a 2.2Ghz machine with a AC97 chip
(yeah crap I know, studio playout machines will probably have a Creative
card) running 4.10-STABLE (Yeasturday's cvsup).
My appologies if this isn't the right place, I'm happy to go pester else
where. Heh.
Many thanks,
Pete.
--
Peter Wood BSc (Hons) :: <[EMAIL PROTECTED]> :: Tel +44 7974 799440
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Simulating network latency

[Mike Jeays]

On Fri, 2004-05-14 at 19:05, Kevin Stevens wrote:
> On Fri, 14 May 2004, Mike Jeays wrote:
> 
> > Is there a way to set up a machine with two network cards, which will
> > simply forward every packet from one card to the other, but will
> > introduce an arbitrary delay period?  Ideally, the delay period should
> > be adjustable, and optionally different in the two directions.
> 
> Yes, man dummynet, and google for examples of people doing this very
> thing.  We've done it in our lab here for that purpose.
> 
> KeS
> ___
> [EMAIL PROTECTED] mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Perfect.  Could have been written to our specifications.  Thanks very
much.

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Simulating network latency

[Kevin Stevens]

On Fri, 14 May 2004, Mike Jeays wrote:

> Is there a way to set up a machine with two network cards, which will
> simply forward every packet from one card to the other, but will
> introduce an arbitrary delay period?  Ideally, the delay period should
> be adjustable, and optionally different in the two directions.

Yes, man dummynet, and google for examples of people doing this very
thing.  We've done it in our lab here for that purpose.

KeS
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Simulating network latency

[Mike Jeays]

We have a proprietary application which will not work properly over our
WAN, even though there is more than enough bandwidth. It works fine on a
LAN.   We suspect a latency problem, mainly because we can't think of
anything else it might be. We are not getting any help from the vendor.

Is there a way to set up a machine with two network cards, which will
simply forward every packet from one card to the other, but will
introduce an arbitrary delay period?  Ideally, the delay period should
be adjustable, and optionally different in the two directions.

Or is there some other way to do the same thing? 

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Latency problem with traffic shaping (ipfw/dummynet)

[Vincent Poy]

On Sun, 21 Mar 2004, Luigi Rizzo wrote:

> On Sat, Mar 20, 2004 at 12:56:08PM -1000, Vincent Poy wrote:
> ...
> > > the above configuration means that if queue 1 is getting a bandwidth
> > > X, then queue 2 will get 0.99X, queue 3 will get 0.98X, queue
> > > 4 will get 0.97X. Hardly matching any reasonable definition of high-mid-low
> > > priority!
> >
> > Hmm, I think I did it that way because 100 is the largest number
> > and I didn't decide on how many queues I may add later so the numbers will
> > change but does the weight number really mean 99%, 98%, 97% priority?  So
> > should it really be 66, 33, and 1?
>
> no, the weights mean exactly what i wrote above, and they
> are weights not priorities. As to the values to use,
> that's entirely up to you.

Just as I thought.  I rebooted and latencies have gone down.  It
seems that latency when the pipes are filled are always 50-100ms slower on
boxes behind the FreeBSD box.  One question though, with ipfw pipe show or
ipfw queue show, is it supposed to show all traffic that matches the queue
rule or just only one?


Cheers,
Vince - [EMAIL PROTECTED] - Vice President    __ 
Unix Networking Operations - FreeBSD-Real Unix for Free / / / / |  / |[__  ]
WurldLink Corporation  / / / /  | /  | __] ]
San Francisco - Honolulu - Hong Kong  / / / / / |/ / | __] ]
HongKong Stars/Gravis UltraSound Mailing Lists Admin /_/_/_/_/|___/|_|[]
[EMAIL PROTECTED] - oahu.DAL.NET Hawaii's DALnet IRC Network Server Admin

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Latency problem with traffic shaping (ipfw/dummynet)

[Luigi Rizzo]

On Sat, Mar 20, 2004 at 12:56:08PM -1000, Vincent Poy wrote:
...
> > the above configuration means that if queue 1 is getting a bandwidth
> > X, then queue 2 will get 0.99X, queue 3 will get 0.98X, queue
> > 4 will get 0.97X. Hardly matching any reasonable definition of high-mid-low
> > priority!
> 
>   Hmm, I think I did it that way because 100 is the largest number
> and I didn't decide on how many queues I may add later so the numbers will
> change but does the weight number really mean 99%, 98%, 97% priority?  So
> should it really be 66, 33, and 1?

no, the weights mean exactly what i wrote above, and they
are weights not priorities. As to the values to use,
that's entirely up to you.

cheers
luigi
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Latency problem with traffic shaping (ipfw/dummynet)

[Vincent Poy]

On Sat, 20 Mar 2004, Luigi Rizzo wrote:

> cannot comment on the reason for the huge delay (but one
> way to check what is going on is to change the pipe's bandwidth
> and see if anything changes), but i see a big
> misunderstanding on weights vs. priorities in your
> configuration:

The delay only seems to be coming from machines behind the FreeBSD
box and not the FreeBSD box itself since every box has static IP's, only
the outgoing is via the FreeBSD box but the downstream is direct from the
modem through the switch and then the machines directly.

> > # Define our upload pipe
> > ${fwcmd} pipe 1 config bw 480Kbit/s
> > # Define a high-priority queue
> > ${fwcmd} queue 1 config pipe 1 weight 100
> > # Define a medium-high-priority queue
> > ${fwcmd} queue 2 config pipe 1 weight 99
> > # Define a medium-low-priority queue
> > ${fwcmd} queue 3 config pipe 1 weight 98
> > # Define a low-priority queue
> > ${fwcmd} queue 4 config pipe 1 weight 97
>
> the above configuration means that if queue 1 is getting a bandwidth
> X, then queue 2 will get 0.99X, queue 3 will get 0.98X, queue
> 4 will get 0.97X. Hardly matching any reasonable definition of high-mid-low
> priority!

Hmm, I think I did it that way because 100 is the largest number
and I didn't decide on how many queues I may add later so the numbers will
change but does the weight number really mean 99%, 98%, 97% priority?  So
should it really be 66, 33, and 1?


Cheers,
Vince - [EMAIL PROTECTED] - Vice President    __ 
Unix Networking Operations - FreeBSD-Real Unix for Free / / / / |  / |[__  ]
WurldLink Corporation  / / / /  | /  | __] ]
San Francisco - Honolulu - Hong Kong  / / / / / |/ / | __] ]
HongKong Stars/Gravis UltraSound Mailing Lists Admin /_/_/_/_/|___/|_|[]
[EMAIL PROTECTED] - oahu.DAL.NET Hawaii's DALnet IRC Network Server Admin




___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Latency problem with traffic shaping (ipfw/dummynet)

[Luigi Rizzo]

cannot comment on the reason for the huge delay (but one
way to check what is going on is to change the pipe's bandwidth
and see if anything changes), but i see a big
misunderstanding on weights vs. priorities in your
configuration:

> # Define our upload pipe
> ${fwcmd} pipe 1 config bw 480Kbit/s
> # Define a high-priority queue
> ${fwcmd} queue 1 config pipe 1 weight 100
> # Define a medium-high-priority queue
> ${fwcmd} queue 2 config pipe 1 weight 99
> # Define a medium-low-priority queue
> ${fwcmd} queue 3 config pipe 1 weight 98
> # Define a low-priority queue
> ${fwcmd} queue 4 config pipe 1 weight 97

the above configuration means that if queue 1 is getting a bandwidth
X, then queue 2 will get 0.99X, queue 3 will get 0.98X, queue
4 will get 0.97X. Hardly matching any reasonable definition of high-mid-low
priority!

cheers
luigi
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Latency problem with traffic shaping (ipfw/dummynet)

[Vincent Poy]

On this subject, I have one of my own...  I have a
6.016Mbps/608kbps ADSL connection with 8 static IP's from my ISP.  I'm
using the FreeBSD box to basically limit my upstream bandwidth to 480kbps
so that the downloads would work while uploading.  In my kernel, I do have
the following options:

options IPFIREWALL  #firewall
options IPDIVERT#divert sockets
options DUMMYNET
options BRIDGE
options HZ=1000
options NMBCLUSTERS=65536

The 8 IP's I'm using is 208.204.244.224-231 on a /24 block with
the gateway on the other side at my ISP being 208.204.244.1.  The FreeBSD
machine is 208.204.244.224 and I do have gateway ip forwarding enabled. My
problem is that while as far as speeds are concerned, it's working
correctly on both the .224 (FreeBSD box) as well as the .225-.231 boxes
behind it.  The issue is that tracerouting from any box other than the
FreeBSD box shows latencies of 1000+ms after the FreeBSD router beginning
with hop 2 when the upstream pipe is being used while the FreeBSD box
shows the latency at 40-50ms which is correct under traffic load.  Anyone
knows what's causing this or is this the way it's supposed to work?  All
the machines are pointing to .224 (FreeBSD box) as the gateway.  All local
traffic doesn't go through dummynet's queues.  This is how I have ipfw
configured.

setup_loopback
# Traffic Shaping for DSL connection 6.016Mbps/608Kbps
# Make packets exiting dummynet not continue down the chain
# If this is not enabled, then packets leaving an early
# queue might enter a later queue if the conditions for
# the later queue are met, which would be completely
# devastating to all the prioritizing we're doing
${fwcmd} enable one_pass
# Add rules so that local routable IP LAN traffic does not use natd
${fwcmd} add 39 divert natd all from 10.0.0.0/8 to any via ${natd_interface}
${fwcmd} add 40 divert natd all from 172.16.0.0/12 to any via ${natd_interface}
${fwcmd} add 41 divert natd all from 192.168.0.0/16 to any via 
${natd_interface}
${fwcmd} add 42 divert natd all from 208.201.244.224/29 to 10.0.0.0/8 via 
${natd_interface}
${fwcmd} add 43 divert natd all from 208.201.244.224/29 to 172.16.0.0/12 via 
${natd_interface}
${fwcmd} add 44 divert natd all from 208.201.244.224/29 to 192.168.0.0/16 via 
${natd_interface}
${fwcmd} add 45 divert natd all from any to 10.0.0.0/8 via ${natd_interface}
${fwcmd} add 46 divert natd all from any to 172.16.0.0/12 via ${natd_interface}
${fwcmd} add 47 divert natd all from any to 192.168.0.0/16 via 
${natd_interface}
${fwcmd} add 48 divert natd all from any to 208.201.244.224/29 via 
${natd_interface}
${fwcmd} add 49 skipto 100 ip from 208.201.244.224/29 to any
${fwcmd} add 50 divert natd all from any to any via ${natd_interface}
${fwcmd} add 100 pass all from any to any via lo0
${fwcmd} add 200 deny all from any to 127.0.0.0/8
${fwcmd} add 300 deny ip from 127.0.0.0/8 to any
# Route LAN and RFC1918 networks without Traffic Shaping
${fwcmd} add 63000 allow all from any to 10.0.0.0/8 out
${fwcmd} add 63001 allow all from any to 172.16.0.0/12 out
${fwcmd} add 63002 allow all from any to 192.168.0.0/16 out
${fwcmd} add 63003 allow all from any to 208.201.244.224/29 out
# Define our upload pipe
${fwcmd} pipe 1 config bw 480Kbit/s
# Define a high-priority queue
${fwcmd} queue 1 config pipe 1 weight 100
# Define a medium-high-priority queue
${fwcmd} queue 2 config pipe 1 weight 99
# Define a medium-low-priority queue
${fwcmd} queue 3 config pipe 1 weight 98
# Define a low-priority queue
${fwcmd} queue 4 config pipe 1 weight 97
# Assign outgoing empty/small ACK packets to the high-priority queue
${fwcmd} add 63004 set 0 queue 1 tcp from any to any tcpflags ack iplen 0-80 
out
# Assign outgoing UDP (DNS/gaming) and SSH traffic to the medium-high-priority queue
${fwcmd} add 63005 set 0 queue 2 tcp from any to any 22,23 out
${fwcmd} add 63006 set 0 queue 2 udp from any to any not 80,443 out
# Assign outgoing HTTP/HTTPS WEB traffic to the medium-low-priority queue
${fwcmd} add 63007 set 0 queue 3 all from any to any 80,443 out
# Assign all other outgoing traffic to the low-priority queue
${fwcmd} add 63008 set 0 queue 4 all from any to any out
# End of Traffic Shaping
${fwcmd} add 65000 pass all from any to any

This is what the latencies look like on the machines behind the
FreeBSD router when there is a upload:

Tracing route to wurldlink.net [66.193.144.22]
over a maximum of 30 hops:

  1<1 ms<1 ms<1 ms  adsl-208-201-244-224.sonic.net [208.201.244.224]
  2   915 ms   933 ms  1025 ms  adsl-208-201-244-1.sonic.net [208.201.244.1]
  3  1082 ms  1015 

Re: FreeBSD box as router adding latency

[Aloha Guy]

Chris Dillon <[EMAIL PROTECTED]> wrote: On Thu, 26 Feb 2004, Aloha Guy wrote:

> Already tried that and it did improve things a little. I tried
> setting the HZ to 1000 and it didn't make much of a difference. Is
> there a larger number that actually works well?

You can try higher HZ numbers, but you might run into other problems.
Experiment and see. Others have experimented with higher HZ numbers
so you might want to check the list archives. Anyway, is a 1ms delay
really that bad?

The 1ms delay isn't that bad if it was 1ms but we're talking about 3-4ms atleast. As 
for HZ numbers, what should I search for in the archives and on which list since it 
seems like HZ is also in the dmesg output for the clock generator so it's one of those 
terms that are used widely. Thanks, John

-
Do you Yahoo!?
Get better spam protection with Yahoo! Mail
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Chris Dillon]

On Thu, 26 Feb 2004, Aloha Guy wrote:

> Already tried that and it did improve things a little. I tried
> setting the HZ to 1000 and it didn't make much of a difference.  Is
> there a larger number that actually works well?

You can try higher HZ numbers, but you might run into other problems.
Experiment and see.  Others have experimented with higher HZ numbers
so you might want to check the list archives.  Anyway, is a 1ms delay
really that bad?


-- 
 Chris Dillon - cdillon(at)wolves.k12.mo.us
 FreeBSD: The fastest, most open, and most stable OS on the planet
 - Available for IA32, IA64, AMD64, PC98, Alpha, and UltraSPARC architectures
 - PowerPC, ARM, MIPS, and S/390 under development
 - http://www.freebsd.org

Q: Because it reverses the logical flow of conversation.
A: Why is putting a reply at the top of the message frowned upon?

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Aloha Guy]

Charles Swiger <[EMAIL PROTECTED]> wrote: On Feb 26, 2004, at 5:59 PM, Aloha Guy wrote:
> Charles Swiger wrote:
>> There's your issue right there: if you care about the millisecond 
>> level
>> granularity of network traffic going by this router, you ought to set
>> HZ to 1000 as documented in "man dummynet".
[ ... ]
> Knew I forgot to read something.  I guess I forgot all about dummynet 
> is the one doing the traffic shaping as I never used traffic shaping 
> on the other boxes when they were used as both Ethernet and T1 
> routers.  I've always had NMBCLUSERS set to 32768 which I assume is 
> fine.

Thats a lot of NMBCLUSTERS, but if you've got the memory you should be 
okay.

> Also, is there a way to use two NICs like a xl0 and a fxp0 and bond 
> them together with just one IP?

Yes, netgraph. See "man ng_one2many"

I actually had the NMBCLUSTERS set that way even with 128MB boxes without issues but 
the box in question has 2GB of ram so it's not much of a big deal.  I tried the 
ng_one2many and it did help bring things closer to 80Mbps from 60Mbps.  I guess the HD 
is the bottleneck as it's only a notebook and even with the 7200rpm 60GB 2.5" drive, 
the sustained transfer rate is limited.  Tried the HZ 1000 setting and recompiled a 
new kernel but it didn't really seem to do anything at all.  I'm wondering what's the 
highest setting it will work with.Thanks,John

-
Do you Yahoo!?
Get better spam protection with Yahoo! Mail
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Aloha Guy]

Chris Dillon <[EMAIL PROTECTED]> wrote: On Thu, 26 Feb 2004, Aloha Guy wrote:

> > What do you have HZ set to (see sysctl kern.clockrate)? I think I
> > remember your original message showing you using pipes and queues
> > and the HZ setting can affect those. Also see if your latency
> > improves if you remove all pipe and queue rules (other ipfw rules
> > are OK).
>
>
> Here is the HZ setting:
>
> kern.clockrate: { hz = 100, tick = 1, profhz = 1024, stathz = 128 }
>
> I'm not sure how to remove the pipe since I don't think the pipe
> works until the queue is defined. When I removed the queues that
> are configured for the pipe, the latency is back to normal though.

Like I said, remove both pipes and queues to test. However, pipes
_can_ be used without queues, but that is irrelevant here. Try
setting HZ to 1000 in your kernel config, recompile, reboot, and test
again. You should see something between a slight improvement to a
ten-fold improvement.


Already tried that and it did improve things a little. I tried setting the HZ to 1000 
and it didn't make much of a difference.  Is there a larger number that actually works 
well?

Thanks,

John


-
Do you Yahoo!?
Get better spam protection with Yahoo! Mail
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Chris Dillon]

On Wed, 25 Feb 2004, Aloha Guy wrote:

> You're right that additional delay while adding a hop is to be
> expected, which is less than 0.1ms to the FreeBSD box but everything
> past the FreeBSD machine is adding atleast 5ms up to 300ms in the
> traceroutes when the normal is no more than 20ms for the same
> traceroute.  I've already checked the NICs and they are all
> configured at their full rated speeds and full duplex.  I even try
> using a Cardbus PCMCIA fxp0 Intel Pro/100S card on the FreeBSD box
> and it still had the same problem.  I am using a September 2003
> -CURRENT so I don't know if it's a issue with the current networking
> code back then or not.

What do you have HZ set to (see sysctl kern.clockrate)?  I think I
remember your original message showing you using pipes and queues and
the HZ setting can affect those.  Also see if your latency improves if
you remove all pipe and queue rules (other ipfw rules are OK).

-- 
 Chris Dillon - cdillon(at)wolves.k12.mo.us
 FreeBSD: The fastest, most open, and most stable OS on the planet
 - Available for IA32, IA64, AMD64, PC98, Alpha, and UltraSPARC architectures
 - PowerPC, ARM, MIPS, and S/390 under development
 - http://www.freebsd.org

Q: Because it reverses the logical flow of conversation.
A: Why is putting a reply at the top of the message frowned upon?

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Aloha Guy]

Chris Dillon <[EMAIL PROTECTED]> wrote: 
On Wed, 25 Feb 2004, Aloha Guy wrote:

> You're right that additional delay while adding a hop is to be
> expected, which is less than 0.1ms to the FreeBSD box but everything
> past the FreeBSD machine is adding atleast 5ms up to 300ms in the
> traceroutes when the normal is no more than 20ms for the same
> traceroute. I've already checked the NICs and they are all
> configured at their full rated speeds and full duplex. I even try
> using a Cardbus PCMCIA fxp0 Intel Pro/100S card on the FreeBSD box
> and it still had the same problem. I am using a September 2003
> -CURRENT so I don't know if it's a issue with the current networking
> code back then or not.

What do you have HZ set to (see sysctl kern.clockrate)? I think I
remember your original message showing you using pipes and queues and
the HZ setting can affect those. Also see if your latency improves if
you remove all pipe and queue rules (other ipfw rules are OK).


Here is the HZ setting: 

kern.clockrate: { hz = 100, tick = 1, profhz = 1024, stathz = 128 }

I'm not sure how to remove the pipe since I don't think the pipe works until the queue 
is defined.  When I removed the queues that are configured for the pipe, the latency 
is back to normal though. 

Thanks,

John


-
Do you Yahoo!?
Get better spam protection with Yahoo! Mail
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Chris Dillon]

On Thu, 26 Feb 2004, Aloha Guy wrote:

> > What do you have HZ set to (see sysctl kern.clockrate)? I think I
> > remember your original message showing you using pipes and queues
> > and the HZ setting can affect those. Also see if your latency
> > improves if you remove all pipe and queue rules (other ipfw rules
> > are OK).
>
>
> Here is the HZ setting:
>
> kern.clockrate: { hz = 100, tick = 1, profhz = 1024, stathz = 128 }
>
> I'm not sure how to remove the pipe since I don't think the pipe
> works until the queue is defined.  When I removed the queues that
> are configured for the pipe, the latency is back to normal though.

Like I said, remove both pipes and queues to test.  However, pipes
_can_ be used without queues, but that is irrelevant here.  Try
setting HZ to 1000 in your kernel config, recompile, reboot, and test
again.  You should see something between a slight improvement to a
ten-fold improvement.


-- 
 Chris Dillon - cdillon(at)wolves.k12.mo.us
 FreeBSD: The fastest, most open, and most stable OS on the planet
 - Available for IA32, IA64, AMD64, PC98, Alpha, and UltraSPARC architectures
 - PowerPC, ARM, MIPS, and S/390 under development
 - http://www.freebsd.org

Q: Because it reverses the logical flow of conversation.
A: Why is putting a reply at the top of the message frowned upon?

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Charles Swiger]

On Feb 26, 2004, at 4:53 PM, Aloha Guy wrote:
Here is the HZ setting:

kern.clockrate: { hz = 100, tick = 1, profhz = 1024, stathz = 128 }
There's your issue right there: if you care about the millisecond level
granularity of network traffic going by this router, you ought to set
HZ to 1000 as documented in "man dummynet".
--
-Chuck
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Aloha Guy]

Charles Swiger <[EMAIL PROTECTED]> wrote: On Feb 26, 2004, at 4:53 PM, Aloha Guy wrote:
> Here is the HZ setting:
>
> kern.clockrate: { hz = 100, tick = 1, profhz = 1024, stathz = 128 }

There's your issue right there: if you care about the millisecond level
granularity of network traffic going by this router, you ought to set
HZ to 1000 as documented in "man dummynet".

-- 
-Chuck


Knew I forgot to read something.  I guess I forgot all about dummynet is the one doing 
the traffic shaping as I never used traffic shaping on the other boxes when they were 
used as both Ethernet and T1 routers.  I've always had NMBCLUSERS set to 32768 which I 
assume is fine.  Also, is there a way to use two NICs like a xl0 and a fxp0 and bond 
them together with just one IP?

John

 


-
Do you Yahoo!?
Get better spam protection with Yahoo! Mail
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Charles Swiger]

On Feb 26, 2004, at 5:59 PM, Aloha Guy wrote:
Charles Swiger <[EMAIL PROTECTED]> wrote:
There's your issue right there: if you care about the millisecond 
level
granularity of network traffic going by this router, you ought to set
HZ to 1000 as documented in "man dummynet".
[ ... ]
Knew I forgot to read something.  I guess I forgot all about dummynet 
is the one doing the traffic shaping as I never used traffic shaping 
on the other boxes when they were used as both Ethernet and T1 
routers.  I've always had NMBCLUSERS set to 32768 which I assume is 
fine.
Thats a lot of NMBCLUSTERS, but if you've got the memory you should be 
okay.

Also, is there a way to use two NICs like a xl0 and a fxp0 and bond 
them together with just one IP?
Yes, netgraph.  See "man ng_one2many"

--
-Chuck
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Aloha Guy]

Chris Dillon <[EMAIL PROTECTED]> wrote: 
On Wed, 25 Feb 2004, Aloha Guy wrote:

> Any ideas what is causing this? Is it the xl0 driver because I've
> used FreeBSD machines as ethernet routers before with a similar
> setup except there was no NAT involved and used the fxp drivers and
> it never had this problem. Thanks for your help in advance!

Additional delay while adding a hop is to be expected, no matter how
fast your network or router is. You only added about 1ms on average,
which is about right. The lost packet in the second traceroute might
be due to a full/half-duplex mismatch between one of the NICs and the
switch.


You're right that additional delay while adding a hop is to be expected, which is less 
than 0.1ms to the FreeBSD box but everything past the FreeBSD machine is adding 
atleast 5ms up to 300ms in the traceroutes when the normal is no more than 20ms for 
the same traceroute.  I've already checked the NICs and they are all configured at 
their full rated speeds and full duplex.  I even try using a Cardbus PCMCIA fxp0 Intel 
Pro/100S card on the FreeBSD box and it still had the same problem.  I am using a 
September 2003 -CURRENT so I don't know if it's a issue with the current networking 
code back then or not.

John


-
Do you Yahoo!?
Yahoo! Mail SpamGuard - Read only the mail you want.
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: FreeBSD box as router adding latency

[Chris Dillon]

On Wed, 25 Feb 2004, Aloha Guy wrote:

> Any ideas what is causing this? Is it the xl0 driver because I've
> used FreeBSD machines as ethernet routers before with a similar
> setup except there was no NAT involved and used the fxp drivers and
> it never had this problem. Thanks for your help in advance!

Additional delay while adding a hop is to be expected, no matter how
fast your network or router is.  You only added about 1ms on average,
which is about right.  The lost packet in the second traceroute might
be due to a full/half-duplex mismatch between one of the NICs and the
switch.


-- 
 Chris Dillon - cdillon(at)wolves.k12.mo.us
 FreeBSD: The fastest, most open, and most stable OS on the planet
 - Available for IA32, IA64, AMD64, PC98, Alpha, and UltraSPARC architectures
 - PowerPC, ARM, MIPS, and S/390 under development
 - http://www.freebsd.org

Q: Because it reverses the logical flow of conversation.
A: Why is putting a reply at the top of the message frowned upon?

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

FreeBSD box as router adding latency

[Aloha Guy]

Greetings everyone:

I'm using a FreeBSD based notebook (P4-M2.6Ghz, 2GB RAM) on the built in 3COM 920c 
(905c compatible) using the xl0 driver with the firewall enabled  and set to open and 
rc.conf basically has:

xl0 configured as 208.204.x.224 netmask 255.255.255.0 with the alias 192.168.0.1 
netmask 255.255.0.0.

natd is enabled with the natd interface as 208.204.x.224

tcp_extensions/RFC1323 is enabled

log_in_vain is set to 1

tcp_keepalive is set to YES

tcp_drop_synfin="NO"

icmp_drop_redirect="NO"

icmp_log_redirect="NO"

defaultrouter="208.201.x.1"

gateway_enable="YES"

forward_sourceroute="YES"

accept_sourceroute="YES"

I also have the following set:

# Don't respond to smurf-type icmp requests 

/sbin/sysctl -w net.inet.icmp.bmcastecho=0 

# Enhance Performance 

/sbin/sysctl -w kern.maxfiles=65536 

/sbin/sysctl -w kern.maxfilesperproc=32768

/sbin/sysctl -w kern.ipc.somaxconn=1024

/sbin/sysctl -w net.inet.ip.redirect=1 

/sbin/sysctl -w net.inet6.ip6.redirect=1 

/sbin/sysctl -w net.link.ether.inet.max_age=1200

The NIC is connected to a HP 2848 Managed 48 port Gigabit switch.

My rc.firewall basically has the following which is for traffic shaping as  well:

setup_loopback () {

${fwcmd} add 48 skipto 100 ip from 208.201.x.224/29 to any

${fwcmd} add 49 skipto 100 ip from any to 208.201.x.224/29

${fwcmd} add 50 divert natd all from any to any via ${natd_interface}

${fwcmd} add 100 pass all from any to any via lo0

${fwcmd} add 200 deny all from any to 127.0.0.0/8

${fwcmd} add 300 deny ip from 127.0.0.0/8 to any

${fwcmd} enable one_pass

${fwcmd} pipe 1 config bw 608Kbit/s

${fwcmd} queue 1 config pipe 1 weight 30

${fwcmd} queue 2 config pipe 1 weight 29

${fwcmd} queue 3 config pipe 1 weight 28

${fwcmd} queue 4 config pipe 1 weight 27

${fwcmd} add 63000 allow all from any to 10.0.0.0/8 out

${fwcmd} add 63001 allow all from any to 172.16.0.0/12 out

${fwcmd} add 63002 allow all from any to 192.168.0.0/16 out

${fwcmd} add 63003 allow all from any to 208.201.x.224/29 out

${fwcmd} add 63004 set 0 queue 1 tcp from any to any tcpflags ack iplen 0-80 out xmit 
xl0

${fwcmd} add 63005 set 0 queue 2 tcp from any to any 22,23 out xmit xl0 

${fwcmd} add 63006 set 0 queue 2 udp from any to any not 80,443 out xmit xl0

${fwcmd} add 63007 set 0 queue 3 all from any to any 80,443 out xmit xl0

${fwcmd} add 63008 set 0 queue 4 all from any to any out xmit xl0

${fwcmd} add 65000 pass all from any to any

and I guess FreeBSD adds the following rule by default:

${fwcmd} add 65535 deny ip from any to any

So anyways, here is the problem, if I traceroute from the FreeBSD machine:

traceroute to yahoo.com (66.218.71.198), 64 hops max, 40 byte packets

1 adsl-208-201-x-1.sonic.net (208.201.x.1) 7.274 ms 8.060 ms 7.384 ms

2 fast1-0-0.border.sr.sonic.net (208.201.224.194) 8.900 ms 8.921 ms 9.584 ms

3 fast0-0.gw.equinix-sj.sonic.net (64.142.0.14) 15.327 ms 14.889 ms  13.765 ms

4 exchange-cust1.sjo.equinix.net (206.223.116.16) 33.692 ms 34.501 ms 33.398 ms

5 ae0-p907.pat1.pao.yahoo.com (216.115.100.17) 19.431 ms 15.831 ms 14.858 ms

6 vlan26.bas1.scd.yahoo.com (216.115.101.34) 15.178 ms 20.284 ms 

vlan29.bas2.scd.yahoo.com (216.115.101.38) 15.301 ms

7 UNKNOWN-66-218-82-234.yahoo.com (66.218.82.234) 15.442 ms 
UNKNOWN-66-218-82-238.yahoo.com (66.218.82.238) 18.271 ms 
UNKNOWN-66-218-82-234.yahoo.com (66.218.82.234) 17.795 ms

8 alteon4.68.scd.yahoo.com (66.218.68.13) 17.168 ms 23.280 ms 19.143 ms

However, if I do the same traceroute from 208.201.x.225 (Intel PRO/1000CT CSA NIC 
connected to the same HP switch) or 208.201.x.226 (3Com 920c (905 compatible connected 
to the same HP switch), it seems to add some latency and timeout between hop 1 and two 
and beyond which is the FreeBSD box and other side of the DSL link as shown below:

Tracing route to yahoo.com [66.218.71.198] over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms adsl-208-201-x-224.sonic.net [208.201.x.224]

2 19 ms * 8 ms adsl-208-201-x-1.sonic.net [208.201.x.1]

3 9 ms 18 ms 10 ms fast1-0-0.border.sr.sonic.net [208.201.224.194]

4 17 ms 14 ms 15 ms fast0-0.gw.equinix-sj.sonic.net [64.142.0.14]

5 40 ms 34 ms 38 ms exchange-cust1.sjo.equinix.net [206.223.116.16]

6 15 ms 16 ms 23 ms ae0-p907.pat1.pao.yahoo.com [216.115.100.17]

7 17 ms 17 ms 18 ms vlan29.bas2.scd.yahoo.com [216.115.101.38]

8 16 ms 18 ms 16 ms UNKNOWN-66-218-82-234.yahoo.com [66.218.82.234]

9 18 ms 17 ms 23 ms w1.rc.vip.scd.yahoo.com [66.218.71.198]

Trace complete.

Any ideas what is causing this? Is it the xl0 driver because I've used FreeBSD 
machines as ethernet routers before with a similar setup except there was no NAT 
involved and used the fxp drivers and it never had this problem. Thanks for your help 
in advance!

John



-
Do you Yahoo!?
Yahoo! Mail SpamGuard - Read only the mail you want.
__

Re: latency

[Cordula's Web]

> How much is latency in FreeBSD?

???

> I would like to listen music playing on FreeBSD PC with hi-fi sound.
> First I need to be sure the system latency is low enough.

I'm listening mp3's (with mpg123) on a old 200 MHz Pentium box, which
is running -STABLE. At the same time setiathome, multiple servers,
XFree86 and a lot of other stuff is running at the same time. No
problems at all with latency AFAICS.

-- 
Cordula's Web. http://www.cordula.ws/

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: latency

[Benjamin Walkenhorst]

Hello,

On Thu, 22 Jan 2004 00:29:23 +0600
Stas <[EMAIL PROTECTED]> wrote:

> Hello,
> 
> How much is latency in FreeBSD?
> 
> I would like to listen music playing on FreeBSD PC with hi-fi sound.
> First I need to be sure the system latency is low enough.

It depends.
a) Where does the music come from? CD, MP3, WAV, MIDI? 
b) What kind of machine do you have? If you want to play Audio-CDs, you
can often plug headphones or speakers to the front-panel of a CD-drive.
Otherwise you need a sound "card", either on-board or as an add-on-card.
Also, if you want to list to MP3, your machine has to be fast enough
(P100 and faster, if I am not mistaken)

On my machine, I listen to MP3s a lot. It's an Athlon XP 2400+ with
256MB RAM and a Soundblaster 64 or 128 PCI, running FreeBSD
5.2-RELEASE: playing MP3s has almost no visible effect on system
performance. The sounds gets laggy, when the system is under *really
heavy* load. But I mean extremely heavy - compiling programs or
rebuilding the system has no notable effect. =) Also, I store my MP3s on
a second machine and share them with my desktop-machine via NFS, so if I
put the network under heavy load, playback gets laggy, too, sometimes.
But the network only consists of the two machines (the other one is
Athlon 700 / 160 MB RAM / NetBSD 1.6.2_RC4) connected via
100Mbit-Ethernet, so this happens very rarely, too. 

If your machine is fast enough, and if you have a soundcard that works
with FreeBSD, you should have no trouble listening to music.
If you have a soundcard installed, you just need to add the line
"device pcm"
to your kernel-config, recompile the kernel and there you go. Now all
you need is a program to play music. I use xmms (www.xmms.org, or you
can install it via ports, too). 

Back in summer, I had FreeBSD 5.0 installed on a Pentium III 450 with
256 MB RAM, listening to MP3s was fine, too. 
And both machines are *far* above the minimum requirements. A Pentium
133 with 64MB RAM should be sufficient, unless you want to run "big
boys" alongside (KDE, Mozilla, ...)

I am not quite sure what you mean by latency, but FreeBSD is a
multi-user, multi-tasking operating system, so even with many things
going on at the same time, music-playback works fine, if your machine is
fast enough.

> -- 
> Best regards,
>  Stas

Kind regards,

Benjamin


pgp0.pgp
Description: PGP signature

Re: latency

[Lance E. Lott]

$19.95.

Sorry, couldn't resist...

Lance



At 12:29 PM 1/21/2004, you wrote:

Hello,

How much is latency in FreeBSD?

I would like to listen music playing on FreeBSD PC with hi-fi sound.
First I need to be sure the system latency is low enough.
--
Best regards,
 Stas
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.564 / Virus Database: 356 - Release Date: 1/19/2004

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.564 / Virus Database: 356 - Release Date: 1/19/2004
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

latency

[Stas]

Hello,

How much is latency in FreeBSD?

I would like to listen music playing on FreeBSD PC with hi-fi sound.
First I need to be sure the system latency is low enough.
-- 
Best regards,
 Stas

___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Sound latency

[Matthew Faircliff]

Hello,

Thanks for the tip...

I have changes the sysctl hw.snd.pcm0.ac97rate to 48000 and all seems
to be working fine now. Funny tho, cause xmms reports all my music to
be encoded at 44100!

Thanks again,

Matthew Faircliff

Telephone: +27 21 531 0304
Fax: +27 21 531 5822
Mobile: +27 83 383 7711 

Public PGP Key: www.databias.co.za/pgp/matt

databias - connect. communicate. create.
www.databias.za.net / www.databias.co.za

On Mon, Jan 19, 2004 at 03:42:03PM +, [EMAIL PROTECTED] wrote:
From: [EMAIL PROTECTED]
To: Matthew Faircliff <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Subject: Re: Sound latency
Date: Mon, 19 Jan 2004 15:42:03 GMT


Hi Matthew!

Matthew Faircliff ?crit:
> Hello folks,
> 
> I'm running FreeBSD 5.1 on a Asus M2E laptop with AC97 sound. Runs
> like a charm but for one thing: when playing mp3s through xmms or
> mpg123 the songs seem to be about 5-10% slower than normal. 

I would suspect that the MP3 file has been recorded at 48kHz and you
replay it at 44.1kHz (which is the standard for AC97 codecs).

Where did you get the MP3s from and does your player support resampling?

Cheers
  Olivier
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: Sound latency

[ogautherot]

Hi Matthew!

Matthew Faircliff écrit:
> Hello folks,
> 
> I'm running FreeBSD 5.1 on a Asus M2E laptop with AC97 sound. Runs
> like a charm but for one thing: when playing mp3s through xmms or
> mpg123 the songs seem to be about 5-10% slower than normal. 

I would suspect that the MP3 file has been recorded at 48kHz and you
replay it at 44.1kHz (which is the standard for AC97 codecs).

Where did you get the MP3s from and does your player support resampling?

Cheers
  Olivier
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Sound latency

[Matthew Faircliff]

Hello folks,

I'm running FreeBSD 5.1 on a Asus M2E laptop with AC97 sound. Runs
like a charm but for one thing: when playing mp3s through xmms or
mpg123 the songs seem to be about 5-10% slower than normal. 

The latency is only noticable if you listen closely, but it is
definately not a figment of my imagination. 

Any ideas???

Matthew Faircliff
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Re: How to simulate high latency links?

[Dan Nelson]

In the last episode (Nov 03), Max Clark said:
> So if I want to do testing between machine A and B I can route all of
> the traffic trough a machine C with dummynet and simulate the network
> environment that I need?

Sure.  You can also add the rules to either A or B and do the same
thing, or even set them up on A's lo0 interface and do all the testing
on A :)
 
> Basically I want to test/experiment with the send/receive settings
> within the servers.
> 
> What about a hardware appliance? Could this be set up using a QOS
> policy or something similar with a switch?

You could set up something similar with the rate-limit command on Cisco
routers.

-- 
Dan Nelson
[EMAIL PROTECTED]

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-questions" in the body of the message

RE: How to simulate high latency links?

[Max Clark]

So if I want to do testing between machine A and B I can route all of
the traffic trough a machine C with dummynet and simulate the network
environment that I need?

Basically I want to test/experiment with the send/receive settings
within the servers.

What about a hardware appliance? Could this be set up using a QOS policy
or something similar with a switch?

Thanks for the advice.
Max

-Original Message-
From: [EMAIL PROTECTED]
[mailto:owner-freebsd-questions@;FreeBSD.ORG] On Behalf Of Dan Nelson
Sent: Sunday, November 03, 2002 6:05 PM
To: Max Clark
Cc: [EMAIL PROTECTED]
Subject: Re: How to simulate high latency links?

In the last episode (Nov 03), Max Clark said:
> Hi,
>
> I am looking for a way to do some performance testing/tuning in a lab
> environment. I have a high latency low speed link (T1/200MS) that I
need
> to replicate.
>
> Are there any ways to do this with FreeBSD?

dummynet is what you want.  By redirecting traffic through a dummynet
pipe with ipfw, you can simulate latency and packet loss.  See the ipfw
and dummynet manpages.

--
Dan Nelson
[EMAIL PROTECTED]

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-questions" in the body of the message



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-questions" in the body of the message

Re: How to simulate high latency links?

[Dan Nelson]

In the last episode (Nov 03), Max Clark said:
> Hi,
> 
> I am looking for a way to do some performance testing/tuning in a lab
> environment. I have a high latency low speed link (T1/200MS) that I need
> to replicate.
> 
> Are there any ways to do this with FreeBSD?

dummynet is what you want.  By redirecting traffic through a dummynet
pipe with ipfw, you can simulate latency and packet loss.  See the ipfw
and dummynet manpages.

-- 
Dan Nelson
[EMAIL PROTECTED]

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-questions" in the body of the message

How to simulate high latency links?

[Max Clark]

Hi,

I am looking for a way to do some performance testing/tuning in a lab
environment. I have a high latency low speed link (T1/200MS) that I need
to replicate.

Are there any ways to do this with FreeBSD?

Thanks in advance,
Max

--
Mailution Email Solutions
http://www.mailution.net



To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-questions" in the body of the message