From kevin.a.mcgregor at gmail.com  Thu Apr  1 13:06:35 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Thu, 1 Apr 2010 13:06:35 -0500
Subject: [RndTbl] Potentially long cron jobs
Message-ID: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>

I want to run something via cron which will take anywhere from less than a
minute to a day or so (possibly). Most of the time it will be well under an
hour, and I'd like it to run hourly, except if it's still running, in which
case nothing more need be done.

What the best way to do that?

Kevin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100401/08a5d90b/attachment.html 

From swalberg at gmail.com  Thu Apr  1 13:18:26 2010
From: swalberg at gmail.com (Sean Walberg)
Date: Thu, 1 Apr 2010 13:18:26 -0500
Subject: [RndTbl] Potentially long cron jobs
In-Reply-To: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
References: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
Message-ID: <r2lcc43a8f81004011118yd7f70edaq67887996637838a9@mail.gmail.com>

#!/bin/bash
TURD=/var/run/mycron.lock
if [ -f $TURD ]; then
 exit
fi
touch $TURD
# do stuff
rm $TURD

(yes, there's a potential race condition. Probably doesn't matter though)

Sean

On Thu, Apr 1, 2010 at 1:06 PM, Kevin McGregor
<kevin.a.mcgregor at gmail.com>wrote:

> I want to run something via cron which will take anywhere from less than a
> minute to a day or so (possibly). Most of the time it will be well under an
> hour, and I'd like it to run hourly, except if it's still running, in which
> case nothing more need be done.
>
> What the best way to do that?
>
> Kevin
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100401/b40948da/attachment.html 

From gedetil at cs.umanitoba.ca  Thu Apr  1 14:02:08 2010
From: gedetil at cs.umanitoba.ca (Gilbert E. Detillieux)
Date: Thu, 01 Apr 2010 14:02:08 -0500
Subject: [RndTbl] Potentially long cron jobs
In-Reply-To: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
References: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
Message-ID: <4BB4EDB0.6030302@cs.umanitoba.ca>

On 2010-04-01 13:06, Kevin McGregor wrote:
> I want to run something via cron which will take anywhere from less than 
> a minute to a day or so (possibly). Most of the time it will be well 
> under an hour, and I'd like it to run hourly, except if it's still 
> running, in which case nothing more need be done.
> 
> What the best way to do that?

This came up on the CentOS-mirror list back in February.  Proposed 
solutions included...

http://djlab.com/2009/10/lockfiles-in-bash/ [similar to Sean's solution]
http://code.google.com/p/withlock/ [requires Python 2.4]
flock(1) [may not be available other than on Linux?]

Hope this helps.

-- 
Gilbert E. Detillieux		E-mail:	<gedetil at cs.umanitoba.ca>
Dept. of Computer Science	Web:	http://www.cs.umanitoba.ca/~gedetil/
University of Manitoba		Phone:	(204)474-8161
Winnipeg MB CANADA  R3T 2N2	Fax:	(204)474-7609

From athompso at muug.mb.ca  Thu Apr  1 16:41:23 2010
From: athompso at muug.mb.ca (athompso at muug.mb.ca)
Date: Thu, 1 Apr 2010 16:41:23 -0500
Subject: [RndTbl] WARNING - RndTbl reply-to setting changed
Message-ID: <00a701cad1e4$12f81f80$38e85e80$@mb.ca>

After a number of complaints (including some from me!) we've changed the 
default setting for Replies on the roundtable at muug.mb.ca (RndTbl) mailing 
list.

When you replied to a message on the list previously, the default was to 
send replies only to the original sender, bypassing the mailing list.
To reply to the list members at large, you probably would have had to use 
your email client's Reply-to-All function.

NOW, when you reply to a message on the list, the default will be to send 
the reply TO THE ENTIRE LIST.

This means the *potential* exists to send embarrassing messages to a much 
larger audience than intended - so the usual rule applies doubly: if you 
don't want everyone to see what you're writing, don't put it in an email. 
And if you do anyway, double-check the To: address before you hit send.

We're all more technically-oriented than the "average user", so I'm hoping 
there won't be very many (if any) ohnosecond[1] moments.

Thank you,

-Adam Thompson
 <athompso at muug.mb.ca>


[1] http://en.wiktionary.org/wiki/ohnosecond




From peter at pogma.com  Thu Apr  1 16:57:59 2010
From: peter at pogma.com (Peter O'Gorman)
Date: Thu, 01 Apr 2010 16:57:59 -0500
Subject: [RndTbl] Potentially long cron jobs
In-Reply-To: <r2lcc43a8f81004011118yd7f70edaq67887996637838a9@mail.gmail.com>
References: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
	<r2lcc43a8f81004011118yd7f70edaq67887996637838a9@mail.gmail.com>
Message-ID: <4BB516E7.8070603@pogma.com>

On 04/01/2010 01:18 PM, Sean Walberg wrote:
> #!/bin/bash
> TURD=/var/run/mycron.lock
> if [ -f $TURD ]; then
>  exit
> fi
> touch $TURD
> # do stuff
> rm $TURD

I find it helpful to have:
...
TURD=/var/run/mycron.lock
trap "rm -f $TURD" 0
...

So that the file will be removed when the shell exits.

I have also used the procmail 'lockfile' command for situations where I
wanted a shorter term lock.

Peter

From trevor at tecnopolis.ca  Thu Apr  1 17:46:11 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Thu, 1 Apr 2010 17:46:11 -0500
Subject: [RndTbl] Potentially long cron jobs
In-Reply-To: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
References: <h2s6756caf11004011106j6f67bb83nb51795637b1ad8f2@mail.gmail.com>
Message-ID: <20100401174611.490b1387@pog.tecnopolis.ca>

On 2010-04-01 Kevin McGregor wrote:
> I want to run something via cron which will take anywhere from less
> than a minute to a day or so (possibly). Most of the time it will be
> well under an hour, and I'd like it to run hourly, except if it's
> still running, in which case nothing more need be done.

To not worry about races, and still keep things simple:

lock=~/lockfile-whatever
lockfile -r 0 -l 86400 $lock >/dev/null 2>&1 || exit
do_work
rm -f $lock

see man lockfile

may not be terribly portable beyond linux(?)

Play with the -l option or omit it.  For your case you'd want to set it
to the absolute max secs it could possibly take your program to run.

Perl also has a ways to do locking.

Note: "Lockfile is NFS-resistant": I have no idea what they mean by
that, will it work or not?  Pretty nifty if they have it working on
NFSvX where X<4.

From kevin.a.mcgregor at gmail.com  Mon Apr  5 17:36:28 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Mon, 5 Apr 2010 17:36:28 -0500
Subject: [RndTbl] PCIe interrupt assignments
Message-ID: <s2o6756caf11004051536v1a323608q76b4fe1a2912f221@mail.gmail.com>

I have an 8-port SATA disk controller in a PCIe slot (thanks, Trevor!), and
it seems to be sharing an interrupt. I'm running Ubuntu Server 9.10 64-bit.
Is there some way to arrange for it to have its own interrupt, unshared?

Kevin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/95dd3860/attachment.html 

From athompso at athompso.net  Mon Apr  5 17:39:50 2010
From: athompso at athompso.net (Adam Thompson)
Date: Mon, 5 Apr 2010 17:39:50 -0500
Subject: [RndTbl] PCIe interrupt assignments
In-Reply-To: <s2o6756caf11004051536v1a323608q76b4fe1a2912f221@mail.gmail.com>
References: <s2o6756caf11004051536v1a323608q76b4fe1a2912f221@mail.gmail.com>
Message-ID: <008f01cad510$e72bc980$b5835c80$@net>

Short answer: no.

 

Sharing interrupts in PCIe is a normal state of affairs.  You?d have to have a BIOS that allowed you to override that.

 

On the upside, the historical case of sharing interrupts being bad, both because of drivers that didn?t allow for that and because of the performance hit, is just that ? historical.  There is a theoretically measurable performance hit even under PCIe, but it?s negligibly small.  (Or at least it?s supposed to be.  Bad drivers can screw anything up.)

 

Why do you want it to have its own interrupt?

 

-Adam Thompson 

 <athompso at athompso.net>

 (204) 291-7950

 

From: roundtable-bounces at muug.mb.ca [mailto:roundtable-bounces at muug.mb.ca] On Behalf Of Kevin McGregor
Sent: Monday, April 05, 2010 17:36
To: MUUG Roundtable
Subject: [RndTbl] PCIe interrupt assignments

 

I have an 8-port SATA disk controller in a PCIe slot (thanks, Trevor!), and it seems to be sharing an interrupt. I'm running Ubuntu Server 9.10 64-bit. Is there some way to arrange for it to have its own interrupt, unshared?

 

Kevin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/4a24779a/attachment.html 

From kevin.a.mcgregor at gmail.com  Mon Apr  5 18:26:07 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Mon, 5 Apr 2010 18:26:07 -0500
Subject: [RndTbl] PCIe interrupt assignments
In-Reply-To: <008f01cad510$e72bc980$b5835c80$@net>
References: <s2o6756caf11004051536v1a323608q76b4fe1a2912f221@mail.gmail.com>
	<008f01cad510$e72bc980$b5835c80$@net>
Message-ID: <x2l6756caf11004051626j7d15ab75ze8514c52b08167bd@mail.gmail.com>

Oh, it was just suggested by a (presumably very outdated) RAID article I
found. I didn't think it would be a big deal for the reasons you mentioned.
Plus it's sharing the interrupt with two USB ports which don't even have the
motherboard adapter connected to them.

Thanks for the confirmation.

On Mon, Apr 5, 2010 at 5:39 PM, Adam Thompson <athompso at athompso.net> wrote:

>  Short answer: no.
>
>
>
> Sharing interrupts in PCIe is a normal state of affairs.  You?d have to
> have a BIOS that allowed you to override that.
>
>
>
> On the upside, the historical case of sharing interrupts being bad, both
> because of drivers that didn?t allow for that and because of the performance
> hit, is just that ? historical.  There is a theoretically measurable
> performance hit even under PCIe, but it?s negligibly small.  (Or at least
> it?s supposed to be.  Bad drivers can screw anything up.)
>
>
>
> Why do you want it to have its own interrupt?
>
>
>
> -Adam Thompson
>
>  <athompso at athompso.net>
>
>  (204) 291-7950
>
>
>
> *From:* roundtable-bounces at muug.mb.ca [mailto:
> roundtable-bounces at muug.mb.ca] *On Behalf Of *Kevin McGregor
> *Sent:* Monday, April 05, 2010 17:36
> *To:* MUUG Roundtable
> *Subject:* [RndTbl] PCIe interrupt assignments
>
>
>
> I have an 8-port SATA disk controller in a PCIe slot (thanks, Trevor!), and
> it seems to be sharing an interrupt. I'm running Ubuntu Server 9.10 64-bit.
> Is there some way to arrange for it to have its own interrupt, unshared?
>
>
>
> Kevin
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/7a675658/attachment.html 

From kevin.a.mcgregor at gmail.com  Mon Apr  5 19:27:03 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Mon, 5 Apr 2010 19:27:03 -0500
Subject: [RndTbl] Network performance tuning
Message-ID: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>

I've been trying to narrow down where my performance bottleneck(s) is (are).
I just ran "iperf" from a Windows PC (XP SP3) to my Ubuntu 9.10 server. With
iperf in "server" mode on the PC, and the client running on the Ubuntu
machine, I get
 0.0-60.0 sec  4.23 GBytes   606 Mbits/sec

Reversed (iperf "server" running on Ubuntu server, client running on the
PC), I get
 0.0-60.0 sec  1.13 GBytes   162 Mbits/sec

Wha...?? Both client and server runs on the PC report TCP window size to be
8 KB, but the Linux client reports 22.4 KB and the Linux server reports 85.3
KB. Increasing that to 256K on both ends has little effect. Does anyone have
any suggestions?

The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the
Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI
Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?

Kevin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/e4d71980/attachment.html 

From swalberg at gmail.com  Mon Apr  5 20:12:26 2010
From: swalberg at gmail.com (Sean Walberg)
Date: Mon, 5 Apr 2010 20:12:26 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
Message-ID: <i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>

On a LAN, window sizes aren't going to make that much of a difference.

First look for the usual suspects - errors on the swich port or NIC.

If you can grab a copy of the traffic, such as with tcpdump -w test.pcap tcp
port 5001, you can pull it into wireshark and look for TCP zero window
conditions, retransmits, and duplicate ACKs. The TCP performance graph will
also show whether or not the transmission is stalling.

Is the performance similar using the UDP test? Does it show loss?

Sean

On Mon, Apr 5, 2010 at 7:27 PM, Kevin McGregor
<kevin.a.mcgregor at gmail.com>wrote:

> I've been trying to narrow down where my performance bottleneck(s) is
> (are). I just ran "iperf" from a Windows PC (XP SP3) to my Ubuntu 9.10
> server. With iperf in "server" mode on the PC, and the client running on the
> Ubuntu machine, I get
>  0.0-60.0 sec  4.23 GBytes   606 Mbits/sec
>
> Reversed (iperf "server" running on Ubuntu server, client running on the
> PC), I get
>  0.0-60.0 sec  1.13 GBytes   162 Mbits/sec
>
> Wha...?? Both client and server runs on the PC report TCP window size to be
> 8 KB, but the Linux client reports 22.4 KB and the Linux server reports 85.3
> KB. Increasing that to 256K on both ends has little effect. Does anyone have
> any suggestions?
>
> The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the
> Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI
> Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?
>
> Kevin
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/68e9612b/attachment-0001.html 

From athompso at athompso.net  Mon Apr  5 21:16:50 2010
From: athompso at athompso.net (Adam Thompson)
Date: Mon, 5 Apr 2010 21:16:50 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
Message-ID: <00c601cad52f$37c5a780$a750f680$@net>

I don?t have much constructive to offer here, but I?ve heard several people encounter much the same thing ? asymmetric network performance between a Windows PC and a Linux file server.  Your observations, however, by using iperf, exclude SAMBA ? which means it?s not a SAMBA-specific problem, it?s a Linux-TCP-stack-to-Windows-TCP-stack problem.

 

Can you tell us what kernel version, and what Windows version, you?re running?

 

Also, do you happen to have a third system from which you could provide comparative results?

 

-Adam Thompson 

 <athompso at athompso.net>

 (204) 291-7950

 

From: roundtable-bounces at muug.mb.ca [mailto:roundtable-bounces at muug.mb.ca] On Behalf Of Kevin McGregor
Sent: Monday, April 05, 2010 19:27
To: MUUG Roundtable
Subject: [RndTbl] Network performance tuning

 

I've been trying to narrow down where my performance bottleneck(s) is (are). I just ran "iperf" from a Windows PC (XP SP3) to my Ubuntu 9.10 server. With iperf in "server" mode on the PC, and the client running on the Ubuntu machine, I get

 0.0-60.0 sec  4.23 GBytes   606 Mbits/sec

 

Reversed (iperf "server" running on Ubuntu server, client running on the PC), I get

 0.0-60.0 sec  1.13 GBytes   162 Mbits/sec

 

Wha...?? Both client and server runs on the PC report TCP window size to be 8 KB, but the Linux client reports 22.4 KB and the Linux server reports 85.3 KB. Increasing that to 256K on both ends has little effect. Does anyone have any suggestions?

 

The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?

 

Kevin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/f7fb2aaf/attachment.html 

From kevin.a.mcgregor at gmail.com  Mon Apr  5 21:43:27 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Mon, 5 Apr 2010 21:43:27 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
Message-ID: <o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>

When I choose UDP, I get
 0.0-15.0 sec  1.88 MBytes  1.05 Mbits/sec  6.154 ms    0/ 1339 (0%)

No packet loss. 0.1% utilization!

On Mon, Apr 5, 2010 at 8:12 PM, Sean Walberg <swalberg at gmail.com> wrote:

> On a LAN, window sizes aren't going to make that much of a difference.
>
> First look for the usual suspects - errors on the swich port or NIC.
>
> If you can grab a copy of the traffic, such as with tcpdump -w test.pcap
> tcp port 5001, you can pull it into wireshark and look for TCP zero window
> conditions, retransmits, and duplicate ACKs. The TCP performance graph will
> also show whether or not the transmission is stalling.
>
> Is the performance similar using the UDP test? Does it show loss?
>
> Sean
>
> On Mon, Apr 5, 2010 at 7:27 PM, Kevin McGregor <kevin.a.mcgregor at gmail.com
> > wrote:
>
>> I've been trying to narrow down where my performance bottleneck(s) is
>> (are). I just ran "iperf" from a Windows PC (XP SP3) to my Ubuntu 9.10
>> server. With iperf in "server" mode on the PC, and the client running on the
>> Ubuntu machine, I get
>>  0.0-60.0 sec  4.23 GBytes   606 Mbits/sec
>>
>> Reversed (iperf "server" running on Ubuntu server, client running on the
>> PC), I get
>>  0.0-60.0 sec  1.13 GBytes   162 Mbits/sec
>>
>> Wha...?? Both client and server runs on the PC report TCP window size to
>> be 8 KB, but the Linux client reports 22.4 KB and the Linux server reports
>> 85.3 KB. Increasing that to 256K on both ends has little effect. Does anyone
>> have any suggestions?
>>
>> The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the
>> Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI
>> Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?
>>
>> Kevin
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>
>
> --
> Sean Walberg <sean at ertw.com>    http://ertw.com/
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/b6c76f8b/attachment.html 

From swalberg at gmail.com  Mon Apr  5 21:52:55 2010
From: swalberg at gmail.com (Sean Walberg)
Date: Mon, 5 Apr 2010 21:52:55 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
Message-ID: <r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>

IIRC UDP needs you to pass the desired bandwidth, otherwise it defaults to a
megabit.

Sean

On Mon, Apr 5, 2010 at 9:43 PM, Kevin McGregor
<kevin.a.mcgregor at gmail.com>wrote:

> When I choose UDP, I get
>  0.0-15.0 sec  1.88 MBytes  1.05 Mbits/sec  6.154 ms    0/ 1339 (0%)
>
> No packet loss. 0.1% utilization!
>
>
> On Mon, Apr 5, 2010 at 8:12 PM, Sean Walberg <swalberg at gmail.com> wrote:
>
>> On a LAN, window sizes aren't going to make that much of a difference.
>>
>> First look for the usual suspects - errors on the swich port or NIC.
>>
>> If you can grab a copy of the traffic, such as with tcpdump -w test.pcap
>> tcp port 5001, you can pull it into wireshark and look for TCP zero window
>> conditions, retransmits, and duplicate ACKs. The TCP performance graph will
>> also show whether or not the transmission is stalling.
>>
>> Is the performance similar using the UDP test? Does it show loss?
>>
>> Sean
>>
>> On Mon, Apr 5, 2010 at 7:27 PM, Kevin McGregor <
>> kevin.a.mcgregor at gmail.com> wrote:
>>
>>> I've been trying to narrow down where my performance bottleneck(s) is
>>> (are). I just ran "iperf" from a Windows PC (XP SP3) to my Ubuntu 9.10
>>> server. With iperf in "server" mode on the PC, and the client running on the
>>> Ubuntu machine, I get
>>>  0.0-60.0 sec  4.23 GBytes   606 Mbits/sec
>>>
>>> Reversed (iperf "server" running on Ubuntu server, client running on the
>>> PC), I get
>>>  0.0-60.0 sec  1.13 GBytes   162 Mbits/sec
>>>
>>> Wha...?? Both client and server runs on the PC report TCP window size to
>>> be 8 KB, but the Linux client reports 22.4 KB and the Linux server reports
>>> 85.3 KB. Increasing that to 256K on both ends has little effect. Does anyone
>>> have any suggestions?
>>>
>>> The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the
>>> Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI
>>> Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?
>>>
>>> Kevin
>>>
>>> _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>
>>>
>>
>>
>> --
>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100405/f2e4158a/attachment.html 

From gedetil at cs.umanitoba.ca  Tue Apr  6 14:44:08 2010
From: gedetil at cs.umanitoba.ca (Gilbert E. Detillieux)
Date: Tue, 6 Apr 2010 14:44:08 -0500 (CDT)
Subject: [RndTbl] MUUG Meeting, April 13, 7:30pm -- A Trio of RTFM!
Message-ID: <201004061944.o36Ji8C11365@iron.cs.umanitoba.ca>

The Manitoba UNIX User Group (MUUG) will be holding its next monthly
meeting on Tuesday, April 13.  The meeting topic for this month is
as follows:

   A Trio of RTFM!

   Weve  had  a  lot  of  positive  feedback to our previous RTFM
   presentations, so this month, were offering you a trio of them!
   To start, Gilbert Detillieux will be introducing the grep family of
   commands (grep, egrep and fgrep) including a brief look at regular
   expression syntax and usage. Next up, Michael Doob will present the
   sed command, showing you how to take regular expressions a little
   further. Finally, Mark Jenkins will present the smartctl command,
   for monitoring hard disk status and health.

The group holds its general meetings at 7:30pm on the second Tuesday of
every month from September to June.  (There are no meetings in July and
August.) Meetings are open to the general public; you don't have to be a
MUUG member to attend.

   **********************************************************************
   Please note our meeting location: The IBM offices, at 400 Ellice Ave.
   (between Edmonton and Kennedy). When you arrive, you will have to
   sign in at the reception desk, and then wait for someone to take you
   (in groups) to the meeting room. Please try to arrive by about 7:15pm, 
   so the meeting can start promptly at 7:30pm. Don't be late, or you may
   not get in.  (But don't come too early either, since security may not
   be there to let you in before 7:15 or so.) Non-members may be required
   to show photo ID at the security desk.

   Limited parking is available for free on the street, either on Ellice
   Ave. or on some of the intersecting streets. Indoor parking is also
   available nearby, at Portage Place, for $5.00 for the evening. Bicycle
   parking is available in a bike rack under video surveillance located
   behind the building on Webb Place.
   **********************************************************************

For more information about MUUG, and its monthly meetings, check out their
Web server:

	http://www.muug.mb.ca/

Help us promote this month's meeting, by putting this poster up on your
workplace bulletin board or other suitable public message board:

	http://www.muug.mb.ca/meetings/MUUGmeeting.pdf

-- 
Gilbert E. Detillieux		E-mail: <gedetil at muug.mb.ca>
Manitoba UNIX User Group	Web:	http://www.muug.mb.ca/
PO Box 130 St-Boniface		Phone:  (204)474-8161
Winnipeg MB CANADA  R2H 3B4	Fax:    (204)474-7609

From kevin.a.mcgregor at gmail.com  Tue Apr  6 18:36:59 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Tue, 6 Apr 2010 18:36:59 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
Message-ID: <r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>

More results: Linux iperf client, Windows iperf server:
# iperf -i 5 -t 15 -c 192.168.27.23 -u -b 700M
------------------------------------------------------------
Client connecting to 192.168.27.23, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size:   126 KByte (default)
------------------------------------------------------------
[  3] local 192.168.27.10 port 56210 connected with 192.168.27.23 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0- 5.0 sec    428 MBytes    718 Mbits/sec
[  3]  5.0-10.0 sec    408 MBytes    684 Mbits/sec
[  3]  0.0-15.0 sec  1.22 GBytes    700 Mbits/sec
[  3] Sent 892738 datagrams
[  3] Server Report:
[  3]  0.0-15.0 sec  1.16 GBytes    663 Mbits/sec  0.015 ms 47569/892392
(5.3%)
[  3]  0.0-15.0 sec  1 datagrams received out-of-order

Reversed:
C:\Temp>iperf -u -t 15 -c 192.168.27.10 -b 750M
------------------------------------------------------------
Client connecting to 192.168.27.10, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size: 8.00 KByte (default)
------------------------------------------------------------
[1912] local 192.168.27.23 port 2100 connected with 192.168.27.10 port 5001
[ ID] Interval       Transfer     Bandwidth
[1912]  0.0-15.0 sec   174 MBytes  97.5 Mbits/sec
[1912] Server Report:
[1912]  0.0-15.0 sec   174 MBytes  97.5 Mbits/sec  1.055 ms    0/124429 (0%)
[1912] Sent 124429 datagrams

And then, running the iperf client on the same hardware as was running
Windows, but running Ubuntu 9.10 (dual boot):
$ iperf -t 15 -c 192.168.27.10 -u -b 750M
------------------------------------------------------------
Client connecting to 192.168.27.10, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size:   112 KByte (default)
------------------------------------------------------------
[  3] local 192.168.27.23 port 51363 connected with 192.168.27.10 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-15.0 sec    316 MBytes    177 Mbits/sec
[  3] Sent 225456 datagrams
[  3] Server Report:
[  3]  0.0-15.0 sec    316 MBytes    177 Mbits/sec  0.207 ms    0/225455
(0%)
[  3]  0.0-15.0 sec  1 datagrams received out-of-order

And with TCP:
$ iperf -t 15 -c 192.168.27.10
------------------------------------------------------------
Client connecting to 192.168.27.10, TCP port 5001
TCP window size: 16.0 KByte (default)
------------------------------------------------------------
[  3] local 192.168.27.23 port 53251 connected with 192.168.27.10 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-15.0 sec    368 MBytes    206 Mbits/sec


The source hardware seems to be having a problem sending. Receiving, less
so:
$ iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 192.168.27.23 port 5001 connected with 192.168.27.10 port 37472
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-15.0 sec    949 MBytes    531 Mbits/sec

$ iperf -t 15 -c 192.168.27.23 -u -b 750M
------------------------------------------------------------
Client connecting to 192.168.27.23, UDP port 5001
Sending 1470 byte datagrams
UDP buffer size:   126 KByte (default)
------------------------------------------------------------
[  3] local 192.168.27.10 port 37815 connected with 192.168.27.23 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-15.0 sec  1.37 GBytes    783 Mbits/sec
[  3] Sent 999330 datagrams
[  3] Server Report:
[  3]  0.0-15.2 sec  1.09 GBytes    613 Mbits/sec  15.609 ms 204247/999327
(20%)
[  3]  0.0-15.2 sec  1 datagrams received out-of-order

...Although push the bandwidth (UDP) too high, and a lot of packets get
lost.

Any further thoughts?

Kevin

On Mon, Apr 5, 2010 at 9:52 PM, Sean Walberg <swalberg at gmail.com> wrote:

> IIRC UDP needs you to pass the desired bandwidth, otherwise it defaults to
> a megabit.
>
> Sean
>
>
> On Mon, Apr 5, 2010 at 9:43 PM, Kevin McGregor <kevin.a.mcgregor at gmail.com
> > wrote:
>
>> When I choose UDP, I get
>>  0.0-15.0 sec  1.88 MBytes  1.05 Mbits/sec  6.154 ms    0/ 1339 (0%)
>>
>> No packet loss. 0.1% utilization!
>>
>>
>> On Mon, Apr 5, 2010 at 8:12 PM, Sean Walberg <swalberg at gmail.com> wrote:
>>
>>> On a LAN, window sizes aren't going to make that much of a difference.
>>>
>>> First look for the usual suspects - errors on the swich port or NIC.
>>>
>>> If you can grab a copy of the traffic, such as with tcpdump -w test.pcap
>>> tcp port 5001, you can pull it into wireshark and look for TCP zero window
>>> conditions, retransmits, and duplicate ACKs. The TCP performance graph will
>>> also show whether or not the transmission is stalling.
>>>
>>> Is the performance similar using the UDP test? Does it show loss?
>>>
>>> Sean
>>>
>>> On Mon, Apr 5, 2010 at 7:27 PM, Kevin McGregor <
>>> kevin.a.mcgregor at gmail.com> wrote:
>>>
>>>> I've been trying to narrow down where my performance bottleneck(s) is
>>>> (are). I just ran "iperf" from a Windows PC (XP SP3) to my Ubuntu 9.10
>>>> server. With iperf in "server" mode on the PC, and the client running on the
>>>> Ubuntu machine, I get
>>>>  0.0-60.0 sec  4.23 GBytes   606 Mbits/sec
>>>>
>>>> Reversed (iperf "server" running on Ubuntu server, client running on the
>>>> PC), I get
>>>>  0.0-60.0 sec  1.13 GBytes   162 Mbits/sec
>>>>
>>>> Wha...?? Both client and server runs on the PC report TCP window size to
>>>> be 8 KB, but the Linux client reports 22.4 KB and the Linux server reports
>>>> 85.3 KB. Increasing that to 256K on both ends has little effect. Does anyone
>>>> have any suggestions?
>>>>
>>>> The Windows NIC is a Realtek RTL8169/8110 Family Gigabit Ethernet; the
>>>> Ubuntu server is using the Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI
>>>> Express Gigabit Ethernet controller (rev 01). Time to replace a NIC or two?
>>>>
>>>> Kevin
>>>>
>>>> _______________________________________________
>>>> Roundtable mailing list
>>>> Roundtable at muug.mb.ca
>>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>>
>>>>
>>>
>>>
>>> --
>>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>>>
>>> _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>
>>>
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>
>
> --
> Sean Walberg <sean at ertw.com>    http://ertw.com/
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100406/507cb0c1/attachment.html 

From athompso at athompso.net  Wed Apr  7 00:00:53 2010
From: athompso at athompso.net (Adam Thompson)
Date: Wed, 7 Apr 2010 00:00:53 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
	<r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
Message-ID: <010c01cad60f$4cd85560$e6890020$@net>

Actually, I was mainly hoping to verify that it was, indeed, a hardware problem.  One person (Trevor) reporting similar results has fairly decent-quality GigE NICs on both sides ? or at least what I *assumed* to be fairly decent-quality NICs!

 

In your case, I?d say yes, it?s time to test different NICs.  Obviously one of your NICs is OK ? although I wouldn?t want to put much money on which one until I tested thoroughly.

 

I?d also not be willing to put money on whether it?s the NIC hardware or the software (i.e. device driver) ? even though you tested under two OSes, the Linux drivers and the Windows drivers share a lot of code for both Intel and Realtek NICs nowadays.

 

-Adam Thompson 

 <athompso at athompso.net>

 (204) 291-7950

 

From: roundtable-bounces at muug.mb.ca [mailto:roundtable-bounces at muug.mb.ca] On Behalf Of Kevin McGregor
Sent: Tuesday, April 06, 2010 18:37
To: Continuation of Round Table discussion
Subject: Re: [RndTbl] Network performance tuning

[...]
The source hardware seems to be having a problem sending. Receiving, less so:

[...]

...Although push the bandwidth (UDP) too high, and a lot of packets get lost.
[...]
Any further thoughts?



 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100407/c6ce1850/attachment-0001.html 

From swalberg at gmail.com  Wed Apr  7 06:28:34 2010
From: swalberg at gmail.com (Sean Walberg)
Date: Wed, 7 Apr 2010 06:28:34 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <010c01cad60f$4cd85560$e6890020$@net>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
	<r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
	<010c01cad60f$4cd85560$e6890020$@net>
Message-ID: <g2rcc43a8f81004070428m5ad2e92bpea4539404ee63b9@mail.gmail.com>

Before you go switching hardware, try swapping cables and switch ports.

Sean

On Wed, Apr 7, 2010 at 12:00 AM, Adam Thompson <athompso at athompso.net>wrote:

>  Actually, I was mainly hoping to verify that it was, indeed, a hardware
> problem.  One person (Trevor) reporting similar results has fairly
> decent-quality GigE NICs on both sides ? or at least what I **assumed** to
> be fairly decent-quality NICs!
>
>
>
> In your case, I?d say yes, it?s time to test different NICs.  Obviously one
> of your NICs is OK ? although I wouldn?t want to put much money on which one
> until I tested thoroughly.
>
>
>
> I?d also not be willing to put money on whether it?s the NIC hardware or
> the software (i.e. device driver) ? even though you tested under two OSes,
> the Linux drivers and the Windows drivers share a lot of code for both Intel
> and Realtek NICs nowadays.
>
>
>
> -Adam Thompson
>
>  <athompso at athompso.net>
>
>  (204) 291-7950
>
>
>
> *From:* roundtable-bounces at muug.mb.ca [mailto:
> roundtable-bounces at muug.mb.ca] *On Behalf Of *Kevin McGregor
> *Sent:* Tuesday, April 06, 2010 18:37
> *To:* Continuation of Round Table discussion
> *Subject:* Re: [RndTbl] Network performance tuning
>
> [...]
>
> The source hardware seems to be having a problem sending. Receiving, less
> so:
>
> [...]
>
> ...Although push the bandwidth (UDP) too high, and a lot of packets get
> lost.
> [...]
> Any further thoughts?
>
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100407/796c8484/attachment.html 

From trevor at tecnopolis.ca  Wed Apr  7 15:13:35 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Wed, 7 Apr 2010 15:13:35 -0500
Subject: [RndTbl] PCIe interrupt assignments
In-Reply-To: <x2l6756caf11004051626j7d15ab75ze8514c52b08167bd@mail.gmail.com>
References: <s2o6756caf11004051536v1a323608q76b4fe1a2912f221@mail.gmail.com>
	<008f01cad510$e72bc980$b5835c80$@net>
	<x2l6756caf11004051626j7d15ab75ze8514c52b08167bd@mail.gmail.com>
Message-ID: <20100407151335.27deb006@pog.tecnopolis.ca>

On 2010-04-05 Kevin McGregor wrote:
> Oh, it was just suggested by a (presumably very outdated) RAID
> article I found. I didn't think it would be a big deal for the
> reasons you mentioned. Plus it's sharing the interrupt with two USB
> ports which don't even have the motherboard adapter connected to them.

a) You can possibly disable those unused usb ports, some mobo bios's
allow you to somewhat granularly disable groups of ports.

b) Try changing the mobo BIOS from "PNP aware OS" enabled to disabled
or vice-versa.  That changes who sets the IRQ's, OS or BIOS.  If your
board has that option... Of course, it's still the luck of the draw in
either case whether IRQs get shared.  My take on it is these days you
just don't worry about it anymore.

From trevor at tecnopolis.ca  Wed Apr  7 16:11:49 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Wed, 7 Apr 2010 16:11:49 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <010c01cad60f$4cd85560$e6890020$@net>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
	<r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
	<010c01cad60f$4cd85560$e6890020$@net>
Message-ID: <20100407161149.33f6e109@pog.tecnopolis.ca>

On 2010-04-07 Adam Thompson wrote:
> Actually, I was mainly hoping to verify that it was, indeed, a
> hardware problem.  One person (Trevor) reporting similar results has
> fairly decent-quality GigE NICs on both sides ? or at least what I
> *assumed* to be fairly decent-quality NICs!

I should hope so, my NICs are Intel server grade gigabit on the server
and Intel high-end workstation grade gigabit on the client ($100-$300
NICs, retail).

Kevin, I didn't have time to scan your exact results, is it mostly
pc->server that's slow or server<-pc?  And your pc is Windows, I gather
(XP?).

My big problem has always been windows->linux performance (but never
linux->windows).  I've given up on it for now, but one thing that made
a HUGE difference was turning OFF jumbo packets.  I instantly got 5X
better performance with jumbo OFF.  Yes, my switch is jumbo capable,
and it was enabled, and set properly on the pc and linux.  Go figure.
I blame the Linksys WebSmart switch, but who knows.


From kevin.a.mcgregor at gmail.com  Sat Apr 10 17:52:58 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Sat, 10 Apr 2010 17:52:58 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <20100407161149.33f6e109@pog.tecnopolis.ca>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
	<r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
	<010c01cad60f$4cd85560$e6890020$@net>
	<20100407161149.33f6e109@pog.tecnopolis.ca>
Message-ID: <r2s6756caf11004101552na8e08f7evc3d96227dafaa6d8@mail.gmail.com>

Well. I plugged in my iMac (Intel, Core 2 Duo T7200, 2.00 GHz) to my gigabit
switch and ran iperf on it in server and client mode. The only copy I could
find for download (executable) was 1.70, and was compiled for the PowerPC
Macs. Here are the results between the iMac and my server:

$ iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 192.168.27.10 port 5001 connected with 192.168.27.29 port 49371
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-20.0 sec  2.17 GBytes    930 Mbits/sec

$ iperf -t 20 -c 192.168.27.29
------------------------------------------------------------
Client connecting to 192.168.27.29, TCP port 5001
TCP window size: 16.0 KByte (default)
------------------------------------------------------------
[  3] local 192.168.27.10 port 44201 connected with 192.168.27.29 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3]  0.0-20.0 sec  2.18 GBytes    936 Mbits/sec

The server and the iMac seem pretty happy to talk to each other -- that's
twice the performance of any other TCP result I've had! Just as a check, I
plugged the PC into the same cable as the iMac had been plugged into:

$ iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 85.3 KByte (default)
------------------------------------------------------------
[  4] local 192.168.27.10 port 5001 connected with 192.168.27.23 port 4701
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0-20.0 sec    386 MBytes    162 Mbits/sec

Much lower results. It seems to me that the problem is with the network
hardware or TCP/IP stack on the PC side. Does anyone else want to venture an
opinion? Or another test to run?

Kevin

On Wed, Apr 7, 2010 at 4:11 PM, Trevor Cordes <trevor at tecnopolis.ca> wrote:

> On 2010-04-07 Adam Thompson wrote:
> > Actually, I was mainly hoping to verify that it was, indeed, a
> > hardware problem.  One person (Trevor) reporting similar results has
> > fairly decent-quality GigE NICs on both sides ? or at least what I
> > *assumed* to be fairly decent-quality NICs!
>
> I should hope so, my NICs are Intel server grade gigabit on the server
> and Intel high-end workstation grade gigabit on the client ($100-$300
> NICs, retail).
>
> Kevin, I didn't have time to scan your exact results, is it mostly
> pc->server that's slow or server<-pc?  And your pc is Windows, I gather
> (XP?).
>
> My big problem has always been windows->linux performance (but never
> linux->windows).  I've given up on it for now, but one thing that made
> a HUGE difference was turning OFF jumbo packets.  I instantly got 5X
> better performance with jumbo OFF.  Yes, my switch is jumbo capable,
> and it was enabled, and set properly on the pc and linux.  Go figure.
> I blame the Linksys WebSmart switch, but who knows.
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100410/a021a0c2/attachment.html 

From gedetil at cs.umanitoba.ca  Wed Apr 14 10:42:07 2010
From: gedetil at cs.umanitoba.ca (Gilbert E. Detillieux)
Date: Wed, 14 Apr 2010 10:42:07 -0500
Subject: [RndTbl] Multiple simultaneous SSH connections
Message-ID: <4BC5E24F.8060700@cs.umanitoba.ca>

At last night's meeting, during the round-table discussion, someone 
asked about setting up simultaneous SSH connections to multiple host 
sytsems.

Here's the program I was thinking of...

ClusterSSH: http://sourceforge.net/projects/clusterssh/

For MacOS X, there's also csshX that interfaces to the Terminal.app...

csshX: http://code.google.com/p/csshx/

Adam also mentioned PSSH...

http://www.theether.org/pssh/docs/0.2.3/pssh-HOWTO.html
http://www.theether.org/pssh/

Both PSSH and Cluster SSH are described in the following article, along 
with yet another option called Multixterm...

http://www.ubuntugeek.com/execute-commands-simultaneously-on-multiple-servers-using-psshcluster-sshmultixterm.html

Hope that helps!

-- 
Gilbert E. Detillieux		E-mail: <gedetil at muug.mb.ca>
Manitoba UNIX User Group	Web:	http://www.muug.mb.ca/
PO Box 130 St-Boniface		Phone:  (204)474-8161
Winnipeg MB CANADA  R2H 3B4	Fax:    (204)474-7609

From sean at ertw.com  Wed Apr 14 10:50:51 2010
From: sean at ertw.com (Sean Walberg)
Date: Wed, 14 Apr 2010 10:50:51 -0500
Subject: [RndTbl] Multiple simultaneous SSH connections
In-Reply-To: <4BC5E24F.8060700@cs.umanitoba.ca>
References: <4BC5E24F.8060700@cs.umanitoba.ca>
Message-ID: <p2mcc43a8f81004140850nb2e9e322w1307b3182a5ab8d9@mail.gmail.com>

If you're at the point where you have to run the same commands on multiple
servers, you should really be looking at puppet or cfengine. They really
reduce the need to be interacting with the servers, and ensure consistency
(and by consistency I mean "fewer fat fingers").

I even gave a presentation mentioning the latter (
http://www.muug.mb.ca/meetings/lamp_tuning_seanwalberg.pdf) and have written
about the former (http://m.linuxjournal.com/article/10046)

Sean

On Wed, Apr 14, 2010 at 10:42 AM, Gilbert E. Detillieux <
gedetil at cs.umanitoba.ca> wrote:

> At last night's meeting, during the round-table discussion, someone
> asked about setting up simultaneous SSH connections to multiple host
> sytsems.
>
> Here's the program I was thinking of...
>
> ClusterSSH: http://sourceforge.net/projects/clusterssh/
>
> For MacOS X, there's also csshX that interfaces to the Terminal.app...
>
> csshX: http://code.google.com/p/csshx/
>
> Adam also mentioned PSSH...
>
> http://www.theether.org/pssh/docs/0.2.3/pssh-HOWTO.html
> http://www.theether.org/pssh/
>
> Both PSSH and Cluster SSH are described in the following article, along
> with yet another option called Multixterm...
>
>
> http://www.ubuntugeek.com/execute-commands-simultaneously-on-multiple-servers-using-psshcluster-sshmultixterm.html
>
> Hope that helps!
>
> --
> Gilbert E. Detillieux           E-mail: <gedetil at muug.mb.ca>
> Manitoba UNIX User Group        Web:    http://www.muug.mb.ca/
> PO Box 130 St-Boniface          Phone:  (204)474-8161
> Winnipeg MB CANADA  R2H 3B4     Fax:    (204)474-7609
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>



-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100414/1c28ab84/attachment.html 

From athompso at athompso.net  Wed Apr 14 11:09:37 2010
From: athompso at athompso.net (Adam Thompson)
Date: Wed, 14 Apr 2010 16:09:37 +0000
Subject: [RndTbl] Multiple simultaneous SSH connections
Message-ID: <136725748-1271261369-cardhu_decombobulator_blackberry.rim.net-416492734-@bda957.bisx.prod.on.blackberry>

'twas I who needed a solution.  The problem with cfengine (and possibly puppet?) is that I cannot install any software on the target servers and have to keep my on-system footprint to an absolute minimum.
I do, however, have one server of my own (so to speak) in the network that I can use for non-invasive monitoring, which can include running commands remotely via SSH.
-Adam


From kevin.a.mcgregor at gmail.com  Wed Apr 14 21:31:38 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Wed, 14 Apr 2010 21:31:38 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <r2s6756caf11004101552na8e08f7evc3d96227dafaa6d8@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
	<r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
	<010c01cad60f$4cd85560$e6890020$@net>
	<20100407161149.33f6e109@pog.tecnopolis.ca>
	<r2s6756caf11004101552na8e08f7evc3d96227dafaa6d8@mail.gmail.com>
Message-ID: <q2m6756caf11004141931g294076f4s61c2e9b8408ac172@mail.gmail.com>

And finally (I'm kinda done messing around with this): I took out all other
cards save for the AGP video card and re-ran iperf, and got 162 Mb/s with
the XP PC sending to the Linux server and 612 Mb/s sending from the server
to the PC. Unpleasantly asymmetric.

An abbreviated "lspci" (from the server) for those interested:
03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B
PCI Express Gigabit Ethernet controller (rev 01)
        Subsystem: ASUSTeK Computer Inc. Device 81aa
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin A routed to IRQ 26
        Region 0: I/O ports at de00 [size=256]
        Region 2: Memory at fdfff000 (64-bit, non-prefetchable) [size=4K]
...
        Kernel driver in use: r8169
        Kernel modules: r8169

Numerically, the PCI codes are 10ec:8168 (rev 01); the PC's card reports as
10ec:8169.

I'll wait a bit to see if any new AMD CPUs are coming out, and then buy a
quad- or hex-core CPU/motherboard/RAM and hope for a decent network
interface.

Kevin

On Sat, Apr 10, 2010 at 5:52 PM, Kevin McGregor
<kevin.a.mcgregor at gmail.com>wrote:

> Well. I plugged in my iMac (Intel, Core 2 Duo T7200, 2.00 GHz) to my
> gigabit switch and ran iperf on it in server and client mode. The only copy
> I could find for download (executable) was 1.70, and was compiled for the
> PowerPC Macs. Here are the results between the iMac and my server:
>
> $ iperf -s
> ------------------------------------------------------------
> Server listening on TCP port 5001
> TCP window size: 85.3 KByte (default)
> ------------------------------------------------------------
> [  4] local 192.168.27.10 port 5001 connected with 192.168.27.29 port 49371
> [ ID] Interval       Transfer     Bandwidth
> [  4]  0.0-20.0 sec  2.17 GBytes    930 Mbits/sec
>
> $ iperf -t 20 -c 192.168.27.29
> ------------------------------------------------------------
> Client connecting to 192.168.27.29, TCP port 5001
> TCP window size: 16.0 KByte (default)
> ------------------------------------------------------------
> [  3] local 192.168.27.10 port 44201 connected with 192.168.27.29 port 5001
> [ ID] Interval       Transfer     Bandwidth
> [  3]  0.0-20.0 sec  2.18 GBytes    936 Mbits/sec
>
> The server and the iMac seem pretty happy to talk to each other -- that's
> twice the performance of any other TCP result I've had! Just as a check, I
> plugged the PC into the same cable as the iMac had been plugged into:
>
> $ iperf -s
> ------------------------------------------------------------
> Server listening on TCP port 5001
> TCP window size: 85.3 KByte (default)
> ------------------------------------------------------------
> [  4] local 192.168.27.10 port 5001 connected with 192.168.27.23 port 4701
> [ ID] Interval       Transfer     Bandwidth
> [  4]  0.0-20.0 sec    386 MBytes    162 Mbits/sec
>
> Much lower results. It seems to me that the problem is with the network
> hardware or TCP/IP stack on the PC side. Does anyone else want to venture an
> opinion? Or another test to run?
>
> Kevin
>
> On Wed, Apr 7, 2010 at 4:11 PM, Trevor Cordes <trevor at tecnopolis.ca>wrote:
>
>> On 2010-04-07 Adam Thompson wrote:
>> > Actually, I was mainly hoping to verify that it was, indeed, a
>> > hardware problem.  One person (Trevor) reporting similar results has
>> > fairly decent-quality GigE NICs on both sides ? or at least what I
>> > *assumed* to be fairly decent-quality NICs!
>>
>> I should hope so, my NICs are Intel server grade gigabit on the server
>> and Intel high-end workstation grade gigabit on the client ($100-$300
>> NICs, retail).
>>
>> Kevin, I didn't have time to scan your exact results, is it mostly
>> pc->server that's slow or server<-pc?  And your pc is Windows, I gather
>> (XP?).
>>
>> My big problem has always been windows->linux performance (but never
>> linux->windows).  I've given up on it for now, but one thing that made
>> a HUGE difference was turning OFF jumbo packets.  I instantly got 5X
>> better performance with jumbo OFF.  Yes, my switch is jumbo capable,
>> and it was enabled, and set properly on the pc and linux.  Go figure.
>> I blame the Linksys WebSmart switch, but who knows.
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100414/52588384/attachment.html 

From trevor at tecnopolis.ca  Thu Apr 15 01:30:53 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Thu, 15 Apr 2010 01:30:53 -0500
Subject: [RndTbl] squid caching of Windblows Updates
Message-ID: <20100415013053.4655ecbf@pog.tecnopolis.ca>

As discussed at the Apr meeting, here is my solution for having squid
cache your Windblows (XP, mainly but should work for Vista/7) updates
downloads.

I have just verified this is still working at all my customer sites by
looking at the squid access logs and seeing HITs for the big CAB files
from windowsupdate.com.  This confirms what I have always been seeing
while manually doing updates.

The magic line you require in your squid.conf is:
refresh_pattern http://www.download.windowsupdate.com/ 0 80% 525600
reload-into-ims
(no wrapping on the above line)

If someone wants my entire squid.conf, let me know.  It's setup for a
pretty common case where you want to squid all of your internal
(192.168.*) computers.  The only other major thing is make sure you
have a pretty big squid cache and max object size.  I recommend at
least 3GB.  If you have tons of disk, make it 50G or something.  More
never hurts.

I use:
cache_swap_low 94
cache_swap_high 97
maximum_object_size 800000 KB

Which should nicely handle the Win Update case we're trying to achieve.

Now, to make it so you don't have to configure ANYTHING on a stock
WinXP box to make it use squid:

Make a file on your web server's document root (ie: /var/www/html or
similar) called wpad.dat.  Make it world readable.  Populate it thusly:

<? header("Content-type: application/x-ns-proxy-autoconfig"); ?>

function FindProxyForURL(url, host)
{
  if (	 isPlainHostName(host)
	|| dnsDomainIs	(host,"tecnopolis.ca")
	|| dnsDomainIs	(host,"nocachingdomain.whatever")
	|| shExpMatch	(host,"192.168.*")
	|| shExpMatch	(url ,"*windowsupdate.microsoft.com/*.js*")
	|| shExpMatch(url ,"*windowsupdate.microsoft.com/*.asp*") 
     )
     return "DIRECT";

  if (url.substring(0, 5) == "http:" ||
      url.substring(0, 4) == "ftp:"  ||
      url.substring(0, 7) == "gopher:")
    return "PROXY 192.168.1.1:8080; DIRECT";

  return "DIRECT";
}

Tweak the DIRECT domain/IP exclusions in the first if set.  Change the
192.168.1.1 to your squid server's IP.  Change the 8080 to whatever
port you want to have squid listen to.  There's a more standard default
port, but I use 8080 as that's what Shaw used when I signed up ages ago
and it's easy to remember :-)

You also require some funky apache config to make it work:
<VirtualHost *:80>
  ServerName    wpad.tecnopolis.ca
  ServerAlias   wpad
  DocumentRoot  /var/www/html
  Alias         /wpad.dat /var/www/html/wpad.dat
</VirtualHost>

(may be a bit redundant, but I am editing my slightly more complex
setup to be usable by others)

Note, wpad.dat only applies to IE (Internet Exploder) to make
Nutscrape-based browsers (Fireflax) you need to make another identical
file (hardlink?) called proxy.pac and do some more funky apache stuff:

<VirtualHost *:80>
  ServerName    proxy.tecnopolis.ca
  ServerAlias   proxy
  DocumentRoot  /var/www/html
  Alias         /wpad.dat /var/www/html/wpad.dat
  Alias         / /var/www/html/proxy.pac
</VirtualHost>

More redundancy, but trust me, getting this stuff to work is mystical
voodoo so cover all your bases.

AND, obviously add proxy and wpad as CNAMEs in your DNS server's
config!  They must resolve to your web server's IP.

Lastly, XP by default has the "autodetect proxy settings" enabled in
IE, so you literally have to do nothing to have your entire network
magically start using the proxy.  If it doesn't seem to be working,
double-check that IE indeed has that option enabled (Tools,
connections, lan settings).

Like I said, it's been 10 years since I set this up and last looked it
it, so hopefully I'm not overlooking some important point.  It
literally has worked all these years without tweaking on dozens of
servers.  3 cheers for UN*X and it's stability of configuration and
migration!

From sean at tinfoilhat.ca  Thu Apr 15 01:39:17 2010
From: sean at tinfoilhat.ca (Sean Cody)
Date: Thu, 15 Apr 2010 01:39:17 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <20100415013053.4655ecbf@pog.tecnopolis.ca>
References: <20100415013053.4655ecbf@pog.tecnopolis.ca>
Message-ID: <EE4C6D80-DACC-4A10-ACA0-7FBEAE86D25D@tinfoilhat.ca>

In case one has some older versions of IE, sym-linking wpad.dat to wpad.da solve some of those corner cases.

Some versions of IE and XP (especially the freshly unpatched kind) have a bug in proxy detection where it doesn't calculate the size of the string properly and null-terminates it one character short which is why the last letter is chopped off of the end (bad coder, no cookie).

May not still be the case anymore but I remember back when we rolled out wpad it gave me a really big headache. :)

On 2010-04-15, at 1:30 AM, Trevor Cordes wrote:

> As discussed at the Apr meeting, here is my solution for having squid
> cache your Windblows (XP, mainly but should work for Vista/7) updates
> downloads.
> 
> I have just verified this is still working at all my customer sites by
> looking at the squid access logs and seeing HITs for the big CAB files
> from windowsupdate.com.  This confirms what I have always been seeing
> while manually doing updates.
> 
> The magic line you require in your squid.conf is:
> refresh_pattern http://www.download.windowsupdate.com/ 0 80% 525600
> reload-into-ims
> (no wrapping on the above line)
> 
> If someone wants my entire squid.conf, let me know.  It's setup for a
> pretty common case where you want to squid all of your internal
> (192.168.*) computers.  The only other major thing is make sure you
> have a pretty big squid cache and max object size.  I recommend at
> least 3GB.  If you have tons of disk, make it 50G or something.  More
> never hurts.
> 
> I use:
> cache_swap_low 94
> cache_swap_high 97
> maximum_object_size 800000 KB
> 
> Which should nicely handle the Win Update case we're trying to achieve.
> 
> Now, to make it so you don't have to configure ANYTHING on a stock
> WinXP box to make it use squid:
> 
> Make a file on your web server's document root (ie: /var/www/html or
> similar) called wpad.dat.  Make it world readable.  Populate it thusly:
> 
> <? header("Content-type: application/x-ns-proxy-autoconfig"); ?>
> 
> function FindProxyForURL(url, host)
> {
>  if (	 isPlainHostName(host)
> 	|| dnsDomainIs	(host,"tecnopolis.ca")
> 	|| dnsDomainIs	(host,"nocachingdomain.whatever")
> 	|| shExpMatch	(host,"192.168.*")
> 	|| shExpMatch	(url ,"*windowsupdate.microsoft.com/*.js*")
> 	|| shExpMatch(url ,"*windowsupdate.microsoft.com/*.asp*") 
>     )
>     return "DIRECT";
> 
>  if (url.substring(0, 5) == "http:" ||
>      url.substring(0, 4) == "ftp:"  ||
>      url.substring(0, 7) == "gopher:")
>    return "PROXY 192.168.1.1:8080; DIRECT";
> 
>  return "DIRECT";
> }
> 
> Tweak the DIRECT domain/IP exclusions in the first if set.  Change the
> 192.168.1.1 to your squid server's IP.  Change the 8080 to whatever
> port you want to have squid listen to.  There's a more standard default
> port, but I use 8080 as that's what Shaw used when I signed up ages ago
> and it's easy to remember :-)
> 
> You also require some funky apache config to make it work:
> <VirtualHost *:80>
>  ServerName    wpad.tecnopolis.ca
>  ServerAlias   wpad
>  DocumentRoot  /var/www/html
>  Alias         /wpad.dat /var/www/html/wpad.dat
> </VirtualHost>
> 
> (may be a bit redundant, but I am editing my slightly more complex
> setup to be usable by others)
> 
> Note, wpad.dat only applies to IE (Internet Exploder) to make
> Nutscrape-based browsers (Fireflax) you need to make another identical
> file (hardlink?) called proxy.pac and do some more funky apache stuff:
> 
> <VirtualHost *:80>
>  ServerName    proxy.tecnopolis.ca
>  ServerAlias   proxy
>  DocumentRoot  /var/www/html
>  Alias         /wpad.dat /var/www/html/wpad.dat
>  Alias         / /var/www/html/proxy.pac
> </VirtualHost>
> 
> More redundancy, but trust me, getting this stuff to work is mystical
> voodoo so cover all your bases.
> 
> AND, obviously add proxy and wpad as CNAMEs in your DNS server's
> config!  They must resolve to your web server's IP.
> 
> Lastly, XP by default has the "autodetect proxy settings" enabled in
> IE, so you literally have to do nothing to have your entire network
> magically start using the proxy.  If it doesn't seem to be working,
> double-check that IE indeed has that option enabled (Tools,
> connections, lan settings).
> 
> Like I said, it's been 10 years since I set this up and last looked it
> it, so hopefully I'm not overlooking some important point.  It
> literally has worked all these years without tweaking on dozens of
> servers.  3 cheers for UN*X and it's stability of configuration and
> migration!
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable

-- 
Sean




From trevor at tecnopolis.ca  Thu Apr 15 02:14:51 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Thu, 15 Apr 2010 02:14:51 -0500
Subject: [RndTbl] Network performance tuning
In-Reply-To: <q2m6756caf11004141931g294076f4s61c2e9b8408ac172@mail.gmail.com>
References: <m2i6756caf11004051727i169498f3me55209f6d20ebcd0@mail.gmail.com>
	<i2qcc43a8f81004051812m2ae7c272y8c5d8498fa02fa3a@mail.gmail.com>
	<o2n6756caf11004051943x6880164cv94b3dd5eaebfef09@mail.gmail.com>
	<r2kcc43a8f81004051952hea95194w8419012aa41b471c@mail.gmail.com>
	<r2n6756caf11004061636j27d5e4a5obfe465ddd504dcde@mail.gmail.com>
	<010c01cad60f$4cd85560$e6890020$@net>
	<20100407161149.33f6e109@pog.tecnopolis.ca>
	<r2s6756caf11004101552na8e08f7evc3d96227dafaa6d8@mail.gmail.com>
	<q2m6756caf11004141931g294076f4s61c2e9b8408ac172@mail.gmail.com>
Message-ID: <20100415021451.04570f6a@pog.tecnopolis.ca>

On 2010-04-14 Kevin McGregor wrote:
> And finally (I'm kinda done messing around with this): I took out all
> other cards save for the AGP video card and re-ran iperf, and got 162
> Mb/s with the XP PC sending to the Linux server and 612 Mb/s sending
> from the server to the PC. Unpleasantly asymmetric.

Finally had a mo to test my systems, which have similar problems.

my comps:
pog   : Fedora 12, Core2Q, PRO/1000 MT Server NIC
piles : Fedora 10, PD 3G , PRO/1000 MT Server NIC
peecee: XP, Core i7, PRO/1000 PT Desktop NIC (PCIe)
switch: Linksys WebSmart SRW2016, jumbo OFF

All NICs have as much offloading turned on as possible.

iperf results in order of slow to fast:

-s	-t
piles   peecee  300
pog     peecee  351
peecee  pog     400
peecee  piles   401
piles	pog     743
pog	piles	744

Now, I can't seem to figure out whether it's the -s that's sending or
the -t, so I just list as above.  Very consistent results with anything
involving the PC (my fastest hardware!) being much slower than I get
linux->linux.  Sure, the peecee has a "lesser" NIC, but it's still an
expensive one.

The above results mirror what I see in daily life going from piles
(file server) to peecee using samba (the only thing I care about from
peecee).

When I have another mo, I'll boot a live linux CD and test peecee with
that to see if it's the hw to blame or simply XP.  I've seen lots of
reports about XP having braindead TCP and regedit tweaks to make it
faster.  I've tried many of them with little success.

If anyone wants to also guess as to why my network speed about doubles
when I turn OFF jumbo packets(!!!), please be my guest!  (Yes, I'm
pretty sure I had jumbos configured properly on all NICs/switch, etc.)

> I'll wait a bit to see if any new AMD CPUs are coming out, and then
> buy a quad- or hex-core CPU/motherboard/RAM and hope for a decent
> network interface.

Or buy a nice NIC now :-)  The onboard NICs are usually always
substandard and often crippled (no jumbo, etc).

From rbdyck2 at shaw.ca  Thu Apr 15 07:32:40 2010
From: rbdyck2 at shaw.ca (Robert Dyck)
Date: Thu, 15 Apr 2010 07:32:40 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <EE4C6D80-DACC-4A10-ACA0-7FBEAE86D25D@tinfoilhat.ca>
Message-ID: <000001cadc97$bd23e8a0$6400a8c0@wp.shawcable.net>

The organization I work for is wants to set up a proxy server for Windows
updates. We work with older computers, XP and 2000. We have a server
available, and a couple of us want to use Ubuntu server. The individual who
suggested a proxy server also suggested Squid. Great ideas. Do you have any
recommendations to setup Squid for Windows 2000 and version of IE that will
run on it?

Trevor, you offered to send your entire squid.conf file; could I get that
please?

Also, my college suggested getting software to make the server a router,
while I had considered getting a hardware router and just configuring the
server as a proxy server. Any recommendations about topology or software?

Thanks,
Rob Dyck

Ps. I'm the guy who was trying to set up a server using CRUX. The problem
was an error message "Spurious ACK on isa0060/serio0. Some program might be
trying access hardware directly" causing the kernel to panic trap. Never was
able to get past that. So I gave up, instead installed Ubuntu 9.10 server.
It was smooth; from start of download to complete up-and-running system was
a single day.




From trevor at tecnopolis.ca  Thu Apr 15 09:29:15 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Thu, 15 Apr 2010 09:29:15 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <000001cadc97$bd23e8a0$6400a8c0@wp.shawcable.net>
References: <EE4C6D80-DACC-4A10-ACA0-7FBEAE86D25D@tinfoilhat.ca>
	<000001cadc97$bd23e8a0$6400a8c0@wp.shawcable.net>
Message-ID: <20100415092915.55d180b2@pog.tecnopolis.ca>

On 2010-04-15 Robert Dyck wrote:
> The organization I work for is wants to set up a proxy server for
> Windows updates. We work with older computers, XP and 2000. We have a

The setup I suggest *should* also support Win2k ok.  I *think* I
remember it being ok back in the day when I had customers with w2k.
For sure it supports wpad.dat, so as long as it hits windowsupdate.com
for updates, it should work.  Even w98 may work.

> server available, and a couple of us want to use Ubuntu server. The

Any distro should work fine.  I prefer Fedora, but that's just me!

> Trevor, you offered to send your entire squid.conf file; could I get
> that please?

I will email it directly.  Obviously it will have to be tweaked for
your site.

> Also, my college suggested getting software to make the server a
> router, while I had considered getting a hardware router and just
> configuring the server as a proxy server. Any recommendations about
> topology or software?

All my squid caches are also routers/firewalls.  Seems silly to
dedicate multiple boxes to what 1 can do quite well, but that's my
personal preference.  All my routers/firewalls are 100% custom.  If
you're looking for an inexpensive turnkey and 100% managed solution,
let me know as that's what my company does, we can even use your old
hardware (tailored for micro and small business).

From high.res.mike at gmail.com  Thu Apr 15 10:32:37 2010
From: high.res.mike at gmail.com (Mike Pfaiffer)
Date: Thu, 15 Apr 2010 10:32:37 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <20100415092915.55d180b2@pog.tecnopolis.ca>
References: <EE4C6D80-DACC-4A10-ACA0-7FBEAE86D25D@tinfoilhat.ca>
	<000001cadc97$bd23e8a0$6400a8c0@wp.shawcable.net>
	<20100415092915.55d180b2@pog.tecnopolis.ca>
Message-ID: <32D49599-BD70-43BD-89AC-D539BF88516D@gmail.com>

On 2010-04-15, at 9:29 AM, Trevor Cordes wrote:

> On 2010-04-15 Robert Dyck wrote:
>> The organization I work for is wants to set up a proxy server for
>> Windows updates. We work with older computers, XP and 2000. We have a
> 
> The setup I suggest *should* also support Win2k ok.  I *think* I
> remember it being ok back in the day when I had customers with w2k.
> For sure it supports wpad.dat, so as long as it hits windowsupdate.com
> for updates, it should work.  Even w98 may work.

	We're part of the same group (it was yours truly who brought up the original topic).

	The thing is we can't really control what machines we get. Sometime it's XP and other times it's 2K. Some times we are lucky and the XP machines have been updated to the previous update. It doesn't look like we'll be getting Vista or W-7 licenses in the near future. The machines couldn't run it anyway.

	In addition to the Windows updates we also have firewall, anti-virus, and anti-spyware. We also install a ZIP package. It used to be Winzip but we found 7-zip does more things. 7-zip installs directly from the net. The idea was we could knock off 45min - 1 hour per machine installation. The connection to the net gets pretty slow sometimes. That's another issue beyond our control.

>> server available, and a couple of us want to use Ubuntu server. The
> 
> Any distro should work fine.  I prefer Fedora, but that's just me!

	The thing is we are "lightweights" when it comes to how to install this sort of thing. The reason I suggested Ubuntu server is because it's available and it is a very quick install. I have an hour at the end of the day to do extra things around the lab so I figured we could do it a little at a time.

	We don't actually have a web server. I mention it because previous discussion suggested it. The machine we have at the moment is a single FreeNAS box. We have hardware ready to go. Just no idea how to proceed.

>> Trevor, you offered to send your entire squid.conf file; could I get
>> that please?
> 
> I will email it directly.  Obviously it will have to be tweaked for
> your site.

	Again, we are "lightweights". The more directions the better.

>> Also, my college suggested getting software to make the server a
>> router, while I had considered getting a hardware router and just
>> configuring the server as a proxy server. Any recommendations about
>> topology or software?
> 
> All my squid caches are also routers/firewalls.  Seems silly to
> dedicate multiple boxes to what 1 can do quite well, but that's my
> personal preference.

	Here is our current setup...
AC Router <--> CLL Hub <--> Various machines

	We don't control the AC Router. We can put in requests but they are having problems with Barracuda at the moment so it will be quite a while before they can help out.

	My thoughts are if we can make the whole process transparent to the machines being installed/repaired it would save us a bunch of time. To do this I figured we'd need a router between the incoming connection and the hub. OTOH, if the machine were on the other side of the hub we'd end up having to configure the machines and restore them when we were done.

	This is what *I* think we'll have to do...
AC Router <--> CLL Router/Squid <--> CLL Hub <--> Various machines

	I don't know enough about Squid to know if we can get rid of the router part of the machine then have it sit between the AC Router and our machines. Our machines are generally set up for DHCP so my concern would be where we get the IPs from.

>  All my routers/firewalls are 100% custom.  If
> you're looking for an inexpensive turnkey and 100% managed solution,
> let me know as that's what my company does, we can even use your old
> hardware (tailored for micro and small business).

	Bear in mind this may or may not last a while. Given the games M$ seems to be playing with licenses it may come down to Linux (and old Macs) or nothing. AFAIC, that would be fine. Unfortunately the short sighted behaviour of many employers means unless we give out M$ machines the clients are mostly ignored. It would be very hard to justify coming up with money to pay someone to install it for us (I don't get paid and I'm the senior volunteer - I don't think the supervisor gets paid either). I think I could persuade them to buy you a lunch at Subway though... :-)

				Later
				Mike



From rbdyck2 at shaw.ca  Thu Apr 15 11:17:42 2010
From: rbdyck2 at shaw.ca (Robert Dyck)
Date: Thu, 15 Apr 2010 11:17:42 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <32D49599-BD70-43BD-89AC-D539BF88516D@gmail.com>
Message-ID: <000b01cadcb7$2d6378a0$6400a8c0@wp.shawcable.net>

On 2010-04-15, at 9:33 AM, Mike Pfaiffer wrote:
> The thing is we are "lightweights"

Hey! I'm still trying to get a job as a system or network admin.

Immodest brag line: 29 years experience in IT, available immediately.

Rob Dyck



From trevor at tecnopolis.ca  Thu Apr 15 12:20:30 2010
From: trevor at tecnopolis.ca (Trevor Cordes)
Date: Thu, 15 Apr 2010 12:20:30 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <32D49599-BD70-43BD-89AC-D539BF88516D@gmail.com>
References: <EE4C6D80-DACC-4A10-ACA0-7FBEAE86D25D@tinfoilhat.ca>
	<000001cadc97$bd23e8a0$6400a8c0@wp.shawcable.net>
	<20100415092915.55d180b2@pog.tecnopolis.ca>
	<32D49599-BD70-43BD-89AC-D539BF88516D@gmail.com>
Message-ID: <20100415122030.302e4431@pog.tecnopolis.ca>

On 2010-04-15 Mike Pfaiffer wrote:
> 	In addition to the Windows updates we also have firewall,
> anti-virus, and anti-spyware. We also install a ZIP package. It used

To save yourself time, you should do as someone mentioned and
slipstream at minimum SP3 into your XP install CD.  This sounds
daunting, but I will give you my cheatsheet and you can do it in 1 page
of commands without any thinking:

(install packages first: wine, convmv, cabextract; and find the
script on the net called geteltorito, or ask me for it)

tcsh
set workdir=/tmp/Slip
set  spfile="/tmp/WinXP_SP3.exe"
set   spdir=$workdir/sp
set   indir=$workdir/new
set   cdrom=/dev/sr0
set   cdmnt=/media/cdrecorder

mkdir $workdir
cd $workdir

cabextract -d $spdir "$spfile"

mkdir $indir

mount $cdrom $cdmnt

cp -r $cdmnt/* $indir
chmod -R 777 $indir

wine $spdir/i386/update/update.exe -s:$indir

convmv -r --upper --notest $indir/*

geteltorito $cdrom > $indir/boot.bin

find $indir | xargs touch -t 200804140000

umount $cdmnt
eject $cdrom

cd $indir
mkisofs -b boot.bin -hide boot.bin -hide boot.catalog -no-emul-boot \
-boot-load-size 4 -iso-level 4 -relaxed-filenames -D -V GRTMPVOL_EN -o \
$workdir/iso .

cdrecord dev=$cdrom $workdir/iso
eject $cdrom
rm $workdir/iso

============

Doing the above (on linux) never ceases to amaze me (I've never used
wine before).

> 	The thing is we are "lightweights" when it comes to how to
> install this sort of thing. The reason I suggested Ubuntu server is

Nothing wrong with Ubuntu at all, just use what you're fastest with
configuring, that's what I say.  You get used to a distro's way of
doing things (mostly file system layout and /etc arrangement), and for
me that's RedHat 6 (c 1999) or so :-)

> 	We don't control the AC Router. We can put in requests but
> they are having problems with Barracuda at the moment so it will be

If you don't control the router then probably best to do the
router-behind-router idea you outlined, just make sure to pick a
different local subnet IP range!

> 	My thoughts are if we can make the whole process transparent
> to the machines being installed/repaired it would save us a bunch of
> time. To do this I figured we'd need a router between the incoming
> connection and the hub. OTOH, if the machine were on the other side
> of the hub we'd end up having to configure the machines and restore
> them when we were done.

For easiest transparency (no settings to be made on client), you want
the wpad to work, which means you must control your DNS, web and squid
server.  If you can live with a 60sec tweak on every client (setting
proxy manually) then you can do away with DNS and web and just setup a
squid server that can be in your existing subnet (no need for another
router).

> 	This is what *I* think we'll have to do...
> AC Router <--> CLL Router/Squid <--> CLL Hub <--> Various machines

As above, best/transparent solution but a fair bit harder to initially
implement.

> 	I don't know enough about Squid to know if we can get rid of
> the router part of the machine then have it sit between the AC Router
> and our machines. Our machines are generally set up for DHCP so my
> concern would be where we get the IPs from.

If you did squid but no DNS/web then the squid server would just be
another peer on the LAN, just another box off the main switch.  Ideally
you'd want to assign it a static IP.

> out M$ machines the clients are mostly ignored. It would be very hard
> to justify coming up with money to pay someone to install it for us
> (I don't get paid and I'm the senior volunteer - I don't think the
> supervisor gets paid either). I think I could persuade them to buy
> you a lunch at Subway though... :-)

I hear ya.  I'm a bit over-subscribed for work, but I could possibly
squeeze in some moments here and there if you can arrange remote ssh
access from my IP.  Going onsite would be a royal pain given my schedule
though.  The roundtable (incl myself) is also here to help out with
questions.

If you've never done any BIND config before, that will probably be the
biggest challenge for you.  The apache stuff should be fairly easy and
the squid stuff extremely easy with my conf file.

From peter at pogma.com  Thu Apr 15 15:26:27 2010
From: peter at pogma.com (Peter O'Gorman)
Date: Thu, 15 Apr 2010 15:26:27 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <20100415013053.4655ecbf@pog.tecnopolis.ca>
References: <20100415013053.4655ecbf@pog.tecnopolis.ca>
Message-ID: <4BC77673.1070500@pogma.com>

On 04/15/2010 01:30 AM, Trevor Cordes wrote:

> 
> AND, obviously add proxy and wpad as CNAMEs in your DNS server's
> config!  They must resolve to your web server's IP.
> 

oooh, I've never seen this before, I had thought you were talking about
DHCP option 252 (DHCP server can tell the client the location of the
wpad file to use). Doing it in DNS is interesting!

Then again, I know almost nothing about Windows :)

Peter


From sean at tinfoilhat.ca  Thu Apr 15 15:31:57 2010
From: sean at tinfoilhat.ca (Sean Cody)
Date: Thu, 15 Apr 2010 15:31:57 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <4BC77673.1070500@pogma.com>
References: <20100415013053.4655ecbf@pog.tecnopolis.ca>
	<4BC77673.1070500@pogma.com>
Message-ID: <B1005663-3AEE-4B27-828A-AE079D1C342D@tinfoilhat.ca>

On 2010-04-15, at 3:26 PM, Peter O'Gorman wrote:
> oooh, I've never seen this before, I had thought you were talking about
> DHCP option 252 (DHCP server can tell the client the location of the
> wpad file to use). Doing it in DNS is interesting!
> 
> Then again, I know almost nothing about Windows :)
> 


I have never seen option 252 work.  I think I have a scar from the wall I banged my head against due to it.
I wish it did, work especially on non-windows hosts as forgetting to set http_proxy is something I ALWAYS forget to do.

-- 
Sean




From athompso at athompso.net  Thu Apr 15 15:38:51 2010
From: athompso at athompso.net (Adam Thompson)
Date: Thu, 15 Apr 2010 20:38:51 +0000
Subject: [RndTbl] squid caching of Windblows Updates
Message-ID: <1958681838-1271363924-cardhu_decombobulator_blackberry.rim.net-380991629-@bda957.bisx.prod.on.blackberry>

IIRC the DHCP option works for OS X.

From peter at pogma.com  Thu Apr 15 15:52:22 2010
From: peter at pogma.com (Peter O'Gorman)
Date: Thu, 15 Apr 2010 15:52:22 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <B1005663-3AEE-4B27-828A-AE079D1C342D@tinfoilhat.ca>
References: <20100415013053.4655ecbf@pog.tecnopolis.ca>	<4BC77673.1070500@pogma.com>
	<B1005663-3AEE-4B27-828A-AE079D1C342D@tinfoilhat.ca>
Message-ID: <4BC77C86.70508@pogma.com>

On 04/15/2010 03:31 PM, Sean Cody wrote:
> On 2010-04-15, at 3:26 PM, Peter O'Gorman wrote:
>> oooh, I've never seen this before, I had thought you were talking about
>> DHCP option 252 (DHCP server can tell the client the location of the
>> wpad file to use). Doing it in DNS is interesting!
>>
>> Then again, I know almost nothing about Windows :)
>>
> 
> 
> I have never seen option 252 work.  I think I have a scar from the wall I banged my head against due to it.
> I wish it did, work especially on non-windows hosts as forgetting to set http_proxy is something I ALWAYS forget to do.
> 

Didn't work when I tried it either - it's just that that's what I
thought Trevor was talking about.

Ah, and the DNS wpad thing works on Mac OS X 10.6 too (as long as the
user has opened System Preferences, clicked on "Network" clicked on
"Advanced..." then on "Proxies" and then checked the box "Auto Proxy
Discovery"... sigh).

Peter

From high.res.mike at gmail.com  Fri Apr 16 07:57:42 2010
From: high.res.mike at gmail.com (Mike Pfaiffer)
Date: Fri, 16 Apr 2010 07:57:42 -0500
Subject: [RndTbl] squid caching of Windblows Updates
In-Reply-To: <20100415122030.302e4431@pog.tecnopolis.ca>
References: <EE4C6D80-DACC-4A10-ACA0-7FBEAE86D25D@tinfoilhat.ca>
	<000001cadc97$bd23e8a0$6400a8c0@wp.shawcable.net>
	<20100415092915.55d180b2@pog.tecnopolis.ca>
	<32D49599-BD70-43BD-89AC-D539BF88516D@gmail.com>
	<20100415122030.302e4431@pog.tecnopolis.ca>
Message-ID: <22DD19A5-DA64-4425-BB1A-E25B4389D966@gmail.com>

On 2010-04-15, at 12:20 PM, Trevor Cordes wrote:

> On 2010-04-15 Mike Pfaiffer wrote:
>> 	In addition to the Windows updates we also have firewall,
>> anti-virus, and anti-spyware. We also install a ZIP package. It used
> 
> To save yourself time, you should do as someone mentioned and
> slipstream at minimum SP3 into your XP install CD.  This sounds
> daunting, but I will give you my cheatsheet and you can do it in 1 page
> of commands without any thinking:

	I think we may have SP3 already slipstreamed. I seem to recall something on a CD. I'll check it out when I get in to the lab in an hour.

> (install packages first: wine, convmv, cabextract; and find the
> script on the net called geteltorito, or ask me for it)
> 
> tcsh
> set workdir=/tmp/Slip
> set  spfile="/tmp/WinXP_SP3.exe"
> set   spdir=$workdir/sp
> set   indir=$workdir/new
> set   cdrom=/dev/sr0
> set   cdmnt=/media/cdrecorder
> 
> mkdir $workdir
> cd $workdir
> 
> cabextract -d $spdir "$spfile"
> 
> mkdir $indir
> 
> mount $cdrom $cdmnt
> 
> cp -r $cdmnt/* $indir
> chmod -R 777 $indir
> 
> wine $spdir/i386/update/update.exe -s:$indir
> 
> convmv -r --upper --notest $indir/*
> 
> geteltorito $cdrom > $indir/boot.bin
> 
> find $indir | xargs touch -t 200804140000
> 
> umount $cdmnt
> eject $cdrom
> 
> cd $indir
> mkisofs -b boot.bin -hide boot.bin -hide boot.catalog -no-emul-boot \
> -boot-load-size 4 -iso-level 4 -relaxed-filenames -D -V GRTMPVOL_EN -o \
> $workdir/iso .
> 
> cdrecord dev=$cdrom $workdir/iso
> eject $cdrom
> rm $workdir/iso
> 
> ============
> 
> Doing the above (on linux) never ceases to amaze me (I've never used
> wine before).

	I had wine configured properly ONCE. It never worked for me again. I'll follow your directions when things get organised. Who knows... Lightning may strike twice. ;-)

>> 	The thing is we are "lightweights" when it comes to how to
>> install this sort of thing. The reason I suggested Ubuntu server is
> 
> Nothing wrong with Ubuntu at all, just use what you're fastest with
> configuring, that's what I say.  You get used to a distro's way of
> doing things (mostly file system layout and /etc arrangement), and for
> me that's RedHat 6 (c 1999) or so :-)

	I like the text interface of Ubuntu server. I'm not sure I'd go with a GUI at this point. No sense in wasting clock cycles unless we have to.

>> 	We don't control the AC Router. We can put in requests but
>> they are having problems with Barracuda at the moment so it will be
> 
> If you don't control the router then probably best to do the
> router-behind-router idea you outlined, just make sure to pick a
> different local subnet IP range!

	Makes sense. We don't want to compete with the AC.

>> 	My thoughts are if we can make the whole process transparent
>> to the machines being installed/repaired it would save us a bunch of
>> time. To do this I figured we'd need a router between the incoming
>> connection and the hub. OTOH, if the machine were on the other side
>> of the hub we'd end up having to configure the machines and restore
>> them when we were done.
> 
> For easiest transparency (no settings to be made on client), you want
> the wpad to work, which means you must control your DNS, web and squid
> server.  If you can live with a 60sec tweak on every client (setting
> proxy manually) then you can do away with DNS and web and just setup a
> squid server that can be in your existing subnet (no need for another
> router).

	I understood about 80% of this paragraph. I think I should be able to understand the rest when we actually start the process.

>> 	This is what *I* think we'll have to do...
>> AC Router <--> CLL Router/Squid <--> CLL Hub <--> Various machines
> 
> As above, best/transparent solution but a fair bit harder to initially
> implement.

	The advantages also include not having to set it up every time.

>> 	I don't know enough about Squid to know if we can get rid of
>> the router part of the machine then have it sit between the AC Router
>> and our machines. Our machines are generally set up for DHCP so my
>> concern would be where we get the IPs from.
> 
> If you did squid but no DNS/web then the squid server would just be
> another peer on the LAN, just another box off the main switch.  Ideally
> you'd want to assign it a static IP.

	I thought this may be the case.

>> out M$ machines the clients are mostly ignored. It would be very hard
>> to justify coming up with money to pay someone to install it for us
>> (I don't get paid and I'm the senior volunteer - I don't think the
>> supervisor gets paid either). I think I could persuade them to buy
>> you a lunch at Subway though... :-)
> 
> I hear ya.  I'm a bit over-subscribed for work, but I could possibly
> squeeze in some moments here and there if you can arrange remote ssh
> access from my IP.  Going onsite would be a royal pain given my schedule
> though.  The roundtable (incl myself) is also here to help out with
> questions.

	I'll have to go through channels to ask. Given the AC has installed Barracuda my guess is they probably will say no.

	We could arrange to have the machine dropped off to you somewhere if that would be any better... 

	Worst case would be we work on it an hour per week and report the progress back here.

	Sounds like this has captured the imagination of the group. I wonder if it might be worth a future presentation... ;-)

> If you've never done any BIND config before, that will probably be the
> biggest challenge for you.  The apache stuff should be fairly easy and
> the squid stuff extremely easy with my conf file.

	As I mentioned I am a lightweight with this. The whole computer area is very broad and this isn't an area I had experience with until very recently.

				Later
				Mike





From peter at pogma.com  Fri Apr 16 10:22:27 2010
From: peter at pogma.com (Peter O'Gorman)
Date: Fri, 16 Apr 2010 10:22:27 -0500
Subject: [RndTbl] Multiple simultaneous SSH connections
In-Reply-To: <136725748-1271261369-cardhu_decombobulator_blackberry.rim.net-416492734-@bda957.bisx.prod.on.blackberry>
References: <136725748-1271261369-cardhu_decombobulator_blackberry.rim.net-416492734-@bda957.bisx.prod.on.blackberry>
Message-ID: <4BC880B3.3010706@pogma.com>

On 04/14/2010 11:09 AM, Adam Thompson wrote:
> 'twas I who needed a solution.  The problem with cfengine (and possibly puppet?) is that I cannot install any software on the target servers and have to keep my on-system footprint to an absolute minimum.
> I do, however, have one server of my own (so to speak) in the network that I can use for non-invasive monitoring, which can include running commands remotely via SSH.

Another possible solution if you use screen(1) is to use its 'At'
command e.g. C-a : to get the screen command line, then 'at "#" stuff
"ls\015"' will put a ls\r in the input buffer of every screen window, as
though you'd typed it there.

A co-worker uses some elisp terminal multiplexer, if you're an emacs
user I can ask him about it.

Peter

From athompso at athompso.net  Fri Apr 16 12:57:08 2010
From: athompso at athompso.net (Adam Thompson)
Date: Fri, 16 Apr 2010 12:57:08 -0500
Subject: [RndTbl] printing from an iPad to an HP printer
Message-ID: <001f01cadd8e$3b2efc70$b18cf550$@net>

It appears that Apple has developed, in conjunction with HP, entirely new 
printing technology for iPad applications.

	http://www.theformgroup.com/about/blog/2010/04/how-to-print-from-an-ipad

:-)

-Adam Thompson
 <athompso at athompso.net>
 (204) 291-7950




From athompso at athompso.net  Fri Apr 16 15:36:36 2010
From: athompso at athompso.net (Adam Thompson)
Date: Fri, 16 Apr 2010 15:36:36 -0500
Subject: [RndTbl] NIC performance with iperf
Message-ID: <004d01cadda4$826088a0$872199e0$@net>

FYI.  Windows networking sucks, across all versions of Windows.  (At 
least, at Gigabit speeds, it does.)

Between my Windows 7 PC
	(3.06Ghz HT Pentium 4, Intel PRO 1000/MT Desktop Adapter [82540EM])
and Gentoo Linux server
	(2 x 2.2GHz Xeon [P4 core], PCI-X [64-bit,66Mhz] Intel PRO 1000/MT 
Dual-Port Server Adapter [82546EB])

Using "iperf -s" and "iperf -c 192.168.x.x", I get - consistently - 
between 238Mbps and 250Mbps **in both directions**.

Interesting to note is that the Windows 7 PC is CPU-limited on both 
sending and receiving, despite having all the offload and acceleration 
features turned on.  Simultaneously, the Gentoo box sees a spike on one 
CPU... to 1%.

In UDP traffic, however, I see a startling asymmetry: PC-to-Linux can only 
send UDP at ~50Mbps, Linux-to-PC sends UDP at ~475Mbps.  WTF?

I also tried the Parallel option, which runs multiple streams in parallel. 
Results are consistent, aggregate bandwidth is still only about 260Mbps. 
Two streams pegs both logical (HT) CPUs on the client at ~95% utilization.

I tried enabling and disabling both the Windows Firewall and my AV 
software, which made essentially no difference at all.

Then I also tried running the same set of tests from a Windows 2003 R2 
server
	(2 x 2.13GHz Athlon MP, PCI-X [64-bit,66Mhz] Broadcom NetXtreme 
[BCM57xx])
	N.B. this is the same model of motherboard Kevin is using in his Linux 
server.
and observed slightly better throughput to the Linux server: about 
380Mbps.  Still better was Linux-to-W2K3R2, at about 440Mbps.  UDP 
performance was similar to Kevin's results, at 103Mbps W2K3-to-Linux & 
950Mbps Linux-to-W2K3.

CPU utilization on the W2K3R2 box exhibited similar properties as the Win7 
client, although none of the tests actually managed to max out either CPU, 
instead peaking at about 70% utilization.

Just for kicks, I ran a test between a RHEL5 VM (under VMWare ESXi 4.0U1, 
VMXNET adapter) and my Linux server... 920+ Mbps every time!  Of course, 
that number may be suspect since the UDP test reports speeds of 45.3Gbps 
:-).

Conclusion: the culprit is the Windows networking stack.  Note that W2K3R2 
is supposed to support zero-copy TCP offload, theoretically allowing 
network communications at wirespeed with extremely low latency.  So much 
for that idea...

-Adam Thompson
 <athompso at athompso.net>
 (204) 291-7950




From kevin.a.mcgregor at gmail.com  Sat Apr 17 07:40:57 2010
From: kevin.a.mcgregor at gmail.com (Kevin McGregor)
Date: Sat, 17 Apr 2010 07:40:57 -0500
Subject: [RndTbl] NIC performance with iperf
In-Reply-To: <004d01cadda4$826088a0$872199e0$@net>
References: <004d01cadda4$826088a0$872199e0$@net>
Message-ID: <i2g6756caf11004170540ge27f44e0kb703f0559de0049a@mail.gmail.com>

Hmm. Now I'll have to do some tests at work next week using various
combinations of OS and VMware ESX/vSphere. I'll get back to you. :)

On Fri, Apr 16, 2010 at 3:36 PM, Adam Thompson <athompso at athompso.net>wrote:

> FYI.  Windows networking sucks, across all versions of Windows.  (At
> least, at Gigabit speeds, it does.)
>
> Between my Windows 7 PC
>        (3.06Ghz HT Pentium 4, Intel PRO 1000/MT Desktop Adapter [82540EM])
> and Gentoo Linux server
>        (2 x 2.2GHz Xeon [P4 core], PCI-X [64-bit,66Mhz] Intel PRO 1000/MT
> Dual-Port Server Adapter [82546EB])
>
> Using "iperf -s" and "iperf -c 192.168.x.x", I get - consistently -
> between 238Mbps and 250Mbps **in both directions**.
>
> Interesting to note is that the Windows 7 PC is CPU-limited on both
> sending and receiving, despite having all the offload and acceleration
> features turned on.  Simultaneously, the Gentoo box sees a spike on one
> CPU... to 1%.
>
> In UDP traffic, however, I see a startling asymmetry: PC-to-Linux can only
> send UDP at ~50Mbps, Linux-to-PC sends UDP at ~475Mbps.  WTF?
>
> I also tried the Parallel option, which runs multiple streams in parallel.
> Results are consistent, aggregate bandwidth is still only about 260Mbps.
> Two streams pegs both logical (HT) CPUs on the client at ~95% utilization.
>
> I tried enabling and disabling both the Windows Firewall and my AV
> software, which made essentially no difference at all.
>
> Then I also tried running the same set of tests from a Windows 2003 R2
> server
>        (2 x 2.13GHz Athlon MP, PCI-X [64-bit,66Mhz] Broadcom NetXtreme
> [BCM57xx])
>        N.B. this is the same model of motherboard Kevin is using in his
> Linux
> server.
> and observed slightly better throughput to the Linux server: about
> 380Mbps.  Still better was Linux-to-W2K3R2, at about 440Mbps.  UDP
> performance was similar to Kevin's results, at 103Mbps W2K3-to-Linux &
> 950Mbps Linux-to-W2K3.
>
> CPU utilization on the W2K3R2 box exhibited similar properties as the Win7
> client, although none of the tests actually managed to max out either CPU,
> instead peaking at about 70% utilization.
>
> Just for kicks, I ran a test between a RHEL5 VM (under VMWare ESXi 4.0U1,
> VMXNET adapter) and my Linux server... 920+ Mbps every time!  Of course,
> that number may be suspect since the UDP test reports speeds of 45.3Gbps
> :-).
>
> Conclusion: the culprit is the Windows networking stack.  Note that W2K3R2
> is supposed to support zero-copy TCP offload, theoretically allowing
> network communications at wirespeed with extremely low latency.  So much
> for that idea...
>
> -Adam Thompson
>  <athompso at athompso.net>
>  (204) 291-7950
>
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100417/2d330a72/attachment.html 

From colin at ninja.ca  Sat Apr 17 13:44:13 2010
From: colin at ninja.ca (Colin Stanners)
Date: Sat, 17 Apr 2010 13:44:13 -0500
Subject: [RndTbl] NIC performance with iperf
In-Reply-To: <004d01cadda4$826088a0$872199e0$@net>
References: <004d01cadda4$826088a0$872199e0$@net>
Message-ID: <4BCA017D.1020203@ninja.ca>

I installed a cheap GigE switch for a customer using 2006-era P4 Dell 
desktops,
running XP with cheap Broadcom GigE NICs. Constant 600-700mbit for file 
transfers.
CPU I think was ~30%, but that may have included the AV scanning the file.

I assumed Iperf is developed/optimized natively under *NIX and the Windows
version is compiled under Cygwin, do you have any indication that it's 
optimized?

From athompso at athompso.net  Sat Apr 17 14:34:39 2010
From: athompso at athompso.net (Adam Thompson)
Date: Sat, 17 Apr 2010 14:34:39 -0500
Subject: [RndTbl] NIC performance with iperf
In-Reply-To: <4BCA017D.1020203@ninja.ca>
References: <004d01cadda4$826088a0$872199e0$@net> <4BCA017D.1020203@ninja.ca>
Message-ID: <003c01cade65$0521e110$0f65a330$@net>

> I installed a cheap GigE switch for a customer using 2006-era P4 Dell
> desktops,
> running XP with cheap Broadcom GigE NICs. Constant 600-700mbit for
> file
> transfers.
> CPU I think was ~30%, but that may have included the AV scanning the
> file.
>
> I assumed Iperf is developed/optimized natively under *NIX and the
> Windows
> version is compiled under Cygwin, do you have any indication that it's
> optimized?

I'm certain iperf is NOT optimized for Windows.  However, empirically I 
can determine using application-level tests (i.e. FTP, SAMBA, etc.) that 
the numbers I'm seeing are consistent.

Worth mentioning is that Microsoft shipped "ttcp.exe" on most versions of 
Windows XP install media, under VALUEADD\NET\something, IIRC.  Presumably 
that binary is reasonably-well optimized.  I recall using it to test 
between Windows XP and Solaris, and seeing half-decent speeds but nothing 
wonderful (~400Mbps).

Equally, I know that I too, have seen lesser Windows clients talking to 
Windows servers at higher speeds.  It's also worth noting that SMB traffic 
to my Linux box is noticeably slower than it was to the Sun Ultra 
Enterprise 450 it replaced.  While there's obviously something wrong with 
the networking stack on my Windows 7 PC (100% CPU utilization just to do 
TCP?!) it's also possible that there's some specific interaction between 
the Linux and Windows stacks.

As I demonstrated, Linux-to-linux works great.  Kevin demonstrated that 
MacOSX-Linux works great.  Many of us have anecdotal evidence that 
Windows-Windows works well, and several people have recently demonstrated 
that Windows-Linux does NOT work well.   And I've personally demonstrated 
(although I haven't posted any numbers here) that Windows-Solaris (*not* 
OpenSolaris, if it makes any difference) works better than Windows-Linux.

(OK, I engaged in a bit of Windows-bashing when I blamed the Windows 
networking stack *exclusively*.)

I don't have any desire to set up a test lab in my living room just to try 
out the various combinations... how much effect can the GigE switch have 
on performance?  I'm using dirt-cheap D-Link 5-port switches if that makes 
any difference.

-Adam Thompson
 <athompso at athompso.net>
 (204) 291-7950




From john at johnlange.ca  Sat Apr 17 22:28:22 2010
From: john at johnlange.ca (john at johnlange.ca)
Date: Sun, 18 Apr 2010 03:28:22 +0000
Subject: [RndTbl] NIC performance with iperf
In-Reply-To: <003c01cade65$0521e110$0f65a330$@net>
References: <004d01cadda4$826088a0$872199e0$@net>
	<4BCA017D.1020203@ninja.ca><003c01cade65$0521e110$0f65a330$@net>
Message-ID: <2086705970-1271561303-cardhu_decombobulator_blackberry.rim.net-1455296573-@bda005.bisx.prod.on.blackberry>

Isn't Windows still running a BSD decendent IP stack?

Just thought I'd troll a little ;)

John


Sent from my BlackBerry device on the Rogers Wireless Network

-----Original Message-----
From: "Adam Thompson" <athompso at athompso.net>
Date: Sat, 17 Apr 2010 14:34:39 
To: 'Continuation of Round Table discussion'<roundtable at muug.mb.ca>
Subject: Re: [RndTbl] NIC performance with iperf

> I installed a cheap GigE switch for a customer using 2006-era P4 Dell
> desktops,
> running XP with cheap Broadcom GigE NICs. Constant 600-700mbit for
> file
> transfers.
> CPU I think was ~30%, but that may have included the AV scanning the
> file.
>
> I assumed Iperf is developed/optimized natively under *NIX and the
> Windows
> version is compiled under Cygwin, do you have any indication that it's
> optimized?

I'm certain iperf is NOT optimized for Windows.  However, empirically I 
can determine using application-level tests (i.e. FTP, SAMBA, etc.) that 
the numbers I'm seeing are consistent.

Worth mentioning is that Microsoft shipped "ttcp.exe" on most versions of 
Windows XP install media, under VALUEADD\NET\something, IIRC.  Presumably 
that binary is reasonably-well optimized.  I recall using it to test 
between Windows XP and Solaris, and seeing half-decent speeds but nothing 
wonderful (~400Mbps).

Equally, I know that I too, have seen lesser Windows clients talking to 
Windows servers at higher speeds.  It's also worth noting that SMB traffic 
to my Linux box is noticeably slower than it was to the Sun Ultra 
Enterprise 450 it replaced.  While there's obviously something wrong with 
the networking stack on my Windows 7 PC (100% CPU utilization just to do 
TCP?!) it's also possible that there's some specific interaction between 
the Linux and Windows stacks.

As I demonstrated, Linux-to-linux works great.  Kevin demonstrated that 
MacOSX-Linux works great.  Many of us have anecdotal evidence that 
Windows-Windows works well, and several people have recently demonstrated 
that Windows-Linux does NOT work well.   And I've personally demonstrated 
(although I haven't posted any numbers here) that Windows-Solaris (*not* 
OpenSolaris, if it makes any difference) works better than Windows-Linux.

(OK, I engaged in a bit of Windows-bashing when I blamed the Windows 
networking stack *exclusively*.)

I don't have any desire to set up a test lab in my living room just to try 
out the various combinations... how much effect can the GigE switch have 
on performance?  I'm using dirt-cheap D-Link 5-port switches if that makes 
any difference.

-Adam Thompson
 <athompso at athompso.net>
 (204) 291-7950



_______________________________________________
Roundtable mailing list
Roundtable at muug.mb.ca
http://www.muug.mb.ca/mailman/listinfo/roundtable


From athompso at athompso.net  Sat Apr 17 22:32:45 2010
From: athompso at athompso.net (Adam Thompson)
Date: Sun, 18 Apr 2010 03:32:45 +0000
Subject: [RndTbl] NIC performance with iperf
Message-ID: <461724043-1271561556-cardhu_decombobulator_blackberry.rim.net-2126233633-@bda957.bisx.prod.on.blackberry>

Supposedly, although much of it got rewritten around the WS2003R2 era to accommodate full hardware TOE devices.

AFAIK, the last MS OS to run anything obviously BSDish was NT4.0.

And of course the BSD stack was never meant for 100mbps, never mind gigabit speeds.

-Adam


From gedetil at cs.umanitoba.ca  Mon Apr 19 13:34:10 2010
From: gedetil at cs.umanitoba.ca (Gilbert E. Detillieux)
Date: Mon, 19 Apr 2010 13:34:10 -0500
Subject: [RndTbl] Fwd: USENIX WebApps '10 Registration Now Open
Message-ID: <4BCCA222.50404@cs.umanitoba.ca>

FYI...

-------- Original Message --------
Subject: USENIX WebApps '10 Registration Now Open
Date: Mon, 19 Apr 2010 11:20:33 -0700
From: Lionel Garth Jones <lgj at usenix.org>
To: info at muug.mb.ca, gedetil at muug.mb.ca

You're invited to join us in Boston, June 23-24, 2010, for the first
USENIX Conference on Web Application Development. WebApps '10 is a new
technical conference designed to bring together experts in all aspects
of developing and deploying Web applications.

Web-based applications are revolutionizing both the features that can be
delivered and the technologies for developing and deploying
applications. They also involve a diverse collection of issues and
technologies. Take advantage of this new opportunity for interaction and
synergy across these areas.

The program includes:

* Invited talks by industry leaders including Adam de Boor, Google, on
"Gmail: Past, Present, and Future"

* Refereed papers on a variety of topics such as new approaches to
Web frameworks, techniques for giving users control over their Web
data, better ways of managing Ajax and Javascript, and a variety
of security issues ranging from better CAPTCHAs to information flow
tracking.

* A Poster Session and Happy Hour held jointly with USENIX ATC '10.
The poster session provides a great way to let other people know about
your work and to get useful feedback from the community. Submissions are
due on June 13, 2010.

* Work-in-Progress Reports. If you have work you would like to share, a
controversial opinion, or a cool idea that's not quite ready for
publication, submit a WiP by June 13, 2010.

Find out more about the posters and WiPs at:
http://www.usenix.org/events/webapps10/posters_wips.html

WebApps '10 is part of the USENIX Federated Conferences Week. Not only
do you get a 2-day conference program filled with the latest research,
but you'll also have additional opportunities to mingle with colleagues
and leading experts in the combined Birds-of-a-Feather sessions (BoFs)
and at the various evening social events, vendor BoFs, and receptions.

Your WebApps '10 registration gets you into all the ConfWeek events
happening those days: tutorials, talks, workshops--you name it. Plus,
registration packages offer expanded discounts. The more days you
attend, the more you save!

http://www.usenix.org/confweek10

The full WebApps '10 program is available at
http://www.usenix.org/webapps10/proga

Register by June 7, 2010, for the greatest savings.

* Additional registration discounts are available!
Discount types include hardship, government, and non-profit:
http://www.usenix.org/events/confweek10/discounts.html

* Help spread the word!
http://www.usenix.org/events/webapps10/promote.html

I look forward to seeing you in Boston!

Sincerely,

John Ousterhout, Stanford University
WebApps '10 Program Chair
webapps10chair at usenix.org

------------------------------------------------------------------------
USENIX Conference on Web Application Development (WebApps '10)
June 23-24, 2010, Boston, MA
http://www.usenix.org/webapps10/proga
http://www.usenix.org/confweek10
Poster and WiPs Submissions Deadline: May 27, 2010
Early Bird Registration Deadline: June 7, 2010
------------------------------------------------------------------------


From ummar143 at shaw.ca  Tue Apr 20 10:51:41 2010
From: ummar143 at shaw.ca (Dan Martin)
Date: Tue, 20 Apr 2010 10:51:41 -0500
Subject: [RndTbl]  trouble with Firewatir
In-Reply-To: <4BCCA222.50404@cs.umanitoba.ca>
References: <4BCCA222.50404@cs.umanitoba.ca>
Message-ID: <6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>

I am learning to navigate and scrape.  I have had more success with  
Firewatir than Ruby mechanize, but both are poorly documented.

Current problem:  logging into a site webpage, Firwatir can see the  
frames, but cannot access them.

 >> browser.show_frames
There are 3 frames
frame: name: topbar
       index: 1
frame: name: langFrame
       index: 2
frame: name: main
       index: 3
=> 0..2

 >> browser.frame(:index, 1)
=> #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>

 >> browser.frame("main")
=> #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">

Firefox shows the webpage source [I show only the top levels]  
structure as:

<html>

<head>
	<meta http-equiv="content-type" content="text/html;charset=iso-8859-1">
	<title>TITLE</title>
	<link rel="stylesheet" href="all.css">

	<script language="JavaScript" src="WEBPAGE"></script>	
	<script language="JavaScript">FUNCTIONS</script>
</head>

	<frameset rows="63,0,*" border="0" id="frameset_ID" framespacing="0"  
frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
		<frame src="/TDM/header.htm" name="topbar" scrolling="no"  
noresize>STUFF IN HERE</frame>

		<frame name="langFrame" src="lang.html">STUFF IN HERE</frame>
		<frame src="" id="main_ID" name="main" noresize>STUFF IN HERE THAT I  
WANT</frame>
	</frameset>

	<noframes>
		STUFF IN HERE
	</noframes>

</html>


I cannot access any forms etc included in the frame unless I can  
name / access the frame.
Any ideas?

Dan Martin
GP Hospital Practitioner
Computer Scientist
ummar143 at shaw.ca
(204) 831-1746
answering machine always on


From sean at ertw.com  Tue Apr 20 11:34:00 2010
From: sean at ertw.com (Sean Walberg)
Date: Tue, 20 Apr 2010 11:34:00 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
Message-ID: <i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>

Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have you tried

browser.frame("main").html

?

On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca> wrote:

> I am learning to navigate and scrape.  I have had more success with
> Firewatir than Ruby mechanize, but both are poorly documented.
>
> Current problem:  logging into a site webpage, Firwatir can see the
> frames, but cannot access them.
>
>  >> browser.show_frames
> There are 3 frames
> frame: name: topbar
>       index: 1
> frame: name: langFrame
>       index: 2
> frame: name: main
>       index: 3
> => 0..2
>
>  >> browser.frame(:index, 1)
> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>
>  >> browser.frame("main")
> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>
> Firefox shows the webpage source [I show only the top levels]
> structure as:
>
> <html>
>
> <head>
>        <meta http-equiv="content-type"
> content="text/html;charset=iso-8859-1">
>        <title>TITLE</title>
>        <link rel="stylesheet" href="all.css">
>
>        <script language="JavaScript" src="WEBPAGE"></script>
>        <script language="JavaScript">FUNCTIONS</script>
> </head>
>
>        <frameset rows="63,0,*" border="0" id="frameset_ID" framespacing="0"
> frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
>                <frame src="/TDM/header.htm" name="topbar" scrolling="no"
> noresize>STUFF IN HERE</frame>
>
>                <frame name="langFrame" src="lang.html">STUFF IN
> HERE</frame>
>                <frame src="" id="main_ID" name="main" noresize>STUFF IN
> HERE THAT I
> WANT</frame>
>        </frameset>
>
>        <noframes>
>                STUFF IN HERE
>        </noframes>
>
> </html>
>
>
> I cannot access any forms etc included in the frame unless I can
> name / access the frame.
> Any ideas?
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>



-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/8b37950c/attachment.html 

From ummar143 at shaw.ca  Tue Apr 20 12:05:48 2010
From: ummar143 at shaw.ca (Dan Martin)
Date: Tue, 20 Apr 2010 12:05:48 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
	<i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
Message-ID: <B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>

As noted in the previous email
 >> browser.frame("main")
=> #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">

"located=false" seems to be Firewatir's way of saying that the object  
doesn't exist until I ask for it.
If I try to use it anyway, I get an unknown object exception

 >> browser.frame("main").html
Watir::Exception::UnknownObjectException: Unable to locate element,  
using :name, "main"
	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
element.rb:907:in `assert_exists'
	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
elements/frame.rb:34:in `html'
	from (irb):62

On 20-Apr-10, at 11:34 AM, Sean Walberg wrote:

>
> Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have you  
> tried
>
> browser.frame("main").html
>
> ?
>
> On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca> wrote:
> I am learning to navigate and scrape.  I have had more success with
> Firewatir than Ruby mechanize, but both are poorly documented.
>
> Current problem:  logging into a site webpage, Firwatir can see the
> frames, but cannot access them.
>
>  >> browser.show_frames
> There are 3 frames
> frame: name: topbar
>       index: 1
> frame: name: langFrame
>       index: 2
> frame: name: main
>       index: 3
> => 0..2
>
>  >> browser.frame(:index, 1)
> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>
>  >> browser.frame("main")
> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>
> Firefox shows the webpage source [I show only the top levels]
> structure as:
>
> <html>
>
> <head>
>        <meta http-equiv="content-type" content="text/ 
> html;charset=iso-8859-1">
>        <title>TITLE</title>
>        <link rel="stylesheet" href="all.css">
>
>        <script language="JavaScript" src="WEBPAGE"></script>
>        <script language="JavaScript">FUNCTIONS</script>
> </head>
>
>        <frameset rows="63,0,*" border="0" id="frameset_ID"  
> framespacing="0"
> frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
>                <frame src="/TDM/header.htm" name="topbar"  
> scrolling="no"
> noresize>STUFF IN HERE</frame>
>
>                <frame name="langFrame" src="lang.html">STUFF IN  
> HERE</frame>
>                <frame src="" id="main_ID" name="main" noresize>STUFF  
> IN HERE THAT I
> WANT</frame>
>        </frameset>
>
>        <noframes>
>                STUFF IN HERE
>        </noframes>
>
> </html>
>
>
> I cannot access any forms etc included in the frame unless I can
> name / access the frame.
> Any ideas?
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
>
> -- 
> Sean Walberg <sean at ertw.com>    http://ertw.com/
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable

Dan Martin
GP Hospital Practitioner
Computer Scientist
ummar143 at shaw.ca
(204) 831-1746
answering machine always on

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/6a3e8912/attachment.html 

From sean at ertw.com  Tue Apr 20 12:35:56 2010
From: sean at ertw.com (Sean Walberg)
Date: Tue, 20 Apr 2010 12:35:56 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
	<i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
	<B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>
Message-ID: <h2vcc43a8f81004201035x4032498eif62642a0b6e0c402@mail.gmail.com>

If you RTSL a bit you'll see that calling browser.frame creates a new
element but doesn't validate it actually found anything. When you call the
html method it first asserts that the object exists and throws an exception
if not.

The located=false comes from the element#inspect method, (which just taught
me a new Ruby idiom: !!@o returns true if @o is a valid object or false if
it isn't.) So located=false means that the object never existed, nor will it
ever.

While poking around I saw a locate_frame method. Maybe that might work
better? (It's actually possible that .frame uses locate_frame, I didn't
trace it far enough)

Failing that, what about opening the frame directly?

Sean



On Tue, Apr 20, 2010 at 12:05 PM, Dan Martin <ummar143 at shaw.ca> wrote:

> As noted in the previous email
> >> browser.frame("main")
> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>
> "located=false" seems to be Firewatir's way of saying that the object
> doesn't exist until I ask for it.
> If I try to use it anyway, I get an unknown object exception
>
> >> browser.frame("main").html
> Watir::Exception::UnknownObjectException: Unable to locate element, using
> :name, "main"
> from
> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:907:in
> `assert_exists'
> from
> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/elements/frame.rb:34:in
> `html'
> from (irb):62
>
> On 20-Apr-10, at 11:34 AM, Sean Walberg wrote:
>
>
> Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have you tried
>
> browser.frame("main").html
>
> ?
>
> On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca> wrote:
>
>> I am learning to navigate and scrape.  I have had more success with
>> Firewatir than Ruby mechanize, but both are poorly documented.
>>
>> Current problem:  logging into a site webpage, Firwatir can see the
>> frames, but cannot access them.
>>
>>  >> browser.show_frames
>> There are 3 frames
>> frame: name: topbar
>>       index: 1
>> frame: name: langFrame
>>       index: 2
>> frame: name: main
>>       index: 3
>> => 0..2
>>
>>  >> browser.frame(:index, 1)
>> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>>
>>  >> browser.frame("main")
>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>
>> Firefox shows the webpage source [I show only the top levels]
>> structure as:
>>
>> <html>
>>
>> <head>
>>        <meta http-equiv="content-type"
>> content="text/html;charset=iso-8859-1">
>>        <title>TITLE</title>
>>        <link rel="stylesheet" href="all.css">
>>
>>        <script language="JavaScript" src="WEBPAGE"></script>
>>        <script language="JavaScript">FUNCTIONS</script>
>> </head>
>>
>>        <frameset rows="63,0,*" border="0" id="frameset_ID"
>> framespacing="0"
>> frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
>>                <frame src="/TDM/header.htm" name="topbar" scrolling="no"
>> noresize>STUFF IN HERE</frame>
>>
>>                <frame name="langFrame" src="lang.html">STUFF IN
>> HERE</frame>
>>                <frame src="" id="main_ID" name="main" noresize>STUFF IN
>> HERE THAT I
>> WANT</frame>
>>        </frameset>
>>
>>        <noframes>
>>                STUFF IN HERE
>>        </noframes>
>>
>> </html>
>>
>>
>> I cannot access any forms etc included in the frame unless I can
>> name / access the frame.
>> Any ideas?
>>
>> Dan Martin
>> GP Hospital Practitioner
>> Computer Scientist
>> ummar143 at shaw.ca
>> (204) 831-1746
>> answering machine always on
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>
>
>
> --
> Sean Walberg <sean at ertw.com>    http://ertw.com/
>  _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/5fa8e15b/attachment.html 

From ummar143 at shaw.ca  Tue Apr 20 13:20:07 2010
From: ummar143 at shaw.ca (Dan Martin)
Date: Tue, 20 Apr 2010 13:20:07 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <h2vcc43a8f81004201035x4032498eif62642a0b6e0c402@mail.gmail.com>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
	<i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
	<B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>
	<h2vcc43a8f81004201035x4032498eif62642a0b6e0c402@mail.gmail.com>
Message-ID: <3F8E2399-4C98-41CB-A542-01C6A1789A72@shaw.ca>

interesting - I get an "unknown frame exception" instead of an  
"unknown object exception" using locate.

 >> browser.frame("main").locate()
Watir::Exception::UnknownFrameException: Unable to locate a frame  
using name and main.
	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
elements/frame.rb:29:in `locate'
	from (irb):66

 >> browser.frame("main").locate_frame(:name, 'main')
Watir::Exception::UnknownObjectException: Unable to locate element,  
using :name, "main"
	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
element.rb:907:in `assert_exists'
	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
element.rb:1266:in `method_missing'
	from (irb):64

How would I open the frame directly?
The frame I want has an ID, even this doesn't help.

An Xpath search fails:
 >> browser.element_by_xpath('//*[@id="main_ID"]')
=> #<FireWatir::Link:0x1a7fdcc located=false how=:jssh_name what=nil>


On 20-Apr-10, at 12:35 PM, Sean Walberg wrote:

> If you RTSL a bit you'll see that calling browser.frame creates a  
> new element but doesn't validate it actually found anything. When  
> you call the html method it first asserts that the object exists and  
> throws an exception if not.
>
> The located=false comes from the element#inspect method, (which just  
> taught me a new Ruby idiom: !!@o returns true if @o is a valid  
> object or false if it isn't.) So located=false means that the object  
> never existed, nor will it ever.
>
> While poking around I saw a locate_frame method. Maybe that might  
> work better? (It's actually possible that .frame uses locate_frame,  
> I didn't trace it far enough)
>
> Failing that, what about opening the frame directly?
>
> Sean
>
>
>
> On Tue, Apr 20, 2010 at 12:05 PM, Dan Martin <ummar143 at shaw.ca> wrote:
> As noted in the previous email
> >> browser.frame("main")
> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>
> "located=false" seems to be Firewatir's way of saying that the  
> object doesn't exist until I ask for it.
> If I try to use it anyway, I get an unknown object exception
>
> >> browser.frame("main").html
> Watir::Exception::UnknownObjectException: Unable to locate element,  
> using :name, "main"
> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
> element.rb:907:in `assert_exists'
> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
> elements/frame.rb:34:in `html'
> 	from (irb):62
>
> On 20-Apr-10, at 11:34 AM, Sean Walberg wrote:
>
>>
>> Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have  
>> you tried
>>
>> browser.frame("main").html
>>
>> ?
>>
>> On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca>  
>> wrote:
>> I am learning to navigate and scrape.  I have had more success with
>> Firewatir than Ruby mechanize, but both are poorly documented.
>>
>> Current problem:  logging into a site webpage, Firwatir can see the
>> frames, but cannot access them.
>>
>>  >> browser.show_frames
>> There are 3 frames
>> frame: name: topbar
>>       index: 1
>> frame: name: langFrame
>>       index: 2
>> frame: name: main
>>       index: 3
>> => 0..2
>>
>>  >> browser.frame(:index, 1)
>> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>>
>>  >> browser.frame("main")
>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>
>> Firefox shows the webpage source [I show only the top levels]
>> structure as:
>>
>> <html>
>>
>> <head>
>>        <meta http-equiv="content-type" content="text/ 
>> html;charset=iso-8859-1">
>>        <title>TITLE</title>
>>        <link rel="stylesheet" href="all.css">
>>
>>        <script language="JavaScript" src="WEBPAGE"></script>
>>        <script language="JavaScript">FUNCTIONS</script>
>> </head>
>>
>>        <frameset rows="63,0,*" border="0" id="frameset_ID"  
>> framespacing="0"
>> frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
>>                <frame src="/TDM/header.htm" name="topbar"  
>> scrolling="no"
>> noresize>STUFF IN HERE</frame>
>>
>>                <frame name="langFrame" src="lang.html">STUFF IN  
>> HERE</frame>
>>                <frame src="" id="main_ID" name="main"  
>> noresize>STUFF IN HERE THAT I
>> WANT</frame>
>>        </frameset>
>>
>>        <noframes>
>>                STUFF IN HERE
>>        </noframes>
>>
>> </html>
>>
>>
>> I cannot access any forms etc included in the frame unless I can
>> name / access the frame.
>> Any ideas?
>>
>> Dan Martin
>> GP Hospital Practitioner
>> Computer Scientist
>> ummar143 at shaw.ca
>> (204) 831-1746
>> answering machine always on
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>>
>> -- 
>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
>
>
> -- 
> Sean Walberg <sean at ertw.com>    http://ertw.com/
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable

Dan Martin
GP Hospital Practitioner
Computer Scientist
ummar143 at shaw.ca
(204) 831-1746
answering machine always on

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/bf68f0b2/attachment-0001.html 

From sean at ertw.com  Tue Apr 20 13:26:23 2010
From: sean at ertw.com (Sean Walberg)
Date: Tue, 20 Apr 2010 13:26:23 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <3F8E2399-4C98-41CB-A542-01C6A1789A72@shaw.ca>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
	<i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
	<B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>
	<h2vcc43a8f81004201035x4032498eif62642a0b6e0c402@mail.gmail.com>
	<3F8E2399-4C98-41CB-A542-01C6A1789A72@shaw.ca>
Message-ID: <u2jcc43a8f81004201126ke148d90ar5c6fd7fa06028aed@mail.gmail.com>

I mean to grab the page the frame references, instead of having Watir figure
it out.

I'm just trying to think what I'd do with Mech, as I've never heard of Watir
until now.

Sean

On Tue, Apr 20, 2010 at 1:20 PM, Dan Martin <ummar143 at shaw.ca> wrote:

> interesting - I get an "unknown frame exception" instead of an "unknown
> object exception" using locate.
>
> >> browser.frame("main").locate()
> Watir::Exception::UnknownFrameException: Unable to locate a frame using
> name and main.
> from
> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/elements/frame.rb:29:in
> `locate'
> from (irb):66
>
> >> browser.frame("main").locate_frame(:name, 'main')
> Watir::Exception::UnknownObjectException: Unable to locate element, using
> :name, "main"
> from
> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:907:in
> `assert_exists'
> from
> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:1266:in
> `method_missing'
> from (irb):64
>
> How would I open the frame directly?
> The frame I want has an ID, even this doesn't help.
>
> An Xpath search fails:
> >> browser.element_by_xpath('//*[@id="main_ID"]')
> => #<FireWatir::Link:0x1a7fdcc located=false how=:jssh_name what=nil>
>
>
> On 20-Apr-10, at 12:35 PM, Sean Walberg wrote:
>
> If you RTSL a bit you'll see that calling browser.frame creates a new
> element but doesn't validate it actually found anything. When you call the
> html method it first asserts that the object exists and throws an exception
> if not.
>
> The located=false comes from the element#inspect method, (which just taught
> me a new Ruby idiom: !!@o returns true if @o is a valid object or false if
> it isn't.) So located=false means that the object never existed, nor will it
> ever.
>
> While poking around I saw a locate_frame method. Maybe that might work
> better? (It's actually possible that .frame uses locate_frame, I didn't
> trace it far enough)
>
> Failing that, what about opening the frame directly?
>
> Sean
>
>
>
> On Tue, Apr 20, 2010 at 12:05 PM, Dan Martin <ummar143 at shaw.ca> wrote:
>
>> As noted in the previous email
>> >> browser.frame("main")
>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>
>> "located=false" seems to be Firewatir's way of saying that the object
>> doesn't exist until I ask for it.
>> If I try to use it anyway, I get an unknown object exception
>>
>> >> browser.frame("main").html
>> Watir::Exception::UnknownObjectException: Unable to locate element, using
>> :name, "main"
>>  from
>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:907:in
>> `assert_exists'
>> from
>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/elements/frame.rb:34:in
>> `html'
>>  from (irb):62
>>
>> On 20-Apr-10, at 11:34 AM, Sean Walberg wrote:
>>
>>
>> Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have you
>> tried
>>
>> browser.frame("main").html
>>
>> ?
>>
>> On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca> wrote:
>>
>>> I am learning to navigate and scrape.  I have had more success with
>>> Firewatir than Ruby mechanize, but both are poorly documented.
>>>
>>> Current problem:  logging into a site webpage, Firwatir can see the
>>> frames, but cannot access them.
>>>
>>>  >> browser.show_frames
>>> There are 3 frames
>>> frame: name: topbar
>>>       index: 1
>>> frame: name: langFrame
>>>       index: 2
>>> frame: name: main
>>>       index: 3
>>> => 0..2
>>>
>>>  >> browser.frame(:index, 1)
>>> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>>>
>>>  >> browser.frame("main")
>>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>>
>>> Firefox shows the webpage source [I show only the top levels]
>>> structure as:
>>>
>>> <html>
>>>
>>> <head>
>>>        <meta http-equiv="content-type"
>>> content="text/html;charset=iso-8859-1">
>>>        <title>TITLE</title>
>>>        <link rel="stylesheet" href="all.css">
>>>
>>>        <script language="JavaScript" src="WEBPAGE"></script>
>>>        <script language="JavaScript">FUNCTIONS</script>
>>> </head>
>>>
>>>        <frameset rows="63,0,*" border="0" id="frameset_ID"
>>> framespacing="0"
>>> frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
>>>                <frame src="/TDM/header.htm" name="topbar" scrolling="no"
>>> noresize>STUFF IN HERE</frame>
>>>
>>>                <frame name="langFrame" src="lang.html">STUFF IN
>>> HERE</frame>
>>>                <frame src="" id="main_ID" name="main" noresize>STUFF IN
>>> HERE THAT I
>>> WANT</frame>
>>>        </frameset>
>>>
>>>        <noframes>
>>>                STUFF IN HERE
>>>        </noframes>
>>>
>>> </html>
>>>
>>>
>>> I cannot access any forms etc included in the frame unless I can
>>> name / access the frame.
>>> Any ideas?
>>>
>>> Dan Martin
>>> GP Hospital Practitioner
>>> Computer Scientist
>>> ummar143 at shaw.ca
>>> (204) 831-1746
>>> answering machine always on
>>>
>>> _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>
>>
>>
>>
>> --
>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>>  _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>>  Dan Martin
>> GP Hospital Practitioner
>> Computer Scientist
>> ummar143 at shaw.ca
>> (204) 831-1746
>> answering machine always on
>>
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>
>
> --
> Sean Walberg <sean at ertw.com>    http://ertw.com/
>  _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/3c6855b0/attachment.html 

From ummar143 at shaw.ca  Tue Apr 20 14:28:39 2010
From: ummar143 at shaw.ca (Dan Martin)
Date: Tue, 20 Apr 2010 14:28:39 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <u2jcc43a8f81004201126ke148d90ar5c6fd7fa06028aed@mail.gmail.com>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
	<i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
	<B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>
	<h2vcc43a8f81004201035x4032498eif62642a0b6e0c402@mail.gmail.com>
	<3F8E2399-4C98-41CB-A542-01C6A1789A72@shaw.ca>
	<u2jcc43a8f81004201126ke148d90ar5c6fd7fa06028aed@mail.gmail.com>
Message-ID: <E2748B8C-2218-47A0-9E29-858FE1965CBB@shaw.ca>

I used a web page reference - consisting of variables enbedded in a  
URL - to "goto" the frame directly.  I can fill the login fields and  
click the button, but then I'm sent to another web page which is  
completely blank.

I may have to skip the login altogether, and do it manually.   
Hopefully I can still navigate within the web site.

It wouldn't surprise me if the difficulty getting in is intentional -  
to discourage robots from guessing names and passwords.  But I had no  
trouble automating my access to a different financial institution's  
site.

On 20-Apr-10, at 1:26 PM, Sean Walberg wrote:

> I mean to grab the page the frame references, instead of having  
> Watir figure it out.
>
> I'm just trying to think what I'd do with Mech, as I've never heard  
> of Watir until now.
>
> Sean
>
> On Tue, Apr 20, 2010 at 1:20 PM, Dan Martin <ummar143 at shaw.ca> wrote:
> interesting - I get an "unknown frame exception" instead of an  
> "unknown object exception" using locate.
>
> >> browser.frame("main").locate()
> Watir::Exception::UnknownFrameException: Unable to locate a frame  
> using name and main.
> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
> elements/frame.rb:29:in `locate'
> 	from (irb):66
>
> >> browser.frame("main").locate_frame(:name, 'main')
> Watir::Exception::UnknownObjectException: Unable to locate element,  
> using :name, "main"
> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
> element.rb:907:in `assert_exists'
> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
> element.rb:1266:in `method_missing'
> 	from (irb):64
>
> How would I open the frame directly?
> The frame I want has an ID, even this doesn't help.
>
> An Xpath search fails:
> >> browser.element_by_xpath('//*[@id="main_ID"]')
> => #<FireWatir::Link:0x1a7fdcc located=false how=:jssh_name what=nil>
>
>
> On 20-Apr-10, at 12:35 PM, Sean Walberg wrote:
>
>> If you RTSL a bit you'll see that calling browser.frame creates a  
>> new element but doesn't validate it actually found anything. When  
>> you call the html method it first asserts that the object exists  
>> and throws an exception if not.
>>
>> The located=false comes from the element#inspect method, (which  
>> just taught me a new Ruby idiom: !!@o returns true if @o is a valid  
>> object or false if it isn't.) So located=false means that the  
>> object never existed, nor will it ever.
>>
>> While poking around I saw a locate_frame method. Maybe that might  
>> work better? (It's actually possible that .frame uses locate_frame,  
>> I didn't trace it far enough)
>>
>> Failing that, what about opening the frame directly?
>>
>> Sean
>>
>>
>>
>> On Tue, Apr 20, 2010 at 12:05 PM, Dan Martin <ummar143 at shaw.ca>  
>> wrote:
>> As noted in the previous email
>> >> browser.frame("main")
>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>
>> "located=false" seems to be Firewatir's way of saying that the  
>> object doesn't exist until I ask for it.
>> If I try to use it anyway, I get an unknown object exception
>>
>> >> browser.frame("main").html
>> Watir::Exception::UnknownObjectException: Unable to locate element,  
>> using :name, "main"
>> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
>> element.rb:907:in `assert_exists'
>> 	from /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/ 
>> elements/frame.rb:34:in `html'
>> 	from (irb):62
>>
>> On 20-Apr-10, at 11:34 AM, Sean Walberg wrote:
>>
>>>
>>> Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have  
>>> you tried
>>>
>>> browser.frame("main").html
>>>
>>> ?
>>>
>>> On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca>  
>>> wrote:
>>> I am learning to navigate and scrape.  I have had more success with
>>> Firewatir than Ruby mechanize, but both are poorly documented.
>>>
>>> Current problem:  logging into a site webpage, Firwatir can see the
>>> frames, but cannot access them.
>>>
>>>  >> browser.show_frames
>>> There are 3 frames
>>> frame: name: topbar
>>>       index: 1
>>> frame: name: langFrame
>>>       index: 2
>>> frame: name: main
>>>       index: 3
>>> => 0..2
>>>
>>>  >> browser.frame(:index, 1)
>>> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>>>
>>>  >> browser.frame("main")
>>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>>
>>> Firefox shows the webpage source [I show only the top levels]
>>> structure as:
>>>
>>> <html>
>>>
>>> <head>
>>>        <meta http-equiv="content-type" content="text/ 
>>> html;charset=iso-8859-1">
>>>        <title>TITLE</title>
>>>        <link rel="stylesheet" href="all.css">
>>>
>>>        <script language="JavaScript" src="WEBPAGE"></script>
>>>        <script language="JavaScript">FUNCTIONS</script>
>>> </head>
>>>
>>>        <frameset rows="63,0,*" border="0" id="frameset_ID"  
>>> framespacing="0"
>>> frameborder="no" onload="loadHomepage(1);"  
>>> onunload="forceLogout();">
>>>                <frame src="/TDM/header.htm" name="topbar"  
>>> scrolling="no"
>>> noresize>STUFF IN HERE</frame>
>>>
>>>                <frame name="langFrame" src="lang.html">STUFF IN  
>>> HERE</frame>
>>>                <frame src="" id="main_ID" name="main"  
>>> noresize>STUFF IN HERE THAT I
>>> WANT</frame>
>>>        </frameset>
>>>
>>>        <noframes>
>>>                STUFF IN HERE
>>>        </noframes>
>>>
>>> </html>
>>>
>>>
>>> I cannot access any forms etc included in the frame unless I can
>>> name / access the frame.
>>> Any ideas?
>>>
>>> Dan Martin
>>> GP Hospital Practitioner
>>> Computer Scientist
>>> ummar143 at shaw.ca
>>> (204) 831-1746
>>> answering machine always on
>>>
>>> _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>
>>>
>>>
>>> -- 
>>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>>> _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>> Dan Martin
>> GP Hospital Practitioner
>> Computer Scientist
>> ummar143 at shaw.ca
>> (204) 831-1746
>> answering machine always on
>>
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>>
>>
>> -- 
>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
>
>
> -- 
> Sean Walberg <sean at ertw.com>    http://ertw.com/
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable

Dan Martin
GP Hospital Practitioner
Computer Scientist
ummar143 at shaw.ca
(204) 831-1746
answering machine always on

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/66096d6d/attachment-0001.html 

From sean at ertw.com  Tue Apr 20 14:34:04 2010
From: sean at ertw.com (Sean Walberg)
Date: Tue, 20 Apr 2010 14:34:04 -0500
Subject: [RndTbl] trouble with Firewatir
In-Reply-To: <E2748B8C-2218-47A0-9E29-858FE1965CBB@shaw.ca>
References: <4BCCA222.50404@cs.umanitoba.ca>
	<6FBD6EFC-A395-4C07-86E1-BB49302702A8@shaw.ca>
	<i2jcc43a8f81004200934zc5ca266cy23fcc3bc346c7fad@mail.gmail.com>
	<B8044E82-8C45-40B1-875B-E463859B66F7@shaw.ca>
	<h2vcc43a8f81004201035x4032498eif62642a0b6e0c402@mail.gmail.com>
	<3F8E2399-4C98-41CB-A542-01C6A1789A72@shaw.ca>
	<u2jcc43a8f81004201126ke148d90ar5c6fd7fa06028aed@mail.gmail.com>
	<E2748B8C-2218-47A0-9E29-858FE1965CBB@shaw.ca>
Message-ID: <l2hcc43a8f81004201234sb8c50149y67a2847c68dc4671@mail.gmail.com>

There might be some session cookies or some javascript-fu going on that you
have to work around. If you're logging in, I'd be betting on the former.

Sean

On Tue, Apr 20, 2010 at 2:28 PM, Dan Martin <ummar143 at shaw.ca> wrote:

> I used a web page reference - consisting of variables enbedded in a URL -
> to "goto" the frame directly.  I can fill the login fields and click the
> button, but then I'm sent to another web page which is completely blank.
>
> I may have to skip the login altogether, and do it manually.  Hopefully I
> can still navigate within the web site.
>
> It wouldn't surprise me if the difficulty getting in is intentional - to
> discourage robots from guessing names and passwords.  But I had no trouble
> automating my access to a different financial institution's site.
>
> On 20-Apr-10, at 1:26 PM, Sean Walberg wrote:
>
> I mean to grab the page the frame references, instead of having Watir
> figure it out.
>
> I'm just trying to think what I'd do with Mech, as I've never heard of
> Watir until now.
>
> Sean
>
> On Tue, Apr 20, 2010 at 1:20 PM, Dan Martin <ummar143 at shaw.ca> wrote:
>
>> interesting - I get an "unknown frame exception" instead of an "unknown
>> object exception" using locate.
>>
>> >> browser.frame("main").locate()
>> Watir::Exception::UnknownFrameException: Unable to locate a frame using
>> name and main.
>> from
>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/elements/frame.rb:29:in
>> `locate'
>>  from (irb):66
>>
>> >> browser.frame("main").locate_frame(:name, 'main')
>> Watir::Exception::UnknownObjectException: Unable to locate element, using
>> :name, "main"
>>  from
>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:907:in
>> `assert_exists'
>> from
>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:1266:in
>> `method_missing'
>>  from (irb):64
>>
>> How would I open the frame directly?
>> The frame I want has an ID, even this doesn't help.
>>
>> An Xpath search fails:
>> >> browser.element_by_xpath('//*[@id="main_ID"]')
>> => #<FireWatir::Link:0x1a7fdcc located=false how=:jssh_name what=nil>
>>
>>
>> On 20-Apr-10, at 12:35 PM, Sean Walberg wrote:
>>
>> If you RTSL a bit you'll see that calling browser.frame creates a new
>> element but doesn't validate it actually found anything. When you call the
>> html method it first asserts that the object exists and throws an exception
>> if not.
>>
>> The located=false comes from the element#inspect method, (which just
>> taught me a new Ruby idiom: !!@o returns true if @o is a valid object or
>> false if it isn't.) So located=false means that the object never existed,
>> nor will it ever.
>>
>> While poking around I saw a locate_frame method. Maybe that might work
>> better? (It's actually possible that .frame uses locate_frame, I didn't
>> trace it far enough)
>>
>> Failing that, what about opening the frame directly?
>>
>> Sean
>>
>>
>>
>> On Tue, Apr 20, 2010 at 12:05 PM, Dan Martin <ummar143 at shaw.ca> wrote:
>>
>>> As noted in the previous email
>>> >> browser.frame("main")
>>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>>
>>> "located=false" seems to be Firewatir's way of saying that the object
>>> doesn't exist until I ask for it.
>>> If I try to use it anyway, I get an unknown object exception
>>>
>>> >> browser.frame("main").html
>>> Watir::Exception::UnknownObjectException: Unable to locate element, using
>>> :name, "main"
>>>  from
>>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/element.rb:907:in
>>> `assert_exists'
>>> from
>>> /Library/Ruby/Gems/1.8/gems/firewatir-1.6.5/lib/firewatir/elements/frame.rb:34:in
>>> `html'
>>>  from (irb):62
>>>
>>> On 20-Apr-10, at 11:34 AM, Sean Walberg wrote:
>>>
>>>
>>> Looking at the rdocs, http://wtr.rubyforge.org/rdoc/1.6.5/, have you
>>> tried
>>>
>>> browser.frame("main").html
>>>
>>> ?
>>>
>>> On Tue, Apr 20, 2010 at 10:51 AM, Dan Martin <ummar143 at shaw.ca> wrote:
>>>
>>>> I am learning to navigate and scrape.  I have had more success with
>>>> Firewatir than Ruby mechanize, but both are poorly documented.
>>>>
>>>> Current problem:  logging into a site webpage, Firwatir can see the
>>>> frames, but cannot access them.
>>>>
>>>>  >> browser.show_frames
>>>> There are 3 frames
>>>> frame: name: topbar
>>>>       index: 1
>>>> frame: name: langFrame
>>>>       index: 2
>>>> frame: name: main
>>>>       index: 3
>>>> => 0..2
>>>>
>>>>  >> browser.frame(:index, 1)
>>>> => #<FireWatir::Frame:0x1a9260c located=false how=:index what=1>
>>>>
>>>>  >> browser.frame("main")
>>>> => #<FireWatir::Frame:0x1a90dd4 located=false how=:name what="main">
>>>>
>>>> Firefox shows the webpage source [I show only the top levels]
>>>> structure as:
>>>>
>>>> <html>
>>>>
>>>> <head>
>>>>        <meta http-equiv="content-type"
>>>> content="text/html;charset=iso-8859-1">
>>>>        <title>TITLE</title>
>>>>        <link rel="stylesheet" href="all.css">
>>>>
>>>>        <script language="JavaScript" src="WEBPAGE"></script>
>>>>        <script language="JavaScript">FUNCTIONS</script>
>>>> </head>
>>>>
>>>>        <frameset rows="63,0,*" border="0" id="frameset_ID"
>>>> framespacing="0"
>>>> frameborder="no" onload="loadHomepage(1);" onunload="forceLogout();">
>>>>                <frame src="/TDM/header.htm" name="topbar" scrolling="no"
>>>> noresize>STUFF IN HERE</frame>
>>>>
>>>>                <frame name="langFrame" src="lang.html">STUFF IN
>>>> HERE</frame>
>>>>                <frame src="" id="main_ID" name="main" noresize>STUFF IN
>>>> HERE THAT I
>>>> WANT</frame>
>>>>        </frameset>
>>>>
>>>>        <noframes>
>>>>                STUFF IN HERE
>>>>        </noframes>
>>>>
>>>> </html>
>>>>
>>>>
>>>> I cannot access any forms etc included in the frame unless I can
>>>> name / access the frame.
>>>> Any ideas?
>>>>
>>>> Dan Martin
>>>> GP Hospital Practitioner
>>>> Computer Scientist
>>>> ummar143 at shaw.ca
>>>> (204) 831-1746
>>>> answering machine always on
>>>>
>>>> _______________________________________________
>>>> Roundtable mailing list
>>>> Roundtable at muug.mb.ca
>>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>>
>>>
>>>
>>>
>>> --
>>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>>>  _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>
>>>
>>>  Dan Martin
>>> GP Hospital Practitioner
>>> Computer Scientist
>>> ummar143 at shaw.ca
>>> (204) 831-1746
>>> answering machine always on
>>>
>>>
>>> _______________________________________________
>>> Roundtable mailing list
>>> Roundtable at muug.mb.ca
>>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>>
>>>
>>
>>
>> --
>> Sean Walberg <sean at ertw.com>    http://ertw.com/
>>  _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>>  Dan Martin
>> GP Hospital Practitioner
>> Computer Scientist
>> ummar143 at shaw.ca
>> (204) 831-1746
>> answering machine always on
>>
>>
>> _______________________________________________
>> Roundtable mailing list
>> Roundtable at muug.mb.ca
>> http://www.muug.mb.ca/mailman/listinfo/roundtable
>>
>>
>
>
> --
> Sean Walberg <sean at ertw.com>    http://ertw.com/
>  _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>
> Dan Martin
> GP Hospital Practitioner
> Computer Scientist
> ummar143 at shaw.ca
> (204) 831-1746
> answering machine always on
>
>
> _______________________________________________
> Roundtable mailing list
> Roundtable at muug.mb.ca
> http://www.muug.mb.ca/mailman/listinfo/roundtable
>
>


-- 
Sean Walberg <sean at ertw.com>    http://ertw.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100420/88d43a27/attachment.html 

From gedetil at cs.umanitoba.ca  Mon Apr 26 16:39:22 2010
From: gedetil at cs.umanitoba.ca (Gilbert E. Detillieux)
Date: Mon, 26 Apr 2010 16:39:22 -0500
Subject: [RndTbl] Fwd: Reminder: USENIX OSDI Submission Deadline is May 7
Message-ID: <4BD6080A.5000405@cs.umanitoba.ca>

FYI...

-------- Original Message --------
Subject: Reminder: USENIX OSDI Submission Deadline is May 7
Date: Mon, 26 Apr 2010 14:34:59 -0700
From: Lionel Garth Jones <lgj at usenix.org>
To: info at muug.mb.ca, gedetil at muug.mb.ca

We're writing to remind you that the submission deadline for the 9th
USENIX Symposium on Operating Systems Design and Implementation (OSDI
'10) is approaching. Please submit your work by Friday, May 7, 2010,
9:00 p.m. PDT. Please note: The page limit for submissions is 12 pages
not including references.

http://www.usenix.org/osdi10/cfpb/

OSDI brings together professionals from academic and industrial
backgrounds in what has become a premier forum for discussing the
design, implementation, and implications of systems software.

The OSDI Symposium emphasizes innovative research as well as quantified
or insightful experiences in systems design and implementation. OSDI
takes a broad view of the systems area and solicits contributions from
many fields of systems practice, including, but not limited to:
* Operating systems
* File and storage systems
* Distributed systems
* Mobile systems
* Secure systems
* Embedded systems
* Virtualization
* Networking as it relates to operating systems
* The interaction of hardware and software development

We particularly encourage contributions containing highly original
ideas, new approaches, and/or groundbreaking results.

Submissions will be judged on originality, significance, interest,
clarity, relevance, and correctness. Accepted papers will be shepherded
through an editorial review process by a member of the program
committee.

For more details on the submission process, please see the complete
Call for Papers at:
http://www.usenix.org/osdi10/cfpb/

We look forward to receiving your submissions!

Remzi Arpaci-Dusseau, University of Wisconsin, Madison
Brad Chen, Google, Inc.
OSDI '10 Program Co-Chairs
osdi10chairs at usenix.org

---------------------------------
Call for Papers
9th USENIX Symposium on Operating Systems Design and Implementation
October 4-6, 2010
Vancouver, BC, Canada
http://www.usenix.org/osdi10/cfpb/
Submissions Deadline: May 7, 2010
Sponsored by USENIX in cooperation with ACM SIGOPS
---------------------------------


From gedetil at cs.umanitoba.ca  Thu Apr 29 10:10:53 2010
From: gedetil at cs.umanitoba.ca (Gilbert E. Detillieux)
Date: Thu, 29 Apr 2010 10:10:53 -0500
Subject: [RndTbl] How-To: Grep Tricks for Linux Users | ITworld
Message-ID: <4BD9A17D.8070309@cs.umanitoba.ca>

Further to my RTFM presentation on "grep" at the April MUUG meeting, the 
following IT World article has a few interesting tips on using GNU grep:

http://www.itworld.com/it-managementstrategy/106032/how-to-grep-tricks-linux-users

Note that the article refers to grep for Linux users, but the options 
described are applicable to the GNU implementation of grep, regardless 
of underlying OS.

GNU grep also has many other potentially useful options, so be sure to 
RTFM: grep(1).

-- 
Gilbert E. Detillieux		E-mail: <gedetil at muug.mb.ca>
Manitoba UNIX User Group	Web:	http://www.muug.mb.ca/
PO Box 130 St-Boniface		Phone:  (204)474-8161
Winnipeg MB CANADA  R2H 3B4	Fax:    (204)474-7609

From drwolfe at shaw.ca  Fri Apr 30 12:09:23 2010
From: drwolfe at shaw.ca (Don R. Wolfe)
Date: Fri, 30 Apr 2010 12:09:23 -0500
Subject: [RndTbl]  Home Automation
Message-ID: <D0FC7386F58E4F9197969014E8F8EEDC@DELL>

Well we meet again, we need to talk serious motor control as princess
auto surplus is waiting. Who in Winnipeg has good knowledge of motor
control, microprocessors, bridge motor control either professionally or
a great hobbyist. Need a mentor.

 

Thanks

 

The Automator

DRWolfe

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.muug.mb.ca/pipermail/roundtable/attachments/20100430/31077acc/attachment.html 

From rbdyck2 at shaw.ca  Fri Apr 30 20:33:59 2010
From: rbdyck2 at shaw.ca (Robert Dyck)
Date: Fri, 30 Apr 2010 20:33:59 -0500
Subject: [RndTbl] Home Automation
In-Reply-To: <D0FC7386F58E4F9197969014E8F8EEDC@DELL>
Message-ID: <002201cae8ce$5fcc1e00$6400a8c0@wp.shawcable.net>

I have done a little bit. I was employed by Micropilot to build an automated
calibration device to calibrate the sensors on the autopilots they
manufacture. This included design and construction of both hardware and
software. Most of my career has been software development, secondly systems
administration, this one job was great fun because it was the only time I
got paid to design and develop hardware. It did involve designing a rack to
precisely rotate the autopilots to use gravity to calibrate their
accelerometers in 3 dimensions, and precisely rotate them to calibrate their
rate gyros. A set of pumps and pressure/vacuum reservoirs calibrated the
pressure transducers used for the pitot tube and altimeter. Quite
interesting; I used some simple easy to find parts, others were high-end
industrial components, and some parts were custom-machined to my
specifications. And I had to install, add additional equipment to, and
program an environmental chamber that used 480-volt heaters and liquid
nitrogen to cycle temperature from -55?C to +75?C. Keeping temperature
adjustments stable and using minimum liquid nitrogen were concerns. Yup, it
meant control of exhaust fans, valves, and duty cycle modulation of the
heaters. The process also required loading software onto the autopilot
microcontrollers. And I did update software onboard the autopilots
themselves for a couple tasks.

Is that good enough?

Rob Dyck

-----Original Message-----
From: roundtable-bounces at muug.mb.ca [mailto:roundtable-bounces at muug.mb.ca]
On Behalf Of Don R. Wolfe
Sent: Friday, April 30, 2010 11:09 AM
To: roundtable at muug.mb.ca
Subject: [RndTbl] Home Automation


Well we meet again, we need to talk serious motor control as princess auto
surplus is waiting. Who in Winnipeg has good knowledge of motor control,
microprocessors, bridge motor control either professionally or a great
hobbyist. Need a mentor.

Thanks

The Automator
DRWolfe