From george at ceetonetechnology.com  Fri Feb  1 11:59:42 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 01 Feb 2013 11:59:42 -0500
Subject: [nycbug-talk] NYC*BUG Upcoming
Message-ID: <510BF47E.7040604@ceetonetechnology.com>

February 7, 645 PM - Location: Suspenders

How SMPng Works and Why It Doesn't Work The Way You Think, John Baldwin

Modern x86 CPUs have hit a wall in frequency scaling and are now 
expanding sideways by adding more cores. Adding more cores does not 
magically multiply performance, however. John talks about some of the 
reasons that it doesn't.

In 2000, FreeBSD launched a project to multithread its kernel to more 
fully take advantage of modern SMP machines. This talk will give an 
overview of that project's history and continuing work on improving 
scalability.

About the speaker:
John first started using FreeBSD in 1996 and has been an active kernel 
developer since 2000. He has worked for various companies that use 
FreeBSD including The Weather Channel and Yahoo!. John lives in New 
Jersey with his wife and three kids.

****

March 6: Brett Wynkoop on Hacking the BeagleBone with FreeBSD

April 3: Brian Callahan on MIPS on OpenBSD


From mark.saad at ymail.com  Fri Feb  1 14:39:07 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Fri, 1 Feb 2013 14:39:07 -0500
Subject: [nycbug-talk] TCP Slow Start - NetBSD
In-Reply-To: <CAJU5ZBMcbY6DyHav=oged6uiCuxT4pA6bRyFdbaNMJ_187maZw@mail.gmail.com>
References: <CAMXt9NZhcZvaHYR9xF7odcOpxE73QYwJfshL=gDphK1AT6G5MA@mail.gmail.com>
 <CAJU5ZBOHmT49OXiTG9cFWu75sQqX7rRg3VHOYk2uA0AiBktF3A@mail.gmail.com>
 <CAJU5ZBMcbY6DyHav=oged6uiCuxT4pA6bRyFdbaNMJ_187maZw@mail.gmail.com>
Message-ID: <CAMXt9NYEhoXWjZXBny5iVATZGv9LvQiKzV7PWOxGSDzJcyC4dw@mail.gmail.com>

Jesse
 I want to play around with adjusting the initial congestion window frame
size defined in rfc 3390 and the the changes proposed in
http://tools.ietf.org/html/draft-ietf-tcpm-initcwnd-00

since mirrors.nycbug.org is netbsd I wanted to see what improvement
changing this made on page load time and download times.




On Thu, Jan 31, 2013 at 7:39 PM, Jesse Callaway <bonsaime at gmail.com> wrote:

> > On Jan 29, 2013 4:39 PM, "Mark Saad" <mark.saad at ymail.com> wrote:
> >>
> >> Hello all
> >>   Any one know if NetBSD allows tuning the TCP Slow start "flightsize".
> In FreeBSD there is s sysctl net.inet.tcp.slowstart_flightsize but I can
> not find any reference to this in NetBSD nor any related sysctls. Any ideas
> ?
> >> --
> >>
> >> Mark Saad | mark.saad at ymail.com
> >>
> >>
> What is fightsize and what would you want to adjust it for... like what
> conditions? If it's sensitive, im not pressing, but would love to hear.
>
> Evil person that i am has not googled before posting this message... doing
> that nowish.
>
> (adjusted post-mortem) ______________________________________________
>
> >> talk mailing list
> >> talk at lists.nycbug.org
> >> http://lists.nycbug.org/mailman/listinfo/talk
> >>
>
>


-- 

Mark Saad | mark.saad at ymail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130201/fbdf6c72/attachment.htm>

From pete at nomadlogic.org  Fri Feb  1 14:46:36 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Fri, 01 Feb 2013 11:46:36 -0800
Subject: [nycbug-talk] TCP Slow Start - NetBSD
In-Reply-To: <CAMXt9NYEhoXWjZXBny5iVATZGv9LvQiKzV7PWOxGSDzJcyC4dw@mail.gmail.com>
References: <CAMXt9NZhcZvaHYR9xF7odcOpxE73QYwJfshL=gDphK1AT6G5MA@mail.gmail.com>
 <CAJU5ZBOHmT49OXiTG9cFWu75sQqX7rRg3VHOYk2uA0AiBktF3A@mail.gmail.com>
 <CAJU5ZBMcbY6DyHav=oged6uiCuxT4pA6bRyFdbaNMJ_187maZw@mail.gmail.com>
 <CAMXt9NYEhoXWjZXBny5iVATZGv9LvQiKzV7PWOxGSDzJcyC4dw@mail.gmail.com>
Message-ID: <510C1B9C.9010806@nomadlogic.org>

On 02/01/13 11:39, Mark Saad wrote:
> Jesse
>  I want to play around with adjusting the initial congestion window 
> frame size defined in rfc 3390 and the the changes proposed in 
> http://tools.ietf.org/html/draft-ietf-tcpm-initcwnd-00
>
> since mirrors.nycbug.org <http://mirrors.nycbug.org> is netbsd I 
> wanted to see what improvement changing this made on page load time 
> and download times.
>


oh rad didn't know that mirror was running netbsd.  as an aside - which 
may be helpful for testing - i've been mucking about with netbsd-6.0.1 
amazon ec2 images.  i've been using the micro instances which are free 
for my account.  they run great in my experience!

-pete

-- 
Pete Wright
pete at nomadlogic.org
twitter =>  @nomadlogicLA


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130201/2deb31b8/attachment.htm>

From edlinuxguru at gmail.com  Sat Feb  2 12:01:17 2013
From: edlinuxguru at gmail.com (Edward Capriolo)
Date: Sat, 2 Feb 2013 12:01:17 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network card
Message-ID: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>

I purchased a new Lenovo Yoga
http://www.lenovo.com/products/us/laptop/ideapad/yoga/yoga-13/. SSD +
18GB ram. I am playing around with different installs right now,
fedora, freebsd etc. The machine is pretty new so many of the
os/distros do not have all the driver information they need. I am
willing to "hang around" and wait for touch screen support.

Right now the big blocker is it does not have a CAT5 and nothing is
picking up it's wireless card. Onto my question.

Does anyone know of an off the shelve USB wireless card that "just
works" with modern OS's. It need not be fast. It could be a A or B
speed, I do not care. Just looking for something that
kickstart/jumpstart/graphical_installers auto-detects reliably without
driver disks etc.

Thanks


From gjb at FreeBSD.org  Sat Feb  2 12:08:55 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Sat, 2 Feb 2013 12:08:55 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
 card
In-Reply-To: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
Message-ID: <20130202170855.GF1397@glenbarber.us>

On Sat, Feb 02, 2013 at 12:01:17PM -0500, Edward Capriolo wrote:
> I purchased a new Lenovo Yoga
> http://www.lenovo.com/products/us/laptop/ideapad/yoga/yoga-13/. SSD +
> 18GB ram. I am playing around with different installs right now,
> fedora, freebsd etc. The machine is pretty new so many of the
> os/distros do not have all the driver information they need. I am
> willing to "hang around" and wait for touch screen support.
> 
> Right now the big blocker is it does not have a CAT5 and nothing is
> picking up it's wireless card. Onto my question.
> 
> Does anyone know of an off the shelve USB wireless card that "just
> works" with modern OS's. It need not be fast. It could be a A or B
> speed, I do not care. Just looking for something that
> kickstart/jumpstart/graphical_installers auto-detects reliably without
> driver disks etc.
> 

I have been using a Belkin F5D7050 USB dongle (rum(4) driver on
FreeBSD).  Works well when I run FreeBSD on my MacBook Pro.  IIRC,
I purchased it at Office Max or Best Buy, or one of those types of
stores.

Glen

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 488 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130202/60ae6747/attachment.bin>

From bcallah at devio.us  Sat Feb  2 12:10:23 2013
From: bcallah at devio.us (Brian Callahan)
Date: Sat, 02 Feb 2013 12:10:23 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
 card
In-Reply-To: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
Message-ID: <510D487F.2050506@devio.us>

On 2/2/2013 12:01 PM, Edward Capriolo wrote:
> I purchased a new Lenovo Yoga
> http://www.lenovo.com/products/us/laptop/ideapad/yoga/yoga-13/. SSD +
> 18GB ram. I am playing around with different installs right now,
> fedora, freebsd etc. The machine is pretty new so many of the
> os/distros do not have all the driver information they need. I am
> willing to "hang around" and wait for touch screen support.
>
> Right now the big blocker is it does not have a CAT5 and nothing is
> picking up it's wireless card. Onto my question.
>
> Does anyone know of an off the shelve USB wireless card that "just
> works" with modern OS's. It need not be fast. It could be a A or B
> speed, I do not care. Just looking for something that
> kickstart/jumpstart/graphical_installers auto-detects reliably without
> driver disks etc.
>
> Thanks
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>

Anything based on RTL8187L/RTL8187B chipset (urtw on OpenBSD) should 
work OOTB on any BSD.

eBay turns this up:
http://www.ebay.com/itm/Alfa-500mW-USB-Adapter-Realtek-RTL8187L-AWUS036EW-5dBi-/380344755381

HTH

~Brian


From bonsaime at gmail.com  Sat Feb  2 12:28:50 2013
From: bonsaime at gmail.com (Jesse Callaway)
Date: Sat, 2 Feb 2013 12:28:50 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
	card
In-Reply-To: <CAJU5ZBOc+L--RpZk1JVgGQ6FOuWrY35Q=dB5zytCmH2+R0vQuw@mail.gmail.com>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
 <510D487F.2050506@devio.us>
 <CAJU5ZBOc+L--RpZk1JVgGQ6FOuWrY35Q=dB5zytCmH2+R0vQuw@mail.gmail.com>
Message-ID: <CAJU5ZBNMuo+v0HCRA2r06NKXOy9VgTifOCPQNFGwCKOhOxs0iw@mail.gmail.com>

On Feb 2, 2013 12:12 PM, "Brian Callahan" <bcallah at devio.us> wrote:
>
> On 2/2/2013 12:01 PM, Edward Capriolo wrote:
>>
>> I purchased a new Lenovo Yoga
>> http://www.lenovo.com/products/us/laptop/ideapad/yoga/yoga-13/. SSD +
>> 18GB ram. I am playing around with different installs right now,
>> fedora, freebsd etc. The machine is pretty new so many of the
>> os/distros do not have all the driver information they need. I am
>> willing to "hang around" and wait for touch screen support.
>>
>> Right now the big blocker is it does not have a CAT5 and nothing is
>> picking up it's wireless card. Onto my question.
>>
>> Does anyone know of an off the shelve USB wireless card that "just
>> works" with modern OS's. It need not be fast. It could be a A or B
>> speed, I do not care. Just looking for something that
>> kickstart/jumpstart/graphical_installers auto-detects reliably without
>> driver disks etc.
>>
>> Thanks
>> _______________________________________________
>> talk mailing list
>> talk at lists.nycbug.org
>> http://lists.nycbug.org/mailman/listinfo/talk
>>
>
> Anything based on RTL8187L/RTL8187B chipset (urtw on OpenBSD) should work
OOTB on any BSD.
>
> eBay turns this up:
>
http://www.ebay.com/itm/Alfa-500mW-USB-Adapter-Realtek-RTL8187L-AWUS036EW-5dBi-/380344755381
>
> HTH
>
> ~Brian
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk

Thanks to you and Glen for this. I find it frustrating to be arned with all
of this informatiin, but still not being too sure of what's inside the
plastic. Specifically, I mean you can do a couple hours of research and
it's  then still a gamble because the mfgr will just up and change the
chipset... then make sure the windows driver supports it.

Yes, this is not news to any of you, but just venting.

Unless a continuous effort is made, with lots of labor to maintain a
database then we will always have to "hunt and gamble". The hunting comes
with owning stuff, but the gamble is no fun.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130202/af34f6ea/attachment.htm>

From bcallah at devio.us  Sat Feb  2 12:56:04 2013
From: bcallah at devio.us (Brian Callahan)
Date: Sat, 02 Feb 2013 12:56:04 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
 card
In-Reply-To: <CAJU5ZBNMuo+v0HCRA2r06NKXOy9VgTifOCPQNFGwCKOhOxs0iw@mail.gmail.com>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
 <510D487F.2050506@devio.us>
 <CAJU5ZBOc+L--RpZk1JVgGQ6FOuWrY35Q=dB5zytCmH2+R0vQuw@mail.gmail.com>
 <CAJU5ZBNMuo+v0HCRA2r06NKXOy9VgTifOCPQNFGwCKOhOxs0iw@mail.gmail.com>
Message-ID: <510D5334.9070401@devio.us>

On 2/2/2013 12:28 PM, Jesse Callaway wrote:
>
> On Feb 2, 2013 12:12 PM, "Brian Callahan" <bcallah at devio.us
> <mailto:bcallah at devio.us>> wrote:
>  >
>  > On 2/2/2013 12:01 PM, Edward Capriolo wrote:
>  >>
>  >> I purchased a new Lenovo Yoga
>  >> http://www.lenovo.com/products/us/laptop/ideapad/yoga/yoga-13/. SSD +
>  >> 18GB ram. I am playing around with different installs right now,
>  >> fedora, freebsd etc. The machine is pretty new so many of the
>  >> os/distros do not have all the driver information they need. I am
>  >> willing to "hang around" and wait for touch screen support.
>  >>
>  >> Right now the big blocker is it does not have a CAT5 and nothing is
>  >> picking up it's wireless card. Onto my question.
>  >>
>  >> Does anyone know of an off the shelve USB wireless card that "just
>  >> works" with modern OS's. It need not be fast. It could be a A or B
>  >> speed, I do not care. Just looking for something that
>  >> kickstart/jumpstart/graphical_installers auto-detects reliably without
>  >> driver disks etc.
>  >>
>  >> Thanks
>  >> _______________________________________________
>  >> talk mailing list
>  >> talk at lists.nycbug.org <mailto:talk at lists.nycbug.org>
>  >> http://lists.nycbug.org/mailman/listinfo/talk
>  >>
>  >
>  > Anything based on RTL8187L/RTL8187B chipset (urtw on OpenBSD) should
> work OOTB on any BSD.
>  >
>  > eBay turns this up:
>  >
> http://www.ebay.com/itm/Alfa-500mW-USB-Adapter-Realtek-RTL8187L-AWUS036EW-5dBi-/380344755381
>  >
>  > HTH
>  >
>  > ~Brian
>  >
>  > _______________________________________________
>  > talk mailing list
>  > talk at lists.nycbug.org <mailto:talk at lists.nycbug.org>
>  > http://lists.nycbug.org/mailman/listinfo/talk
>
> Thanks to you and Glen for this. I find it frustrating to be arned with
> all of this informatiin, but still not being too sure of what's inside
> the plastic. Specifically, I mean you can do a couple hours of research
> and it's  then still a gamble because the mfgr will just up and change
> the chipset... then make sure the windows driver supports it.
>
> Yes, this is not news to any of you, but just venting.
>
> Unless a continuous effort is made, with lots of labor to maintain a
> database then we will always have to "hunt and gamble". The hunting
> comes with owning stuff, but the gamble is no fun.
>

I've found the Realtek stuff never changes the chipset on a product. A 
different chipset will have a different product name/number. Plus they 
tend to advertise the chipset openly and directly. Much different from 
some of the big-name vendors, indeed!

The manpages usually have a list of hardware known to work with each 
driver. Admittedly, it's sometimes non-exhaustive.

Of course, if you have something you know has a good chipset and it 
comes up unidentified, write up a patch or let a dev know 
(me/okan/others? for OpenBSD, I'm sure there are devs for other BSDs 
lurking here) so we can write up a patch.

I wonder if there's a way we can take the dmesgd entries, pull out the 
networking info, and keep that in its own separate database? Just a 
thought :)

~Brian


From george at ceetonetechnology.com  Sat Feb  2 13:42:52 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sat, 02 Feb 2013 13:42:52 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
 card
In-Reply-To: <510D5334.9070401@devio.us>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
 <510D487F.2050506@devio.us>
 <CAJU5ZBOc+L--RpZk1JVgGQ6FOuWrY35Q=dB5zytCmH2+R0vQuw@mail.gmail.com>
 <CAJU5ZBNMuo+v0HCRA2r06NKXOy9VgTifOCPQNFGwCKOhOxs0iw@mail.gmail.com>
 <510D5334.9070401@devio.us>
Message-ID: <510D5E2C.2020101@ceetonetechnology.com>

On 02/02/13 12:56, Brian Callahan wrote:
> On 2/2/2013 12:28 PM, Jesse Callaway wrote:
>>
>> On Feb 2, 2013 12:12 PM, "Brian Callahan" <bcallah at devio.us
>> <mailto:bcallah at devio.us>> wrote:
>>  >
>>  > On 2/2/2013 12:01 PM, Edward Capriolo wrote:
>>  >>
>>  >> I purchased a new Lenovo Yoga
>>  >> http://www.lenovo.com/products/us/laptop/ideapad/yoga/yoga-13/. SSD +
>>  >> 18GB ram. I am playing around with different installs right now,
>>  >> fedora, freebsd etc. The machine is pretty new so many of the
>>  >> os/distros do not have all the driver information they need. I am
>>  >> willing to "hang around" and wait for touch screen support.
>>  >>
>>  >> Right now the big blocker is it does not have a CAT5 and nothing is
>>  >> picking up it's wireless card. Onto my question.
>>  >>
>>  >> Does anyone know of an off the shelve USB wireless card that "just
>>  >> works" with modern OS's. It need not be fast. It could be a A or B
>>  >> speed, I do not care. Just looking for something that
>>  >> kickstart/jumpstart/graphical_installers auto-detects reliably
>> without
>>  >> driver disks etc.
>>  >>
>>  >> Thanks
>>  >> _______________________________________________
>>  >> talk mailing list
>>  >> talk at lists.nycbug.org <mailto:talk at lists.nycbug.org>
>>  >> http://lists.nycbug.org/mailman/listinfo/talk
>>  >>
>>  >
>>  > Anything based on RTL8187L/RTL8187B chipset (urtw on OpenBSD) should
>> work OOTB on any BSD.
>>  >
>>  > eBay turns this up:
>>  >
>> http://www.ebay.com/itm/Alfa-500mW-USB-Adapter-Realtek-RTL8187L-AWUS036EW-5dBi-/380344755381
>>
>>  >
>>  > HTH
>>  >
>>  > ~Brian
>>  >
>>  > _______________________________________________
>>  > talk mailing list
>>  > talk at lists.nycbug.org <mailto:talk at lists.nycbug.org>
>>  > http://lists.nycbug.org/mailman/listinfo/talk
>>
>> Thanks to you and Glen for this. I find it frustrating to be arned with
>> all of this informatiin, but still not being too sure of what's inside
>> the plastic. Specifically, I mean you can do a couple hours of research
>> and it's  then still a gamble because the mfgr will just up and change
>> the chipset... then make sure the windows driver supports it.
>>
>> Yes, this is not news to any of you, but just venting.
>>
>> Unless a continuous effort is made, with lots of labor to maintain a
>> database then we will always have to "hunt and gamble". The hunting
>> comes with owning stuff, but the gamble is no fun.
>>
>
> I've found the Realtek stuff never changes the chipset on a product. A
> different chipset will have a different product name/number. Plus they
> tend to advertise the chipset openly and directly. Much different from
> some of the big-name vendors, indeed!
>

The nice thing is that with the declining margins/cheaper production, a 
la carte components aren't the monopoly of the big manufacturers that 
you have to wait on them for "niceties" like documentation. 
AFAIK/remember, it was the Taiwanese manufacturers who the OpenBSD 
people had an easier time accessing, not the bcms of the world.

> The manpages usually have a list of hardware known to work with each
> driver. Admittedly, it's sometimes non-exhaustive.

Which

>
> Of course, if you have something you know has a good chipset and it
> comes up unidentified, write up a patch or let a dev know
> (me/okan/others? for OpenBSD, I'm sure there are devs for other BSDs
> lurking here) so we can write up a patch.
>
> I wonder if there's a way we can take the dmesgd entries, pull out the
> networking info, and keep that in its own separate database? Just a
> thought :)

You mean dynamically?  Like queries across all the files based on 
categories?  hmmm.. . . good idea. the problem is there a way to 
indicate network stuff across all the dmesg types?

OTOH, I'm sure we could just do a dump of the contents if you wanted...

g



From bcallah at devio.us  Sat Feb  2 13:53:42 2013
From: bcallah at devio.us (Brian Callahan)
Date: Sat, 02 Feb 2013 13:53:42 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
 card
In-Reply-To: <510D5E2C.2020101@ceetonetechnology.com>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
 <510D487F.2050506@devio.us>
 <CAJU5ZBOc+L--RpZk1JVgGQ6FOuWrY35Q=dB5zytCmH2+R0vQuw@mail.gmail.com>
 <CAJU5ZBNMuo+v0HCRA2r06NKXOy9VgTifOCPQNFGwCKOhOxs0iw@mail.gmail.com>
 <510D5334.9070401@devio.us> <510D5E2C.2020101@ceetonetechnology.com>
Message-ID: <510D60B6.5060308@devio.us>

On 2/2/2013 1:42 PM, George Rosamond wrote:
*snip*
>>
>> I've found the Realtek stuff never changes the chipset on a product. A
>> different chipset will have a different product name/number. Plus they
>> tend to advertise the chipset openly and directly. Much different from
>> some of the big-name vendors, indeed!
>>
>
> The nice thing is that with the declining margins/cheaper production, a
> la carte components aren't the monopoly of the big manufacturers that
> you have to wait on them for "niceties" like documentation.
> AFAIK/remember, it was the Taiwanese manufacturers who the OpenBSD
> people had an easier time accessing, not the bcms of the world.
>

Yup. For this reason, I make sure to buy Realtek wireless stuff. As an 
added benefit, urtw(4) doesn't require any firmware, so it's perfectly 
usable in bsd.rd (in fact, I use wireless all the time to install new 
snapshots to my yeeloongs).

>> The manpages usually have a list of hardware known to work with each
>> driver. Admittedly, it's sometimes non-exhaustive.
>
> Which
>

ath(4) comes to mind...

>>
>> Of course, if you have something you know has a good chipset and it
>> comes up unidentified, write up a patch or let a dev know
>> (me/okan/others? for OpenBSD, I'm sure there are devs for other BSDs
>> lurking here) so we can write up a patch.
>>
>> I wonder if there's a way we can take the dmesgd entries, pull out the
>> networking info, and keep that in its own separate database? Just a
>> thought :)
>
> You mean dynamically?  Like queries across all the files based on
> categories?  hmmm.. . . good idea. the problem is there a way to
> indicate network stuff across all the dmesg types?
>
> OTOH, I'm sure we could just do a dump of the contents if you wanted...
>

I do mean dynamically. I'm sure there are plenty of people in the 
"hoping the right chip is inside the plastic" camp as Jesse pointed out.

Even a dump would be useful for people looking to buy wireless cards.

~Brian

> g
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk



From george at ceetonetechnology.com  Sat Feb  2 22:55:34 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sat, 02 Feb 2013 22:55:34 -0500
Subject: [nycbug-talk] NYC*BUG Upcoming
In-Reply-To: <510BF47E.7040604@ceetonetechnology.com>
References: <510BF47E.7040604@ceetonetechnology.com>
Message-ID: <510DDFB6.6040106@ceetonetechnology.com>

On 02/01/13 11:59, George Rosamond wrote:
> February 7, 645 PM - Location: Suspenders
> 
> How SMPng Works and Why It Doesn't Work The Way You Think, John Baldwin
> 

Correction on this: it's February 6th.

Thanks Spork!

Okan... note www

g



From mike at myownsoho.net  Sun Feb  3 12:03:37 2013
From: mike at myownsoho.net (mike at myownsoho.net)
Date: Sun, 03 Feb 2013 12:03:37 -0500
Subject: [nycbug-talk] Searching for a "Plug and play" wireless network
	card
In-Reply-To: <510D60B6.5060308@devio.us>
References: <CAENxBwxgd7gStq+rHrCQAFbbVNp6t6u1kx45pCBRZJLTq7N=Jw@mail.gmail.com>
 <510D487F.2050506@devio.us>
 <CAJU5ZBOc+L--RpZk1JVgGQ6FOuWrY35Q=dB5zytCmH2+R0vQuw@mail.gmail.com>
 <CAJU5ZBNMuo+v0HCRA2r06NKXOy9VgTifOCPQNFGwCKOhOxs0iw@mail.gmail.com>
 <510D5334.9070401@devio.us> <510D5E2C.2020101@ceetonetechnology.com>
 <510D60B6.5060308@devio.us>
Message-ID: <d6a38e44-fb71-48b1-831b-574fdfdffab7@email.android.com>

My favorite move is to get a mini-pci to pci adapter and reuse cards from dead laptops.
I see the chip, most are BRCM or Atheros, then plug it all together. Cheap move IMO.

Mike--

Brian Callahan <bcallah at devio.us> wrote:

>On 2/2/2013 1:42 PM, George Rosamond wrote:
>*snip*
>>>
>>> I've found the Realtek stuff never changes the chipset on a product.
>A
>>> different chipset will have a different product name/number. Plus
>they
>>> tend to advertise the chipset openly and directly. Much different
>from
>>> some of the big-name vendors, indeed!
>>>
>>
>> The nice thing is that with the declining margins/cheaper production,
>a
>> la carte components aren't the monopoly of the big manufacturers that
>> you have to wait on them for "niceties" like documentation.
>> AFAIK/remember, it was the Taiwanese manufacturers who the OpenBSD
>> people had an easier time accessing, not the bcms of the world.
>>
>
>Yup. For this reason, I make sure to buy Realtek wireless stuff. As an 
>added benefit, urtw(4) doesn't require any firmware, so it's perfectly 
>usable in bsd.rd (in fact, I use wireless all the time to install new 
>snapshots to my yeeloongs).
>
>>> The manpages usually have a list of hardware known to work with each
>>> driver. Admittedly, it's sometimes non-exhaustive.
>>
>> Which
>>
>
>ath(4) comes to mind...
>
>>>
>>> Of course, if you have something you know has a good chipset and it
>>> comes up unidentified, write up a patch or let a dev know
>>> (me/okan/others? for OpenBSD, I'm sure there are devs for other BSDs
>>> lurking here) so we can write up a patch.
>>>
>>> I wonder if there's a way we can take the dmesgd entries, pull out
>the
>>> networking info, and keep that in its own separate database? Just a
>>> thought :)
>>
>> You mean dynamically?  Like queries across all the files based on
>> categories?  hmmm.. . . good idea. the problem is there a way to
>> indicate network stuff across all the dmesg types?
>>
>> OTOH, I'm sure we could just do a dump of the contents if you
>wanted...
>>
>
>I do mean dynamically. I'm sure there are plenty of people in the 
>"hoping the right chip is inside the plastic" camp as Jesse pointed
>out.
>
>Even a dump would be useful for people looking to buy wireless cards.
>
>~Brian
>
>> g
>>
>> _______________________________________________
>> talk mailing list
>> talk at lists.nycbug.org
>> http://lists.nycbug.org/mailman/listinfo/talk
>
>_______________________________________________
>talk mailing list
>talk at lists.nycbug.org
>http://lists.nycbug.org/mailman/listinfo/talk

-- 
Sent from my Android phone with K-9 Mail. Please excuse my brevity.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130203/c41093b2/attachment.htm>

From mcevoy.pat at gmail.com  Mon Feb  4 14:08:13 2013
From: mcevoy.pat at gmail.com (Pat McEvoy)
Date: Mon, 4 Feb 2013 14:08:13 -0500
Subject: [nycbug-talk] What's new...the video
Message-ID: <CABPwdMUrP2DsBd757AKkpNoh0TKHr4rdmg0u1uT1260GVNt2eQ@mail.gmail.com>

Hey Folks,
Just put the finishing touches on Eitan Adler's "What's New With FreeBSD"
talk from the Jan NYCBUG meeting.
Many thanks to Nikolai for the audio and Eitan for the slides. Enjoy
P


http://youtu.be/NPD75P-TTno
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130204/666cb9c7/attachment.htm>

From george at ceetonetechnology.com  Wed Feb  6 10:17:40 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Wed, 06 Feb 2013 10:17:40 -0500
Subject: [nycbug-talk] NYC*BUG Tonight
Message-ID: <51127414.2010108@ceetonetechnology.com>

February 6, 645 PM - Location: Suspenders

(note date correction.  thanks Spork!)
(cross-posting to announce@ and talk@ as a temporary measure only)

->How SMPng Works and Why It Doesn't Work The Way You Think, John Baldwin

Modern x86 CPUs have hit a wall in frequency scaling and are now 
expanding sideways by adding more cores. Adding more cores does not 
magically multiply performance, however. John talks about some of the 
reasons that it doesn't.

In 2000, FreeBSD launched a project to multithread its kernel to more 
fully take advantage of modern SMP machines. This talk will give an 
overview of that project's history and continuing work on improving 
scalability.

About the speaker:

John first started using FreeBSD in 1996 and has been an active kernel 
developer since 2000. He has worked for various companies that use 
FreeBSD including The Weather Channel and Yahoo!. John lives in New 
Jersey with his wife and three kids.

****

March 6: Brett Wynkoop on Hacking the BeagleBone with FreeBSD

April 3: Brian Callahan on MIPS on OpenBSD


From nycbug at wynn.com  Sun Feb 10 22:44:21 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Sun, 10 Feb 2013 22:44:21 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
	BeagleBone
Message-ID: <20130210224421.037a854e@ivory.local>

Greeting-

Those of you that were at the meeting know I mentioned a possibility of
taking up a collection to post a bounty for the completion of USB
support for the BeagleBone.  I have 3 people in for $100 each.  

Please let me know how much you would be interested in contributing to
this goal.  Once I have a total I will approach a couple of people who
have already done some USB arm work.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

April 19, 1775 An English attempt to confiscate guns from Americans
triggered a successful revolution......

     Dear Congress, that's a hint.


From george at ceetonetechnology.com  Mon Feb 11 10:18:03 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Feb 2013 10:18:03 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
	BeagleBone
In-Reply-To: <20130210224421.037a854e@ivory.local>
References: <20130210224421.037a854e@ivory.local>
Message-ID: <51190BAB.3040809@ceetonetechnology.com>

On 02/10/13 22:44, Brett Wynkoop wrote:
> Greeting-
>
> Those of you that were at the meeting know I mentioned a possibility of
> taking up a collection to post a bounty for the completion of USB
> support for the BeagleBone.  I have 3 people in for $100 each.
>
> Please let me know how much you would be interested in contributing to
> this goal.  Once I have a total I will approach a couple of people who
> have already done some USB arm work.

I'm all for this and certainly could use USB support, but I think it's 
probably a good idea to connect with a dev who'd be interested in doing 
the actual work.  Then if necessary, we could provide contact info to 
the engineer at CircuitCo.

A bunch of us (including Brett) are involved in the FBSD-arm list 
already, so it might be a good idea to find out the route to getting it 
done, and who might be interested...

I know how a 'bounty' works, but in this context, getting a sense of the 
path might be a useful first step.

And if $100 is the perceived bar of entry, then it's not exactly 
conducive to an array of people contributing...

my $0.02.

g



From nycbug at wynn.com  Mon Feb 11 11:31:13 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Mon, 11 Feb 2013 11:31:13 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
 BeagleBone
In-Reply-To: <51190BAB.3040809@ceetonetechnology.com>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
Message-ID: <20130211113113.57fcfc7f@ivory.local>

On Mon, 11 Feb 2013 10:18:03 -0500
George Rosamond <george at ceetonetechnology.com> wrote:

> I'm all for this and certainly could use USB support, but I think
> it's probably a good idea to connect with a dev who'd be interested
> in doing the actual work.  Then if necessary, we could provide
> contact info to the engineer at CircuitCo.
> 
> A bunch of us (including Brett) are involved in the FBSD-arm list 
> already, so it might be a good idea to find out the route to getting
> it done, and who might be interested...
> 
> I know how a 'bounty' works, but in this context, getting a sense of
> the path might be a useful first step.
> 
> And if $100 is the perceived bar of entry, then it's not exactly 
> conducive to an array of people contributing...
> 
> my $0.02.
>

Greeting-

I think there is a small disconnect in your understanding, and it is
probably due to my not explaining things well enough.

I have already polled some devs.  There is one that is a good choice,
but at the moment he is playing with other things and may or may not
get back to usb.  Another of the devs suggested him and suggested
offering him an incentive to do the rest of the USB work.

I said that at this point we had 3 folks in for $100 each.  

   3 * 100 = 300

I would like to get pledges of more than that before I approach the dev
in question.  I will also contact the FreeBSD Foundation so we can make
our contributions there and they pay the dev so we can all get a tax
deduction.

I would love to hear your thoughts given my fuller explanation.

-Brett
-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

Gun Control: The theory that a woman found dead in an alley, raped and
strangled with her own pantyhose, is somehow morally superior to a
woman explaining to police how her attacker got that fatal bullet wound



From george at ceetonetechnology.com  Mon Feb 11 11:47:13 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Feb 2013 11:47:13 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
	BeagleBone
In-Reply-To: <20130211113113.57fcfc7f@ivory.local>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
 <20130211113113.57fcfc7f@ivory.local>
Message-ID: <51192091.7050300@ceetonetechnology.com>

On 02/11/13 11:31, Brett Wynkoop wrote:
> On Mon, 11 Feb 2013 10:18:03 -0500
> George Rosamond <george at ceetonetechnology.com> wrote:
>
>> I'm all for this and certainly could use USB support, but I think
>> it's probably a good idea to connect with a dev who'd be interested
>> in doing the actual work.  Then if necessary, we could provide
>> contact info to the engineer at CircuitCo.
>>
>> A bunch of us (including Brett) are involved in the FBSD-arm list
>> already, so it might be a good idea to find out the route to getting
>> it done, and who might be interested...
>>
>> I know how a 'bounty' works, but in this context, getting a sense of
>> the path might be a useful first step.
>>
>> And if $100 is the perceived bar of entry, then it's not exactly
>> conducive to an array of people contributing...
>>
>> my $0.02.
>>
>
> Greeting-
>
> I think there is a small disconnect in your understanding, and it is
> probably due to my not explaining things well enough.
>
> I have already polled some devs.  There is one that is a good choice,
> but at the moment he is playing with other things and may or may not
> get back to usb.  Another of the devs suggested him and suggested
> offering him an incentive to do the rest of the USB work.
>

Cool.  Good stuff.

> I said that at this point we had 3 folks in for $100 each.
>
>     3 * 100 = 300
>
> I would like to get pledges of more than that before I approach the dev
> in question.  I will also contact the FreeBSD Foundation so we can make
> our contributions there and they pay the dev so we can all get a tax
> deduction.

Agree.  All for self-initiative, and I am personally 100% behind you on 
this.  Don't get me wrong.

We should consider adding this to the next announce@ email.

>
> I would love to hear your thoughts given my fuller explanation.

It is a good idea (but not a requirement IMHO) to have the Foundation 
deal directly with it.

Maybe they could setup a specific method of donations directly for this 
to broaden the appeal?

(i'd cc Dru and GNN, but I know they're on this list)

There is a bunch of ARM board work (including new support for 
VersatilePB) going on right now.

My overall feeling (representing me and only me) is that this is great. 
  You are to be applauded Brett.  We need more of this in NYC*BUG, not 
less.  May you set a tone for others around our scene.

Side note: I wonder how applicable the BBone USB work is to other ARM 
boards.

g


From george at ceetonetechnology.com  Mon Feb 11 12:58:49 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Feb 2013 12:58:49 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
	BeagleBone
In-Reply-To: <20130211125430.75396821@ivory.local>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
 <20130211113113.57fcfc7f@ivory.local>
 <51192091.7050300@ceetonetechnology.com>
 <20130211125430.75396821@ivory.local>
Message-ID: <51193159.8040005@ceetonetechnology.com>

On 02/11/13 12:54, Brett Wynkoop wrote:
> On Mon, 11 Feb 2013 11:47:13 -0500
> George Rosamond <george at ceetonetechnology.com> wrote:
>
>> Side note: I wonder how applicable the BBone USB work is to other ARM
>> boards.
>>
>> g
>
> Greeting-
>
> A couple of the devs told me just about every ARM board does USB a
> little different.  I do not understand it, but then I am not a kernel
> hacker.

Santa Ike?

Had to start ripping SD cards out of old devices... time to get some new 
ones.

>
> I am, thanks to Ike passing me a Pi, starting to build up ARM
> packages.  I plan to have a web server running with the packages on it
> sometime in the next week or two.

Which www server are you going to use?  thttpd?  Hiawatha as in 
wwww.freebsdarm.org?

g




From nycbug at wynn.com  Mon Feb 11 14:23:22 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Mon, 11 Feb 2013 14:23:22 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
 BeagleBone
In-Reply-To: <51192091.7050300@ceetonetechnology.com>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
 <20130211113113.57fcfc7f@ivory.local>
 <51192091.7050300@ceetonetechnology.com>
Message-ID: <20130211142322.041d0afa@ivory.local>

On Mon, 11 Feb 2013 11:47:13 -0500
George Rosamond <george at ceetonetechnology.com> wrote:

> Side note: I wonder how applicable the BBone USB work is to other ARM 
> boards.
> 
> g

Greeting-

A couple of the devs told me just about every ARM board does USB a
little different.  I do not understand it, but then I am not a kernel
hacker.

I am, thanks to Ike passing me a Pi, starting to build up ARM
packages.  I plan to have a web server running with the packages on it
sometime in the next week or two.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

I would never invade the United States.  There would be a gun behind
every blade of grass.  --Isoroku Yamamoto



From lists at eitanadler.com  Mon Feb 11 14:30:37 2013
From: lists at eitanadler.com (Eitan Adler)
Date: Mon, 11 Feb 2013 14:30:37 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
	BeagleBone
In-Reply-To: <20130211113113.57fcfc7f@ivory.local>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
 <20130211113113.57fcfc7f@ivory.local>
Message-ID: <CAF6rxgnufWVW9365R6Oco8H4x378uVRvryX3-poZOic9EOsG1A@mail.gmail.com>

On 11 February 2013 11:31, Brett Wynkoop <nycbug at wynn.com> wrote:
> On Mon, 11 Feb 2013 10:18:03 -0500
> George Rosamond <george at ceetonetechnology.com> wrote:

>
> I would like to get pledges of more than that before I approach the dev
> in question.  I will also contact the FreeBSD Foundation so we can make
> our contributions there and they pay the dev so we can all get a tax
> deduction.

They *will not* accept this for legal reasons.  You can provide the
project you are paying the foundation for.  What *can* happen is the
dev submits a proposal for funding and you contribute to the general
fun of foundation monies.

-- 
Eitan Adler


From george at ceetonetechnology.com  Mon Feb 11 14:35:05 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Feb 2013 14:35:05 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
	BeagleBone
In-Reply-To: <CAF6rxgnufWVW9365R6Oco8H4x378uVRvryX3-poZOic9EOsG1A@mail.gmail.com>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
 <20130211113113.57fcfc7f@ivory.local>
 <CAF6rxgnufWVW9365R6Oco8H4x378uVRvryX3-poZOic9EOsG1A@mail.gmail.com>
Message-ID: <511947E9.6070200@ceetonetechnology.com>

On 02/11/13 14:30, Eitan Adler wrote:
> On 11 February 2013 11:31, Brett Wynkoop <nycbug at wynn.com> wrote:
>> On Mon, 11 Feb 2013 10:18:03 -0500
>> George Rosamond <george at ceetonetechnology.com> wrote:
>
>>
>> I would like to get pledges of more than that before I approach the dev
>> in question.  I will also contact the FreeBSD Foundation so we can make
>> our contributions there and they pay the dev so we can all get a tax
>> deduction.
>
> They *will not* accept this for legal reasons.  You can provide the
> project you are paying the foundation for.  What *can* happen is the
> dev submits a proposal for funding and you contribute to the general
> fun of foundation monies.
>


Oh, yeah, right, I forget about those rule things and all.

That's the way Eitan!

Or we could just informally in NYC.  We have no structure.

g


From nycbug at wynn.com  Mon Feb 11 14:51:17 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Mon, 11 Feb 2013 14:51:17 -0500
Subject: [nycbug-talk] How many NYCBUG folks want USB working on the
 BeagleBone
In-Reply-To: <51193159.8040005@ceetonetechnology.com>
References: <20130210224421.037a854e@ivory.local>
 <51190BAB.3040809@ceetonetechnology.com>
 <20130211113113.57fcfc7f@ivory.local>
 <51192091.7050300@ceetonetechnology.com>
 <20130211125430.75396821@ivory.local>
 <51193159.8040005@ceetonetechnology.com>
Message-ID: <20130211145117.737234bd@ivory.local>

On Mon, 11 Feb 2013 12:58:49 -0500
George Rosamond <george at ceetonetechnology.com> wrote:

> >
> > I am, thanks to Ike passing me a Pi, starting to build up ARM
> > packages.  I plan to have a web server running with the packages on
> > it sometime in the next week or two.
> 
> Which www server are you going to use?  thttpd?  Hiawatha as in 
> wwww.freebsdarm.org?
> 

Greeting-

I am actually thinking of using sws (http://prd4.wynn.com:8080/) right
on the Pi. I would like to get sws into ports, but I am not sure how to
go about that process and I suck at writing makefiles.

Right now I am in the middle of a buildworld/buildkernel cycle on the
Pi because the image I grabbed to bootstrap myself was missing some
things like named and man pages and some other things I am sure I do
not recall right now.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

April 19, 1775 An English attempt to confiscate guns from Americans
triggered a successful revolution......

     Dear Congress, that's a hint.


From nycbug at wynn.com  Wed Feb 13 04:14:30 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Wed, 13 Feb 2013 04:14:30 -0500
Subject: [nycbug-talk] BIND 10 BETA
Message-ID: <20130213041430.6495138a@ivory.wynn.com>

Greeting-

I mentioned this at the last two meetings, but it can not hurt to
repeat it here. I strongly suggest that folks that use BIND for name
service visit www.isc.org and grab the most recent BIND 10 beta to try
and pass feedback to the devs at isc.org.

Trust me if you use bind you want to do this and put it on a test box
right away to start giving feedback to the devs.  This is your
chance to influence the BIND 10 release with your feedback.  

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435


From george at ceetonetechnology.com  Wed Feb 13 14:51:41 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Wed, 13 Feb 2013 14:51:41 -0500
Subject: [nycbug-talk] pkgng question
Message-ID: <511BEECD.5050809@ceetonetechnology.com>

I can't seem to find it anywhere... so I thought I'd ping here.

In old pkg days, you just use/used:

# pkg_create -Rxjb <installed-port>

to create a package *with dependencies*

How do you do this in pkgng?  I'd rather not create a package for each...

Doesn't seem to be in man page or online...

TIA

g

PS I'm seriously thinking about creating a fat repo of arm pkgng 
packages on our mirrors. site.


From gjb at FreeBSD.org  Wed Feb 13 15:01:49 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Wed, 13 Feb 2013 15:01:49 -0500
Subject: [nycbug-talk] pkgng question
In-Reply-To: <511BEECD.5050809@ceetonetechnology.com>
References: <511BEECD.5050809@ceetonetechnology.com>
Message-ID: <20130213200149.GB1481@glenbarber.us>

On Wed, Feb 13, 2013 at 02:51:41PM -0500, George Rosamond wrote:
> I can't seem to find it anywhere... so I thought I'd ping here.
> 
> In old pkg days, you just use/used:
> 
> # pkg_create -Rxjb <installed-port>
> 
> to create a package *with dependencies*
> 
> How do you do this in pkgng?  I'd rather not create a package for each...
> 
> Doesn't seem to be in man page or online...
> 

I use pkg-info to generate the dependency list, then do something like
this:

    # pkg create -o /usr/ports/packages/All `pkg info -qr foo`

Glen

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 488 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130213/51b66470/attachment.bin>

From nycbug at wynn.com  Wed Feb 13 15:02:54 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Wed, 13 Feb 2013 15:02:54 -0500
Subject: [nycbug-talk] pkgng question
In-Reply-To: <511BEECD.5050809@ceetonetechnology.com>
References: <511BEECD.5050809@ceetonetechnology.com>
Message-ID: <20130213150254.6afff311@ivory.wynn.com>

On Wed, 13 Feb 2013 14:51:41 -0500
George Rosamond <george at ceetonetechnology.com> wrote:

> I can't seem to find it anywhere... so I thought I'd ping here.
> 

Greeting-

I have no idea how to do it from binary already installed packages, but
when you build a port a simple 

make package-recursive 

seems to make all the needed packages and puts them
in /usr/ports/packages if the dir exists.

I have built a few ports on my Pi in the last few days and they all did
the right thing with make package-recursive.  I had to force pkgng onto
the Bone to then install them on the Bone because the Bone expected to
use pkg, not pkgng.  I am not sure why the disconenct between the two
images.

I am going to write something up on this in the next few days.

-Brett


-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

I would never invade the United States.  There would be a gun behind
every blade of grass.  --Isoroku Yamamoto



From pete at nomadlogic.org  Wed Feb 13 15:07:07 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Wed, 13 Feb 2013 12:07:07 -0800
Subject: [nycbug-talk] pkgng question
In-Reply-To: <511BEECD.5050809@ceetonetechnology.com>
References: <511BEECD.5050809@ceetonetechnology.com>
Message-ID: <511BF26B.4040309@nomadlogic.org>

On 02/13/13 11:51, George Rosamond wrote:
> I can't seem to find it anywhere... so I thought I'd ping here.
>
> In old pkg days, you just use/used:
>
> # pkg_create -Rxjb <installed-port>
>
> to create a package *with dependencies*
>
> How do you do this in pkgng?  I'd rather not create a package for each...
>
> Doesn't seem to be in man page or online...
>
> TIA


the easiest method i've found to do this automagically is using 
poudriere from the ports tree.  it requires zfs though :/  you can also 
use tinderbox - although i have no personal experience using that for 
pkgng builds.

http://blog.etoilebsd.net/post/Home_made_pkgng_repo
https://fossil.etoilebsd.net/poudriere/doc/trunk/doc/pkgng_repos.wiki

it looks like for tinderbox to work you need to use the dev version for 
pkgng support:
https://wiki.freebsd.org/pkgng
"(Note: pkgng support in the stable tinderbox port 
<http://www.freshports.org/ports-mgmt/tinderbox> will be available from 
the upcoming version 4.0; until then you will need to use a development 
version <http://www.freshports.org/ports-mgmt/tinderbox-devel> or 
checkout sources from MarcusCom CVS 
<http://www.marcuscom.com:8080/cgi-bin/cvsweb.cgi/>.) "


-pete




-- 
Pete Wright
pete at nomadlogic.org
twitter =>  @nomadlogicLA


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130213/8bc03c00/attachment.htm>

From pete at nomadlogic.org  Wed Feb 13 15:13:01 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Wed, 13 Feb 2013 12:13:01 -0800
Subject: [nycbug-talk] pkgng question
In-Reply-To: <20130213200149.GB1481@glenbarber.us>
References: <511BEECD.5050809@ceetonetechnology.com>
 <20130213200149.GB1481@glenbarber.us>
Message-ID: <511BF3CD.3050702@nomadlogic.org>

On 02/13/13 12:01, Glen Barber wrote:
> On Wed, Feb 13, 2013 at 02:51:41PM -0500, George Rosamond wrote:
>> I can't seem to find it anywhere... so I thought I'd ping here.
>>
>> In old pkg days, you just use/used:
>>
>> # pkg_create -Rxjb<installed-port>
>>
>> to create a package *with dependencies*
>>
>> How do you do this in pkgng?  I'd rather not create a package for each...
>>
>> Doesn't seem to be in man page or online...
>>
> I use pkg-info to generate the dependency list, then do something like
> this:
>
>      # pkg create -o /usr/ports/packages/All `pkg info -qr foo`

awesome!

-p


-- 
Pete Wright
pete at nomadlogic.org
twitter =>  @nomadlogicLA




From george at ceetonetechnology.com  Wed Feb 13 15:15:37 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Wed, 13 Feb 2013 15:15:37 -0500
Subject: [nycbug-talk] pkgng question
In-Reply-To: <511BF26B.4040309@nomadlogic.org>
References: <511BEECD.5050809@ceetonetechnology.com>
 <511BF26B.4040309@nomadlogic.org>
Message-ID: <511BF469.80602@ceetonetechnology.com>

On 02/13/13 15:07, Pete Wright wrote:
> On 02/13/13 11:51, George Rosamond wrote:
>> I can't seem to find it anywhere... so I thought I'd ping here.
>>
>> In old pkg days, you just use/used:
>>
>> # pkg_create -Rxjb <installed-port>
>>
>> to create a package *with dependencies*
>>
>> How do you do this in pkgng?  I'd rather not create a package for each...
>>
>> Doesn't seem to be in man page or online...
>>
>> TIA
>
>
> the easiest method i've found to do this automagically is using
> poudriere from the ports tree.  it requires zfs though :/  you can also
> use tinderbox - although i have no personal experience using that for
> pkgng builds.
>
> http://blog.etoilebsd.net/post/Home_made_pkgng_repo
> https://fossil.etoilebsd.net/poudriere/doc/trunk/doc/pkgng_repos.wiki
>
> it looks like for tinderbox to work you need to use the dev version for
> pkgng support:
> https://wiki.freebsd.org/pkgng
> "(Note: pkgng support in the stable tinderbox port
> <http://www.freshports.org/ports-mgmt/tinderbox> will be available from
> the upcoming version 4.0; until then you will need to use a development
> version <http://www.freshports.org/ports-mgmt/tinderbox-devel> or
> checkout sources from MarcusCom CVS
> <http://www.marcuscom.com:8080/cgi-bin/cvsweb.cgi/>.) "


yeah... I would definitely use poudriere for the arm package repository.

But it also means us getting a real box in the colo that can handle ZFS...

At this point, the colo is a lot of dated hardware, except for a handful 
of boxes which are out of our direct control: pfSense, BSD Cert..

g



From nycbug at wynn.com  Thu Feb 14 10:40:53 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 14 Feb 2013 10:40:53 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
Message-ID: <20130214104053.37716dea@ivory.wynn.com>

Greeting-

As a VERY LONG TIME Systems Admin, as in I have been doing it longer
than many developers have been alive my issues with the
move to svn from cvs are totally related to SIZE & LICENSE.
1.  SIZE
    It is obscene to require a source fetching tool that is so huge
    compared to the size of the base OS.  This is especially true for
    those of us that deal with small systems.

2.  There are times when one wants to keep NON-BSD-Licensed code off
    of a system. I do not at this time have any need to do such a thing,
    but in the past I had that need sometimes.

When I started working with Unix full time in the early 1980s I used
EMACS as my editor of choice.  At some point Richard got EMACS up to
about 50Mb, which considering at the time I was running sun 4-110s with
120Mb disks was HUGE.  I dropped EMACS.  

The idea that systems administrators should be forced to
install/maintain a huge tool to keep their systems up to date with
source is silly at best.

I believe that several years ago when portsnap was brought to life that
a similar tool for grabbing /usr/src should have been introduced.  I
know I would have jumped all over that.  At the time of the
introduction of portsnap I was working at a government agency where I
did not have control over the firewall and the network crew would not
open up the needed ports for CVS.  I was forced to tunnel all cvsup
traffic via an outside FreeBSD Box that was running ssh on a port
assigned to something the firewall folks did allow.  What a pain.  Then
came portsnap running via http and life was easier!

So this begs the question why not something like "basesnap fetch"?

In any case I feel that long before now the discussion of how to fetch
base should have been started and a suitable tool should have been
proposed and coded.  This "how to fetch base" issue shows the same lack
of forethought exhibited by changing the system installer and breaking
sysinstall by changing the layout of the FTP servers before the new
installer actually worked. AFIK the new installer is still broken, but
I have not tried to install using it since I discovered the MFSBSD iso
images, which are small and give a simple fool proof install method.

I do not know the ins and outs of svn or how hard it is to build a
fetch tool.  I am probably not a good enough programmer to do the job
either, but I would urge someone who is a good programmer to take on
the problem and provide a solution before /usr/src can no longer be
fetched with csup.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

"The strongest reason for the people to retain the right to keep 
and bear arms is, as a last resort, to protect themselves against 
tyranny in government" - Thomas Jefferson. 



From bonsaime at gmail.com  Thu Feb 14 12:38:08 2013
From: bonsaime at gmail.com (Jesse Callaway)
Date: Thu, 14 Feb 2013 12:38:08 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130214104053.37716dea@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
Message-ID: <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>

Create your own SVN mirror and host it somewhere.

Provide HEAD or whatever other tags you are interested in accessible over
http/s or ftp or whatever you like. Your build environment would then need
curl or lftp instead of subversion or cvs clients. A source torrent might
not be a bad idea.

Does that work? I'm not sure what your environment and constraints are
like. I forget the FreeBSD build steps, but I think that fetching source is
an exercise left to the user and is not baked in.


On Thu, Feb 14, 2013 at 10:40 AM, Brett Wynkoop <nycbug at wynn.com> wrote:

> Greeting-
>
> As a VERY LONG TIME Systems Admin, as in I have been doing it longer
> than many developers have been alive my issues with the
> move to svn from cvs are totally related to SIZE & LICENSE.
> 1.  SIZE
>     It is obscene to require a source fetching tool that is so huge
>     compared to the size of the base OS.  This is especially true for
>     those of us that deal with small systems.
>
> 2.  There are times when one wants to keep NON-BSD-Licensed code off
>     of a system. I do not at this time have any need to do such a thing,
>     but in the past I had that need sometimes.
>
> When I started working with Unix full time in the early 1980s I used
> EMACS as my editor of choice.  At some point Richard got EMACS up to
> about 50Mb, which considering at the time I was running sun 4-110s with
> 120Mb disks was HUGE.  I dropped EMACS.
>
> The idea that systems administrators should be forced to
> install/maintain a huge tool to keep their systems up to date with
> source is silly at best.
>
> I believe that several years ago when portsnap was brought to life that
> a similar tool for grabbing /usr/src should have been introduced.  I
> know I would have jumped all over that.  At the time of the
> introduction of portsnap I was working at a government agency where I
> did not have control over the firewall and the network crew would not
> open up the needed ports for CVS.  I was forced to tunnel all cvsup
> traffic via an outside FreeBSD Box that was running ssh on a port
> assigned to something the firewall folks did allow.  What a pain.  Then
> came portsnap running via http and life was easier!
>
> So this begs the question why not something like "basesnap fetch"?
>
> In any case I feel that long before now the discussion of how to fetch
> base should have been started and a suitable tool should have been
> proposed and coded.  This "how to fetch base" issue shows the same lack
> of forethought exhibited by changing the system installer and breaking
> sysinstall by changing the layout of the FTP servers before the new
> installer actually worked. AFIK the new installer is still broken, but
> I have not tried to install using it since I discovered the MFSBSD iso
> images, which are small and give a simple fool proof install method.
>
> I do not know the ins and outs of svn or how hard it is to build a
> fetch tool.  I am probably not a good enough programmer to do the job
> either, but I would urge someone who is a good programmer to take on
> the problem and provide a solution before /usr/src can no longer be
> fetched with csup.
>
> -Brett
>
> --
>
> wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
> 917-642-6925
> 718-717-5435
>
> "The strongest reason for the people to retain the right to keep
> and bear arms is, as a last resort, to protect themselves against
> tyranny in government" - Thomas Jefferson.
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>



-- 
-jesse
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130214/36135f0f/attachment.htm>

From bcallah at devio.us  Thu Feb 14 12:55:08 2013
From: bcallah at devio.us (Brian Callahan)
Date: Thu, 14 Feb 2013 12:55:08 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
Message-ID: <511D24FC.9040300@devio.us>

On 2/14/2013 12:38 PM, Jesse Callaway wrote:
> Create your own SVN mirror and host it somewhere.
>
> Provide HEAD or whatever other tags you are interested in accessible
> over http/s or ftp or whatever you like. Your build environment would
> then need curl or lftp instead of subversion or cvs clients. A source
> torrent might not be a bad idea.
>
> Does that work? I'm not sure what your environment and constraints are
> like. I forget the FreeBSD build steps, but I think that fetching source
> is an exercise left to the user and is not baked in.
>

You need to extrapolate his problem outward. Even assuming that solves 
his particular case, do you really want to "official" solution to be "go 
host your own server elsewhere, deal with all the hassles of chasing 
HEAD, maintain svn (which he says he doesn't want to do!), etc., etc."?

~Brian

>
> On Thu, Feb 14, 2013 at 10:40 AM, Brett Wynkoop <nycbug at wynn.com
> <mailto:nycbug at wynn.com>> wrote:
>
>     Greeting-
>
>     As a VERY LONG TIME Systems Admin, as in I have been doing it longer
>     than many developers have been alive my issues with the
>     move to svn from cvs are totally related to SIZE & LICENSE.
>     1.  SIZE
>          It is obscene to require a source fetching tool that is so huge
>          compared to the size of the base OS.  This is especially true for
>          those of us that deal with small systems.
>
>     2.  There are times when one wants to keep NON-BSD-Licensed code off
>          of a system. I do not at this time have any need to do such a
>     thing,
>          but in the past I had that need sometimes.
>
>     When I started working with Unix full time in the early 1980s I used
>     EMACS as my editor of choice.  At some point Richard got EMACS up to
>     about 50Mb, which considering at the time I was running sun 4-110s with
>     120Mb disks was HUGE.  I dropped EMACS.
>
>     The idea that systems administrators should be forced to
>     install/maintain a huge tool to keep their systems up to date with
>     source is silly at best.
>
>     I believe that several years ago when portsnap was brought to life that
>     a similar tool for grabbing /usr/src should have been introduced.  I
>     know I would have jumped all over that.  At the time of the
>     introduction of portsnap I was working at a government agency where I
>     did not have control over the firewall and the network crew would not
>     open up the needed ports for CVS.  I was forced to tunnel all cvsup
>     traffic via an outside FreeBSD Box that was running ssh on a port
>     assigned to something the firewall folks did allow.  What a pain.  Then
>     came portsnap running via http and life was easier!
>
>     So this begs the question why not something like "basesnap fetch"?
>
>     In any case I feel that long before now the discussion of how to fetch
>     base should have been started and a suitable tool should have been
>     proposed and coded.  This "how to fetch base" issue shows the same lack
>     of forethought exhibited by changing the system installer and breaking
>     sysinstall by changing the layout of the FTP servers before the new
>     installer actually worked. AFIK the new installer is still broken, but
>     I have not tried to install using it since I discovered the MFSBSD iso
>     images, which are small and give a simple fool proof install method.
>
>     I do not know the ins and outs of svn or how hard it is to build a
>     fetch tool.  I am probably not a good enough programmer to do the job
>     either, but I would urge someone who is a good programmer to take on
>     the problem and provide a solution before /usr/src can no longer be
>     fetched with csup.
>
>     -Brett
>
>     --
>
>     wynkoop at wynn.com <mailto:wynkoop at wynn.com>
>     http://prd4.wynn.com/wynkoop/pgp-keys.txt
>     917-642-6925 <tel:917-642-6925>
>     718-717-5435 <tel:718-717-5435>
>
>     "The strongest reason for the people to retain the right to keep
>     and bear arms is, as a last resort, to protect themselves against
>     tyranny in government" - Thomas Jefferson.
>
>     _______________________________________________
>     talk mailing list
>     talk at lists.nycbug.org <mailto:talk at lists.nycbug.org>
>     http://lists.nycbug.org/mailman/listinfo/talk
>
>
>
>
> --
> -jesse
>
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>



From pete at nomadlogic.org  Thu Feb 14 13:20:31 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Thu, 14 Feb 2013 10:20:31 -0800
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <20130214104053.37716dea@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
Message-ID: <511D2AEF.9030606@nomadlogic.org>

On 02/14/13 07:40, Brett Wynkoop wrote:
> Greeting-
>
>
> The idea that systems administrators should be forced to
> install/maintain a huge tool to keep their systems up to date with
> source is silly at best.

IMHO in 2013 requiring admins to maintain a source tree to keep their 
individual systems up-to-date is a silly idea at best.

for large sites, having a source checkout that includes all revision 
history seems like a very good idea.  the other option being a download 
of an exported snapshot of the source tree w/o that history.  by be able 
to just submit an "svn diff" against your local check-out seems much 
more user friendly to the devs.


-pete

-- 
Pete Wright
pete at nomadlogic.org
twitter =>  @nomadlogicLA




From nycbug at wynn.com  Thu Feb 14 13:49:02 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 14 Feb 2013 13:49:02 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <511D24FC.9040300@devio.us>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us>
Message-ID: <20130214134902.2e532e5d@ivory.wynn.com>

On Thu, 14 Feb 2013 12:55:08 -0500
Brian Callahan <bcallah at devio.us> wrote:

> On 2/14/2013 12:38 PM, Jesse Callaway wrote:
> > Create your own SVN mirror and host it somewhere.

SNIP

> You need to extrapolate his problem outward. Even assuming that
> solves his particular case, do you really want to "official" solution
> to be "go host your own server elsewhere, deal with all the hassles
> of chasing HEAD, maintain svn (which he says he doesn't want to do!),
> etc., etc."?
> 
> ~Brian
> 

Brian is correct, while what you propose might work it is not
reasonable.  Am I to tell each client that has FreeBSD installed that
they need to set up a special update server and maintain that to allow
for grabbing the sources.  Nope.

I am dealing with similar issues on a development project at work where
the hard core programming staff thinks it is nothing to change what is
required of sys-admins to install our product.  My first install of the
product in it's new form required 3 developers to help me over a period
of several weeks before the product was installed and functional in my
lab.  Needless to say I yelled and screamed which lead to folks above
me saying,  "Wynkoop has a point, we better figure out how to make it
as easy as it used to be"....that we turned into "Brett how about
figuring it out and circulate your solution to dev for feedback".  

The FreeBSD project needs to worry about mindset and market share just
like a for profit software operations or we will lose users, financial
supporters and eventually developers because even if we have the best
kernel or the greatest smp or superpages, or ZFS or
BIG_MAGIC_OF_SOME_SORT people have only so much willingness to deal
with pain.  I am nearing the end of my pain tolerance with FreeBSD, I
am sure I am not the only one.

Jesse I assure you I can come up with HACKENSTEIN solutions to the
problem, but the point is that a good general solution needs to happen
for the entire community of users, large and small.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

"The strongest reason for the people to retain the right to keep 
and bear arms is, as a last resort, to protect themselves against 
tyranny in government" - Thomas Jefferson. 



From mspitzer at gmail.com  Thu Feb 14 13:51:12 2013
From: mspitzer at gmail.com (Marc Spitzer)
Date: Thu, 14 Feb 2013 13:51:12 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130214104053.37716dea@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
Message-ID: <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>

Brett,

Don't take this the wrong way but why do you need to build from source at
all on all your servers?  Why not set up a freebsd-update server that
serves your current build and install the build only tools, including svn,
on 1 crappy box(8gb ram, 500gb disk, 4 cores)  and build binary updates for
all your serves and then use the binary update tools that come with the os
to patch?  This is much more maintainable, scaleable and auditable.

If all your servers are special and unique the odds are you are doing it
wrong, very wrong in this age of cheap commodity hardware.   Even if all
your servers are special and unique there is nothing stopping you from
setting up a freebsd-update server for each of them so you never have to
install source build tools on the operational systems.

marc



On Thu, Feb 14, 2013 at 10:40 AM, Brett Wynkoop <nycbug at wynn.com> wrote:

> Greeting-
>
> As a VERY LONG TIME Systems Admin, as in I have been doing it longer
> than many developers have been alive my issues with the
> move to svn from cvs are totally related to SIZE & LICENSE.
> 1.  SIZE
>     It is obscene to require a source fetching tool that is so huge
>     compared to the size of the base OS.  This is especially true for
>     those of us that deal with small systems.
>
> 2.  There are times when one wants to keep NON-BSD-Licensed code off
>     of a system. I do not at this time have any need to do such a thing,
>     but in the past I had that need sometimes.
>
> When I started working with Unix full time in the early 1980s I used
> EMACS as my editor of choice.  At some point Richard got EMACS up to
> about 50Mb, which considering at the time I was running sun 4-110s with
> 120Mb disks was HUGE.  I dropped EMACS.
>
> The idea that systems administrators should be forced to
> install/maintain a huge tool to keep their systems up to date with
> source is silly at best.
>
> I believe that several years ago when portsnap was brought to life that
> a similar tool for grabbing /usr/src should have been introduced.  I
> know I would have jumped all over that.  At the time of the
> introduction of portsnap I was working at a government agency where I
> did not have control over the firewall and the network crew would not
> open up the needed ports for CVS.  I was forced to tunnel all cvsup
> traffic via an outside FreeBSD Box that was running ssh on a port
> assigned to something the firewall folks did allow.  What a pain.  Then
> came portsnap running via http and life was easier!
>
> So this begs the question why not something like "basesnap fetch"?
>
> In any case I feel that long before now the discussion of how to fetch
> base should have been started and a suitable tool should have been
> proposed and coded.  This "how to fetch base" issue shows the same lack
> of forethought exhibited by changing the system installer and breaking
> sysinstall by changing the layout of the FTP servers before the new
> installer actually worked. AFIK the new installer is still broken, but
> I have not tried to install using it since I discovered the MFSBSD iso
> images, which are small and give a simple fool proof install method.
>
> I do not know the ins and outs of svn or how hard it is to build a
> fetch tool.  I am probably not a good enough programmer to do the job
> either, but I would urge someone who is a good programmer to take on
> the problem and provide a solution before /usr/src can no longer be
> fetched with csup.
>
> -Brett
>
> --
>
> wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
> 917-642-6925
> 718-717-5435
>
> "The strongest reason for the people to retain the right to keep
> and bear arms is, as a last resort, to protect themselves against
> tyranny in government" - Thomas Jefferson.
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>



-- 
Freedom is nothing but a chance to be better.
--Albert Camus

The inherent vice of capitalism is the unequal sharing of blessings; the
inherent virtue of socialism is the equal sharing of miseries.
-- Winston Churchill

Do the arithmetic or be doomed to talk nonsense.
--John McCarthy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130214/22a9f00b/attachment.htm>

From nycbug at wynn.com  Thu Feb 14 14:02:25 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 14 Feb 2013 14:02:25 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
Message-ID: <20130214140225.480aa649@ivory.wynn.com>

On Thu, 14 Feb 2013 13:51:12 -0500
Marc Spitzer <mspitzer at gmail.com> wrote:

> Brett,
> 
> Don't take this the wrong way but why do you need to build from
> source at all on all your servers?  Why not set up a freebsd-update
> server that serves your current build and install the build only
> tools, including svn, on 1 crappy box(8gb ram, 500gb disk, 4 cores)
> and build binary updates for all your serves and then use the binary
> update tools that come with the os to patch?  This is much more
> maintainable, scaleable and auditable.
> 
> If all your servers are special and unique the odds are you are doing
> it wrong, very wrong in this age of cheap commodity hardware.   Even
> if all your servers are special and unique there is nothing stopping
> you from setting up a freebsd-update server for each of them so you
> never have to install source build tools on the operational systems.
> 
> marc

No Marc the odds are that I have systems I maintain at CUSTOMER-A
CUSTOMER-B CUSTOMER-C, and now I have x86,AMD64 and ARM at my own place
as well as mixtures of those at clients.


Nothing you mentioned excuses the bloat of svn being required for
updates.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

"The strongest reason for the people to retain the right to keep 
and bear arms is, as a last resort, to protect themselves against 
tyranny in government" - Thomas Jefferson. 



From mspitzer at gmail.com  Thu Feb 14 14:24:45 2013
From: mspitzer at gmail.com (Marc Spitzer)
Date: Thu, 14 Feb 2013 14:24:45 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130214140225.480aa649@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
Message-ID: <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>

On Thu, Feb 14, 2013 at 2:02 PM, Brett Wynkoop <nycbug at wynn.com> wrote:

>
> No Marc the odds are that I have systems I maintain at CUSTOMER-A
> CUSTOMER-B CUSTOMER-C, and now I have x86,AMD64 and ARM at my own place
> as well as mixtures of those at clients.
>
>
> Nothing you mentioned excuses the bloat of svn being required for
> updates.
>
> -Brett
>

The "bloat" of subversion is not required for updates, the "bloat" of
subversion is required once for building binary updates there is a
difference.  And the tools may have warts that I am not aware of, but any
real issues will be addressed in a reasonable time frame.

And even if you have multiple clients there is nothing stopping you from
doing exactly what I said you could do and make an update server, not
a separate computer, for each of them on a small little box you take around
or use ssh tunnels to access back to your server.  See no subversion needed
on your production boxes, in fact no dev tool chain needed on production
boxes and every file installed has a checsum and master copy stored
elsewhere for audit/recovery purposes.

marc

-- 
Freedom is nothing but a chance to be better.
--Albert Camus

The inherent vice of capitalism is the unequal sharing of blessings; the
inherent virtue of socialism is the equal sharing of miseries.
-- Winston Churchill

Do the arithmetic or be doomed to talk nonsense.
--John McCarthy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130214/019d91a0/attachment.htm>

From bcallah at devio.us  Thu Feb 14 15:26:53 2013
From: bcallah at devio.us (Brian Callahan)
Date: Thu, 14 Feb 2013 15:26:53 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
Message-ID: <511D488D.1030209@devio.us>

On 2/14/2013 2:24 PM, Marc Spitzer wrote:
>
>
>
> On Thu, Feb 14, 2013 at 2:02 PM, Brett Wynkoop <nycbug at wynn.com
> <mailto:nycbug at wynn.com>> wrote:
>
>
>     No Marc the odds are that I have systems I maintain at CUSTOMER-A
>     CUSTOMER-B CUSTOMER-C, and now I have x86,AMD64 and ARM at my own place
>     as well as mixtures of those at clients.
>
>
>     Nothing you mentioned excuses the bloat of svn being required for
>     updates.
>
>     -Brett
>
>
> The "bloat" of subversion is not required for updates, the "bloat" of
> subversion is required once for building binary updates there is a
> difference.  And the tools may have warts that I am not aware of, but
> any real issues will be addressed in a reasonable time frame.
>
> And even if you have multiple clients there is nothing stopping you from
> doing exactly what I said you could do and make an update server, not
> a separate computer, for each of them on a small little box you take
> around or use ssh tunnels to access back to your server.  See no
> subversion needed on your production boxes, in fact no dev tool chain
> needed on production boxes and every file installed has a checsum and
> master copy stored elsewhere for audit/recovery purposes.
>
> marc
>

I'm going to throw something out here, and feel free to disagree with me.

Easy access to source, even if we grant that it is "silly in 2013" as 
has been stated before, is no excuse for no easy access to source.

There are many reasons one would want to have easy access to source. 
Some people want to learn, others may want to be able to quick reference 
things for projects (school/work/etc.), others still may enjoy being 
weekend tinkerers. Even more reasons exist for wanting easy access to 
source.

Some of those people who fit in the above paragraph today will go on to 
become devs tomorrow. Case in point, me. I went from a weekend tinkerer 
to a dev. It would have been nigh impossible if obsd didn't have an easy 
and well-documented way of getting all sources from a base install.

If even one person (/business/manufacturer/whatever) skips out on fbsd 
because of this svn debacle, it should be considered a massive failure 
in the eyes of the project and the devs, because who knows - that one 
person could have gone on to become the next core team member.

I get that I am and always will be on the outside of this, not being a 
fbsd person. But this is how I see it.

FWIW, I'd take Brett's suggestion of a basesnap program and run with it. 
Fbsd should already have the infrastructure necessary to do this (I 
imagine it would be similar to "do whatever you do for portsnap in the 
ports dir, and do this for the base dir").

~Brian


From bonsaime at gmail.com  Thu Feb 14 15:55:26 2013
From: bonsaime at gmail.com (Jesse Callaway)
Date: Thu, 14 Feb 2013 15:55:26 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130214134902.2e532e5d@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us>
 <20130214134902.2e532e5d@ivory.wynn.com>
Message-ID: <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>

It should be easy enough to set up an https service with the source you
need which is kept in sync with the FreeBSD source.

Not sure what is so difficult or controversial about this. I usually do
this sort of a thing when I have more than 10 boxes to take care of, just
out of bandwidth and speed concerns.


On Thu, Feb 14, 2013 at 1:49 PM, Brett Wynkoop <nycbug at wynn.com> wrote:

> On Thu, 14 Feb 2013 12:55:08 -0500
> Brian Callahan <bcallah at devio.us> wrote:
>
> > On 2/14/2013 12:38 PM, Jesse Callaway wrote:
> > > Create your own SVN mirror and host it somewhere.
>
> SNIP
>
> > You need to extrapolate his problem outward. Even assuming that
> > solves his particular case, do you really want to "official" solution
> > to be "go host your own server elsewhere, deal with all the hassles
> > of chasing HEAD, maintain svn (which he says he doesn't want to do!),
> > etc., etc."?
> >
> > ~Brian
> >
>
> Brian is correct, while what you propose might work it is not
> reasonable.  Am I to tell each client that has FreeBSD installed that
> they need to set up a special update server and maintain that to allow
> for grabbing the sources.  Nope.
>
> I am dealing with similar issues on a development project at work where
> the hard core programming staff thinks it is nothing to change what is
> required of sys-admins to install our product.  My first install of the
> product in it's new form required 3 developers to help me over a period
> of several weeks before the product was installed and functional in my
> lab.  Needless to say I yelled and screamed which lead to folks above
> me saying,  "Wynkoop has a point, we better figure out how to make it
> as easy as it used to be"....that we turned into "Brett how about
> figuring it out and circulate your solution to dev for feedback".
>
> The FreeBSD project needs to worry about mindset and market share just
> like a for profit software operations or we will lose users, financial
> supporters and eventually developers because even if we have the best
> kernel or the greatest smp or superpages, or ZFS or
> BIG_MAGIC_OF_SOME_SORT people have only so much willingness to deal
> with pain.  I am nearing the end of my pain tolerance with FreeBSD, I
> am sure I am not the only one.
>
> Jesse I assure you I can come up with HACKENSTEIN solutions to the
> problem, but the point is that a good general solution needs to happen
> for the entire community of users, large and small.
>
> -Brett
>
> --
>
> wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
> 917-642-6925
> 718-717-5435
>
> "The strongest reason for the people to retain the right to keep
> and bear arms is, as a last resort, to protect themselves against
> tyranny in government" - Thomas Jefferson.
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>



-- 
-jesse
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130214/e3d70ffa/attachment.htm>

From jhb at freebsd.org  Thu Feb 14 16:17:25 2013
From: jhb at freebsd.org (John Baldwin)
Date: Thu, 14 Feb 2013 16:17:25 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <511D488D.1030209@devio.us>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
Message-ID: <201302141617.25528.jhb@freebsd.org>

On Thursday, February 14, 2013 3:26:53 pm Brian Callahan wrote:
> FWIW, I'd take Brett's suggestion of a basesnap program and run with it. 
> Fbsd should already have the infrastructure necessary to do this (I 
> imagine it would be similar to "do whatever you do for portsnap in the 
> ports dir, and do this for the base dir").

The problem with a basesnap (as compared to csup) is that it would only let 
you move forward.  It wouldn't let you do bisecting operations.  With ports 
only moving forward is fine.  But for source it is very common for users to 
narrow down which commit causes breakage that they are seeing, and that 
requires being able to move both forwards and backwards in time as well as
moving to arbitrary points in time (individual revisions).  An svnup that 
functions as a light-weight svn client would provide that, but to date it has 
proven a non-trivial task. :(

-- 
John Baldwin


From mspitzer at gmail.com  Thu Feb 14 18:28:20 2013
From: mspitzer at gmail.com (Marc Spitzer)
Date: Thu, 14 Feb 2013 18:28:20 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <511D488D.1030209@devio.us>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
Message-ID: <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>

On Thu, Feb 14, 2013 at 3:26 PM, Brian Callahan <bcallah at devio.us> wrote:

> On 2/14/2013 2:24 PM, Marc Spitzer wrote:
>
>>
>>
>>
>> On Thu, Feb 14, 2013 at 2:02 PM, Brett Wynkoop <nycbug at wynn.com
>> <mailto:nycbug at wynn.com>> wrote:
>>
>>
>>     No Marc the odds are that I have systems I maintain at CUSTOMER-A
>>     CUSTOMER-B CUSTOMER-C, and now I have x86,AMD64 and ARM at my own
>> place
>>     as well as mixtures of those at clients.
>>
>>
>>     Nothing you mentioned excuses the bloat of svn being required for
>>     updates.
>>
>>     -Brett
>>
>>
>> The "bloat" of subversion is not required for updates, the "bloat" of
>> subversion is required once for building binary updates there is a
>> difference.  And the tools may have warts that I am not aware of, but
>> any real issues will be addressed in a reasonable time frame.
>>
>> And even if you have multiple clients there is nothing stopping you from
>> doing exactly what I said you could do and make an update server, not
>> a separate computer, for each of them on a small little box you take
>> around or use ssh tunnels to access back to your server.  See no
>> subversion needed on your production boxes, in fact no dev tool chain
>> needed on production boxes and every file installed has a checsum and
>> master copy stored elsewhere for audit/recovery purposes.
>>
>> marc
>>
>>
> I'm going to throw something out here, and feel free to disagree with me.
>
> Easy access to source, even if we grant that it is "silly in 2013" as has
> been stated before, is no excuse for no easy access to source.
>

How is source access not easy?  just install svn and you have it.  Is there
a hard step I am missing?


>
> There are many reasons one would want to have easy access to source. Some
> people want to learn, others may want to be able to quick reference things
> for projects (school/work/etc.), others still may enjoy being weekend
> tinkerers. Even more reasons exist for wanting easy access to source.
>
Some of those people who fit in the above paragraph today will go on to
> become devs tomorrow. Case in point, me. I went from a weekend tinkerer to
> a dev. It would have been nigh impossible if obsd didn't have an easy and
> well-documented way of getting all sources from a base install.
>

again install svn and done, ie cd /usr/ports/devel/subversion && make &&
make install is the hard way.  To be honest I have small hope of anyone
becoming a developer if svn is a show stopper.



>
> If even one person (/business/manufacturer/**whatever) skips out on fbsd
> because of this svn debacle, it should be considered a massive failure in
> the eyes of the project and the devs, because who knows - that one person
> could have gone on to become the next core team member.
>

one person is not a failure, let alone a debacle.  I think freebsd made the
right decision for freebsd from the available choices as I understand it.
 the community of freebsd devs made the decision for valid reasons as
believed by the people who get to decide and some people are not happy
about it and blowing it out of proportion, well ok life goes on.


>
> I get that I am and always will be on the outside of this, not being a
> fbsd person. But this is how I see it.
>
> FWIW, I'd take Brett's suggestion of a basesnap program and run with it.
> Fbsd should already have the infrastructure necessary to do this (I imagine
> it would be similar to "do whatever you do for portsnap in the ports dir,
> and do this for the base dir").
>
> ~Brian
>

to be honest I think that the binary updates are vastly superior to source
updates, I get check sums, reinstall/clone is faster and when build tools
show up in my web server I just may have a problem.

marc
-- 
Freedom is nothing but a chance to be better.
--Albert Camus

The inherent vice of capitalism is the unequal sharing of blessings; the
inherent virtue of socialism is the equal sharing of miseries.
-- Winston Churchill

Do the arithmetic or be doomed to talk nonsense.
--John McCarthy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130214/66ca3f46/attachment.htm>

From bcallah at devio.us  Thu Feb 14 21:13:05 2013
From: bcallah at devio.us (Brian Callahan)
Date: Thu, 14 Feb 2013 21:13:05 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
Message-ID: <511D99B1.40501@devio.us>

On 2/14/2013 6:28 PM, Marc Spitzer wrote:
>
>     I'm going to throw something out here, and feel free to disagree
>     with me.
>
>     Easy access to source, even if we grant that it is "silly in 2013"
>     as has been stated before, is no excuse for no easy access to source.
>
>
> How is source access not easy?  just install svn and you have it.  Is
> there a hard step I am missing?
>

What's the undue hardship of having a tool in base that can do this? As 
far as I understand it, the issue here boils down to:
1. I have FreeBSD.
2. I want sources.
3. I do not want to/cannot run svn.
How can I get sources?

"You can't" should be an unacceptable answer in my opinion.

Anyhow, I've put in my two cents. I'm done with this topic.

~Brian


From lists at eitanadler.com  Thu Feb 14 21:17:41 2013
From: lists at eitanadler.com (Eitan Adler)
Date: Thu, 14 Feb 2013 21:17:41 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <511D99B1.40501@devio.us>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <511D99B1.40501@devio.us>
Message-ID: <CAF6rxgko5ngC-1MBJY9PC9QVrFachdxEPDkir-KqLorvHBO-WA@mail.gmail.com>

On 14 February 2013 21:13, Brian Callahan <bcallah at devio.us> wrote:
> On 2/14/2013 6:28 PM, Marc Spitzer wrote:
>>
>>
>>     I'm going to throw something out here, and feel free to disagree
>>     with me.
>>
>>     Easy access to source, even if we grant that it is "silly in 2013"
>>     as has been stated before, is no excuse for no easy access to source.
>>
>>
>> How is source access not easy?  just install svn and you have it.  Is
>> there a hard step I am missing?
>>
>
> What's the undue hardship of having a tool in base that can do this? As far
> as I understand it, the issue here boils down to:
> 1. I have FreeBSD.
> 2. I want sources.
> 3. I do not want to/cannot run svn.
> How can I get sources?

freebsd-update ?


-- 
Eitan Adler


From mspitzer at gmail.com  Thu Feb 14 22:00:52 2013
From: mspitzer at gmail.com (Marc Spitzer)
Date: Thu, 14 Feb 2013 22:00:52 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <511D99B1.40501@devio.us>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <511D99B1.40501@devio.us>
Message-ID: <CA+qr9e6tSpZiW8sHPx_03HPF3TQ76yzns1YuUtbu00LXLenqbw@mail.gmail.com>

On Thu, Feb 14, 2013 at 9:13 PM, Brian Callahan <bcallah at devio.us> wrote:

> On 2/14/2013 6:28 PM, Marc Spitzer wrote:
>
>>
>>     I'm going to throw something out here, and feel free to disagree
>>     with me.
>>
>>     Easy access to source, even if we grant that it is "silly in 2013"
>>     as has been stated before, is no excuse for no easy access to source.
>>
>>
>> How is source access not easy?  just install svn and you have it.  Is
>> there a hard step I am missing?
>>
>>
> What's the undue hardship of having a tool in base that can do this? As
> far as I understand it, the issue here boils down to:
> 1. I have FreeBSD.
> 2. I want sources.
> 3. I do not want to/cannot run svn.
> How can I get sources?
>

Since binary patching is how things are getting done for the os patching,
why should there be a tool in base to grab source?  and you can download
the system source code using wget or curl, unless they are also too big to
install on your system.  Source is not as critical to operational
management of your system as it used to be, things change.


>
> "You can't" should be an unacceptable answer in my opinion.
>
> Anyhow, I've put in my two cents. I'm done with this topic.


As have I, remember when you could buy bubble gum with two cents?

marc

-- 
Freedom is nothing but a chance to be better.
--Albert Camus

The inherent vice of capitalism is the unequal sharing of blessings; the
inherent virtue of socialism is the equal sharing of miseries.
-- Winston Churchill

Do the arithmetic or be doomed to talk nonsense.
--John McCarthy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130214/771fa1a2/attachment.htm>

From lists at eitanadler.com  Thu Feb 14 22:05:54 2013
From: lists at eitanadler.com (Eitan Adler)
Date: Thu, 14 Feb 2013 22:05:54 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e6tSpZiW8sHPx_03HPF3TQ76yzns1YuUtbu00LXLenqbw@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <511D99B1.40501@devio.us>
 <CA+qr9e6tSpZiW8sHPx_03HPF3TQ76yzns1YuUtbu00LXLenqbw@mail.gmail.com>
Message-ID: <CAF6rxgmEr7EHoivj6j0vFFTZduk1JSgiTXxaCr73OJBPo89c0A@mail.gmail.com>

On 14 February 2013 22:00, Marc Spitzer <mspitzer at gmail.com> wrote:>
>
> Since binary patching is how things are getting done for the os patching,
> why should there be a tool in base to grab source?

There *is* a base-tool to obtain sources for releases:  freebsd-update

The config file has

 14 # Components of the base system which should be kept updated.
 15 Components src world kernel


-- 
Eitan Adler


From nycbug at wynn.com  Fri Feb 15 00:35:57 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Fri, 15 Feb 2013 00:35:57 -0500
Subject: [nycbug-talk] Fw: BIND 10 - 1.0.0 Release Candidate
Message-ID: <20130215003557.6008c593@ivory.wynn.com>



Begin forwarded message:

Date: Thu, 14 Feb 2013 21:25:44 -0600 (CST)
From: "Jeremy C. Reed" <jreed at isc.org>
To: staff at isc.org
Subject: BIND 10 - 1.0.0 Release Candidate


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

	BIND 10 - 1.0.0 Release Candidate

Welcome to the first release candidate toward the first production
BIND 10 1.0.0 release.  BIND 10 provides a C++ library for DNS
(with python wrappers) and several cooperating daemons for providing
authoritative DNS service (with in-memory and SQLite3 backends and
DNSSEC support), dynamic DNS, zone transfers, and experimental
forwarding and recursive name service.  Supplementary components
are included for statistics collection and reporting and remote
configuration and control.

This version of BIND 10 also includes the latest snapshot of the
BIND 10 DHCP development.  The snapshot includes a C++ library for
DHCP and two DHCP servers, one for IPv4 and one for IPv6. Features
of these servers are:

* Able to allocate and renew addresses, and handle lease expiration
  and releases.
* Supports a subset of clients:
  - DHCPv4 clients connected to the server via a relay.
  - DHCPv6 clients on the same LAN as the server.
* Able to configure values for standard options returned to a client,
  either globally or on a per-subnet basis.
* Able to define new options and configure them in the same way as
  standard options.
* Leases are stored in a MySQL database.
* Configuration, logging and process control uses the same mechanisms
  as the BIND 10 DNS server.

Note: The default testing account and password for bindctl/b10-cmdctl
is now removed; a new account for remote configuration and control
can be created with b10-cmdctl-usermgr, for example:
	b10-cmdctl-usermgr
--file /usr/local/etc/bind10/cmdctl-accounts.csv

We are looking for testers to provide feedback about using this
release candidate. For more information about BIND 10, the release
schedule, and the community testing plans, please see:

	http://bind10.isc.org/wiki/ProductionRelease

Documentation is included and also available via the BIND 10 website
at http://bind10.isc.org/

The bind10-1.0.0-rc source may be downloaded from:

	ftp://ftp.isc.org/isc/bind10/1.0.0-rc/bind10-1.0.0-rc.tar.gz

A PGP signature of the distribution is at

	ftp://ftp.isc.org/isc/bind10/1.0.0-rc/bind10-1.0.0-rc.tar.gz.sha512.asc

The signature was generated with the ISC code signing key which is
available at https://www.isc.org/about/openpgp

A summary of the significant changes since the previous release
include (from the ChangeLog):

580.	[func]*		muks
	There is no longer a default user account. The old default
account with username 'root' has been removed. In a fresh installation
of BIND 10, the administrator has to configure a user account using
	the b10-cmdctl-usermgr program.
	(Trac #2641, git 54e8f4061f92c2f9e5b8564240937515efa6d934)

579.	[bug]		jinmei
	libdatasrc/b10-auth: corrected some corner cases in query
handling of in-memory data source that led to the following invalid/odd
	responses from b10-auth:
	- duplicate RRs in answer and additional for type ANY query
	- incorrect NSEC for no error, no data (NXRRSET) response that
	  matches a wildcard
	(Trac #2585, git abe78fae4ba3aca5eb01806dd4e05607b1241745)

578.	[bug]		jinmei
	b10-auth now returns closest encloser NSEC3 proof to queries for
	an empty non terminal derived from an Opt-Out NSEC RR, as
clarified in errata 3441 for RFC5155.  Previously it regarded such case
as broken zone and returned SERVFAIL.
	(Trac #2659, git 24c235cb1b379c6472772d340e21577c3460b742)

577.	[func]		muks
	Added an SQLite3 index on records(rname, rdtype). This decreases
	insert performance by ~28% and adds about ~20% to the file size,
	but increases zone iteration performance. As it introduces a new
	index, a database upgrade would be required.
	(Trac #1756, git 9b3c959af13111af1fa248c5010aa33ee7e307ee)

576.	[bug]		tmark, tomek
	b10-dhcp6: Fixed bug when the server aborts operation when
	receiving renew and there are no IPv6 subnets configured.
	(Trac #2719, git 3132b8b19495470bbfd0f2ba0fe7da443926034b)

575.	[bug]		marcin
	b10-dhcp6: Fixed the bug whereby the subnet for the incoming
	packet was selected using only its source address. The subnet
	is now selected using either source address or the name of the
	server's interface on which the packet has been received.
	(Trac #2704, git 1cbacf19a28bdae50bb9bd3767bca0147fde37ed)

574.	[func]		tmark
	b10-dhcp4, b10-dhcp6: Composite key indexes were added to the
lease tables to reduce lease search time. The lease4 table now has two
	additional indexes: a) hwaddr/subnet_id and b)
client_id/subnet_id. The lease6 now has the one additional index:
iaid/subnet_id/duid. Adding these indexes significantly improves lease
acquisition performance.
	(Trac #2699,#2703, git 54bbed5fcbe237c5a49b515ae4c55148723406ce)

573.	[bug]		stephen
	Fixed problem whereby the DHCP server crashed if it ran out of
	addresses.  Such a condition now causes a packet to be returned
	to the client refusing the allocation of an address.
	(Trac #2681, git 87ce14cdb121b37afb5b1931af51bed7f6323dd6)

572.	[bug]		marcin
	perfdhcp: Fixed bug where the command line switches used to
	run the perfdhcp where printed as ASCII codes.
	(Trac #2700, git b8d6b949eb7f4705e32fbdfd7694ca2e6a6a5cdc)

571.	[build]		jinmei
	The ./configure script can now handle output from python-config
	--ldflags that contains a space after -L switches.  This fixes
	failure reported on some Solaris environments.
	(Trac #2661, git e6f86f2f5eec8e6003c13d36804a767a840d96d6)

570.	[bug]		tmark, marcin, tomek
	b10-dhcp4: Address renewal now works properly for DHCPv4 clients
	that do not send client ID.
	(Trac #2702, git daf2abe68ce9c111334a15c14e440730f3a085e2)

569.	[bug]		tomek
	b10-dhcp4: Fix bug whereby a DHCP packet without a client ID
	could crash the MySQL lease database backend.
	(Trac #2697, git b5e2be95d21ed750ad7cf5e15de2058aa8bc45f4)

568.	[func]		muks
	Various message IDs have been renamed to remove the word 'ERROR'
	from them when they are not logged at ERROR severity level.
	(Trac #2672, git 660a0d164feaf055677f375977f7ed327ead893e)

567.	[doc]		marcin, stephen, tomek
	Update DHCP sections of the BIND 10 guide.
	(Trac #2657, git 1d0c2004865d1bf322bf78d13630d992e39179fd)

566.	[func]*		jinmei
	libdns++/Python isc.dns: In Python isc.dns, function style
	constants for RRType, RRClass, Rcode and Opcode were deprecated
	and replaced with straightforward object constants, e.g., from
	RRType.AAAA() to RRType.AAAA.  This is a backward incompatible
	change (see the Trac ticket for a conversion script if needed).
	Also, these constants are now more consistent between C++
	and Python, and RRType constants for all currently standardized
	types are now supported (even if Rdata for these are not yet
	available).
	(Trac #1866 and #2409, git
e5005185351cf73d4a611407c2cfcd163f80e428)

565.	[func]*		jelte
	The main initializer script (formerly known as either 'bind10',
	'boss', or 'bob'), has been renamed to b10-init (and Init in
	configuration). Configuring which components are run is
henceforth done through '/Init/components', and the sbin/bind10 script
is now simply a shellscript that runs b10-init. Existing configuration
is automatically updated. NOTE: once configuration with this update
	has been saved (by committing any new change with bindctl), you
	cannot run older versions of BIND 10 anymore with this
configuration. (Trac #1901, git
bae3798603affdb276f370c1ac6b33b011a5ed4f)

564.	[func]		muks
	libdns++: the CNAME, DNAME, MX, NS, PTR and SRV Rdata classes
now use the generic lexer in constructors from text.  This means that
	the name fields in such RRs in a zone file can now be
non-absolute (the origin name in that context will be used), e.g., when
loaded by b10-loadzone. One additional change to the libdns++ API is
that the existing string constructors for these Rdata classes also use
	the generic lexer, and they now expect an absolute name (with
the trailing '.') in the name fields.
	(Trac #2390, git a01569277cda3f78b1171bbf79f15ecf502e81e2)
	(Trac #2656, git 5a0d055137287f81e23fbeedd35236fee274596d)

563.	[build]		jinmei
	Added --disable-rpath configure option to avoid embedding
library paths to binaries.  Patch from Adam Tkac.
	(Trac #2667, git 1c50c5a6ee7e9675e3ab154f2c7f975ef519fca2)

562.	[func]*		vorner
	The b10-xfrin now performs basic sanity check on just received
	zone. It'll reject severely broken zones (such as missing NS
	records).
	(Trac #2439, git 44699b4b18162581cd1dd39be5fb76ca536012e6)

561.	[bug]		kambe, jelte
	b10-stats-httpd no longer dumps request information to the
console, but uses the bind10 logging system. Additionally, the logging
	identifiers have been changed from STATHTTPD_* to STATSHTTPD_*
	(Trac #1897, git 93716b025a4755a8a2cbf250a9e4187741dbc9bb)

560.	[bug]		jinmei
	b10-auth now sets the TTL of SOA RR for negative responses to
	the minimum of the RR TTL and the minimum TTL of the SOA RDATA
	as specified in RFC2308; previously the RR TTL was always used.
	The ZoneFinder class was extended partly for implementing this
	and partly for allowing further optimization.
	(Trac #2309 and #2635, git
ee17e979fcde48b59d91c74ac368244169065f3b)

559.	[bug]		jelte
	b10-cmdctl no longer aborts on basic file issues with its https
	certificate or private key file. It performs additional checks,
and provides better error logs if these fail. Additionally, bindctl
	provides a better error report if it is unable to connect over
	https connection. This issue could occur if BIND 10 was
installed with root privileges but then started as a normal user.
	(Trac #2595, git 09b1a2f927483b407d70e98f5982f424cc872149)

558.	[func]		marcin
	b10-dhcp4: server now adds configured options to its
	responses to a client when client requests them.
	A few basic options: Routers, Domain Name, Domain
	Name Servers and Subnet Mask are added regardless
	if client requested them or not.
	(Trac #2591, git aeec2dc1b9c511d17971ac63138576c37e7c5164)

557.	[doc]		stephen
	Update DHCP sections of the BIND 10 guide.
	(Trac #2642, git e5faeb5fa84b7218fde486347359504cf692510e)

556.	[bug]		marcin
	Fixed DHCP servers configuration whereby the servers did not
	receive a configuration stored in the database on their startup.
	Also, the configuration handler function now uses full
configuration instead of partial to configure the server. This
guarantees that dependencies between various configuration parameters
are fulfilled.
	(Trac #2637, git 91aa998226f1f91a232f2be59a53c9568c4ece77)

555.	[func]		marcin
	The encapsulated option space name can be specified for
	a DHCP option. It comprises sub-options being sent within
	an option that encapsulates this option space.
	(Trac #2314, git 27e6119093723a1e46a239ec245a8b4b10677635)

554.	[func]		jinmei
	b10-loadzone: improved completion log message and intermediate
	reports: It now logs the precise number of loaded RRs on
	completion, and intermediate reports show additional information
	such as the estimated progress in percentage and estimated time
	to complete.
	(Trac #2574, git 5b8a824054313bdecb8988b46e55cb2e94cb2d6c)

553.	[func]		stephen
	Values of the parameters to access the DHCP server lease
database can now be set through the BIND 10 configuration mechanism.
	(Trac #2559, git 6c6f405188cc02d2358e114c33daff58edabd52a)

552.	[bug]		shane
	Build on Raspberry PI.
	The main issue was use of char for reading from input streams,
	which is incorrect, as EOF is returned as an int -1, which would
	then get cast into a char -1.
	A number of other minor issues were also fixed.
	(Trac #2571, git 525333e187cc4bbbbde288105c9582c1024caa4a)

551.	[bug]		shane
	Kill msgq if we cannot connect to it on startup.
	When the boss process was unable to connect to the msgq, it
would exit. However, it would leave the msgq process running. This has
	been fixed, and the msgq is now stopped in this case.
	(Trac #2608, git 016925ef2437e0396127e135c937d3a55539d224)

550.	[func]		tomek
	b10-dhcp4: The DHCPv4 server now generates a server identifier
	the first time it is run. The identifier is preserved in a file
	across server restarts.
	b10-dhcp6: The server identifier is now preserved in a file
across server restarts.
	(Trac #2597, git fa342a994de5dbefe32996be7eebe58f6304cff7)

549.	[func]		tomek
	b10-dhcp6: It is now possible to specify that a configured
subnet is reachable locally over specified interface (see "interface"
	parameter in Subnet6 configuration).
	(Trac #2596, git a70f6172194a976b514cd7d67ce097bbca3c2798)

548.	[func]		vorner
	The message queue daemon now appears on the bus. This has two
	effects, one is it obeys logging configuration and logs to the
	correct place like the rest of the modules. The other is it
	appears in bindctl as module (but it doesn't have any commands
or configuration yet).
	(Trac #2582, git ced31d8c5a0f2ca930b976d3caecfc24fc04634e)

547.	[func]*		vorner
	The b10-loadzone now performs more thorough sanity check on the
	loaded data.  Some of the checks are now fatal and zone failing
	them will be rejected.
	(Trac #2436, git 48d999f1cb59f308f9f30ba2639521d2a5a85baa)

546.	[func]		marcin
	DHCP option definitions can be now created using the
	Configuration Manager. The option definition specifies
	the option code, name and the types of the data being
	carried by the option.  The Configuration Manager
	reports an error on attempt to override standard DHCP
	option definition.
	(Trac #2317, git 71e25eb81e58a695cf3bad465c4254b13a50696e)

545.	[func]		jinmei
	libdns++: the SOA Rdata class now uses the generic lexer in
	constructors from text.  This means that the MNAME and RNAME of
an SOA RR in a zone file can now be non absolute (the origin name
	in that context will be used), e.g., when loaded by
b10-loadzone. (Trac #2500, git 019ca218027a218921519f205139b96025df2bb5)

544.	[func]		tomek
	b10-dhcp4: Allocation engine support for IPv4 added. Currently
	supported operations are server selection (Discover/Offer),
	address assignment (Request/Ack), address renewal (Request/Ack),
	and address release (Release). Expired leases can be reused.
	Some options (e.g. Router Option) are still hardcoded, so the
	DHCPv4 server is not yet usable, although its address allocation
	is operational.
	(Trac #2320, git 60606cabb1c9584700b1f642bf2af21a35c64573)

543.	[func]*		jelte
	When calling getFullConfig() as a module, , the configuration
is now returned as properly-structured JSON.  Previously, the structure
had been flattened, with all data being labelled by fully-qualified
	element names.
	(Trac #2619, git bed3c88c25ea8f7e951317775e99ebce3340ca22)

542.	[func]		marcin
	Created OptionSpace and OptionSpace6 classes to represent DHCP
	option spaces. The option spaces are used to group instances
	and definitions of options having uniqe codes. A special type
	of option space is the so-called "vendor specific option space"
	which groups sub-options sent within Vendor Encapsulated
Options. The new classes are not used yet but they will be used once
	the creation of option spaces by configuration manager is
	implemented.
	(Trac #2313, git 37a27e19be874725ea3d560065e5591a845daa89)

541.	[func]		marcin
	Added routines to search for configured DHCP options and their
	definitions using name of the option space they belong to.
	New routines are called internally from the DHCPv4 and DHCPv6
	servers code.
	(Trac #2315, git 741fe7bc96c70df35d9a79016b0aa1488e9b3ac8)

540.	[func]		marcin
	DHCP Option values can be now specified using a string of
	tokens separated with comma sign. Subsequent tokens are used
	to set values for corresponding data fields in a particular
	DHCP option. The format of the token matches the data type
	of the corresponding option field: e.g. "192.168.2.1" for IPv4
	address, "5" for integer value etc.
	(Trac #2545, git 792c129a0785c73dd28fd96a8f1439fe6534a3f1)

539.	[func]		stephen
	Add logging to the DHCP server library.
	(Trac #2524, git b55b8b6686cc80eed41793c53d1779f4de3e9e3c)

538.	[bug]		muks
	Added escaping of special characters (double-quotes, semicolon,
	backslash, etc.) in text-like RRType's toText() implementation.
	Without this change, some TXT and SPF RDATA were incorrectly
	stored in SQLite3 datasource as they were not escaped.
	(Trac #2535, git f516fc484544b7e08475947d6945bc87636d4115)

537.	[func]		tomek
	b10-dhcp6: Support for RELEASE message has been added. Clients
	are now able to release their non-temporary IPv6 addresses.
	(Trac #2326, git 0974318566abe08d0702ddd185156842c6642424)

536.	[build]		jinmei
	Detect a build issue on FreeBSD with g++ 4.2 and Boost
installed via FreeBSD ports at ./configure time.  This seems to be a
bug of FreeBSD	ports setup and has been reported to the
maintainer: http://www.freebsd.org/cgi/query-pr.cgi?pr=174753
	Until it's fixed, you need to build BIND 10 for FreeBSD that has
	this problem with specifying --without-werror, with clang++
	(development version), or with manually extracted Boost header
	files (no compiled Boost library is necessary).
	(Trac #1991, git 6b045bcd1f9613e3835551cdebd2616ea8319a36)

535.	[bug]		jelte
	The log4cplus internal logging mechanism has been disabled, and
no output from the log4cplus library itself should be printed to
	stderr anymore. This output can be enabled by using the
	compile-time option --enable-debug.
	(Trac #1081, git db55f102b30e76b72b134cbd77bd183cd01f95c0)

534.	[func]*		vorner
	The b10-msgq now uses the same logging format as the rest
	of the system. However, it still doesn't obey the common
	configuration, as due to technical issues it is not able
	to read it yet.
	(git 9e6e821c0a33aab0cd0e70e51059d9a2761f76bb)

Thanks again to those who contributed bug reports, code, and reviews.
Bugs may be reported as tickets via the developers website (after
logging into Trac) at:

	http://bind10.isc.org/

Please feel free to participate and share your feedback on the BIND
10 mailing lists:

	https://lists.isc.org/mailman/listinfo/bind10-users
	https://lists.isc.org/mailman/listinfo/bind10-dev

Jeremy C. Reed
ISC Release Engineering

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (NetBSD)

iEYEARECAAYFAlEdqlYACgkQs9Bv5D4YwC3t9QCdFmHE9bVZq0WRa4E1pq5t1JtK
CMgAoNTXHYMMlvMU6bzARXBOsgYq2ZW5
=JulM
-----END PGP SIGNATURE-----


-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

I would never invade the United States.  There would be a gun behind
every blade of grass.  --Isoroku Yamamoto



From nycbug at wynn.com  Fri Feb 15 00:58:38 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Fri, 15 Feb 2013 00:58:38 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <511D488D.1030209@devio.us>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
Message-ID: <20130215005838.392d0c54@ivory.wynn.com>

On Thu, 14 Feb 2013 15:26:53 -0500
Brian Callahan <bcallah at devio.us> wrote:

Much of Brian's very good insight and explanation of my thought process
on this issue snipped.

> FWIW, I'd take Brett's suggestion of a basesnap program and run with
> it. Fbsd should already have the infrastructure necessary to do this
> (I imagine it would be similar to "do whatever you do for portsnap in
> the ports dir, and do this for the base dir").
> 
> ~Brian

Brian I am glad you get it!  I hope others will get it as well and that
the FreeBSD Project starts to mitigate the pain for users and not
increase it as has been the course the last few years.

I am about to contribute my first port, but I am also thinking of
moving to something else.  If the pain continues my ARM boards will go
to some version of GNU/Linux because I do not think any BSD is as far
along on the BeagleBone and Raspberry Pi as FreeBSD.

I think we have some folks on this list that are somewhat closer to the
center of the FreeBSD universe than I am.  Please take my ideas and
thoughts to others that have the ability to influence the direction on
the SVN debacle.

-Brett


-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

I would never invade the United States.  There would be a gun behind
every blade of grass.  --Isoroku Yamamoto



From nycbug at wynn.com  Fri Feb 15 01:09:33 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Fri, 15 Feb 2013 01:09:33 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us>
 <20130214134902.2e532e5d@ivory.wynn.com>
 <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>
Message-ID: <20130215010933.551e994b@ivory.wynn.com>

On Thu, 14 Feb 2013 15:55:26 -0500
Jesse Callaway <bonsaime at gmail.com> wrote:

> It should be easy enough to set up an https service with the source
> you need which is kept in sync with the FreeBSD source.
> 
> Not sure what is so difficult or controversial about this. I usually
> do this sort of a thing when I have more than 10 boxes to take care
> of, just out of bandwidth and speed concerns.

Jesse-

You just do not get it.  I am an admin with 30+ years in the field.  I
can do any of the various hackenstein solutions that you or anyone else
has come up with, and I probably have come up with them before you, but
that does not address the real problem which Brian obviously did
understand.  The FreeBSD Project has been making things harder rather
than easier for the last several years.

I am saying the project needs a non-hackenstien solution like it had
before.  

So I say again someone better than me needs to write a basesnap utility
and the server infrastructure needs to be put in place.  This should be
pretty trivial considering it will be basically a clone of portsnap!

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

April 19, 1775 An English attempt to confiscate guns from Americans
triggered a successful revolution......

     Dear Congress, that's a hint.


From george at ceetonetechnology.com  Fri Feb 15 01:18:07 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 15 Feb 2013 01:18:07 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <20130215010933.551e994b@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us> <20130214134902.2e532e5d@ivory.wynn.com>
 <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>
 <20130215010933.551e994b@ivory.wynn.com>
Message-ID: <511DD31F.1070606@ceetonetechnology.com>

On 02/15/13 01:09, Brett Wynkoop wrote:
> On Thu, 14 Feb 2013 15:55:26 -0500
> Jesse Callaway <bonsaime at gmail.com> wrote:
>
>> It should be easy enough to set up an https service with the source
>> you need which is kept in sync with the FreeBSD source.
>>
>> Not sure what is so difficult or controversial about this. I usually
>> do this sort of a thing when I have more than 10 boxes to take care
>> of, just out of bandwidth and speed concerns.
>
> Jesse-
>
> You just do not get it.  I am an admin with 30+ years in the field.  I
> can do any of the various hackenstein solutions that you or anyone else
> has come up with, and I probably have come up with them before you, but
> that does not address the real problem which Brian obviously did
> understand.  The FreeBSD Project has been making things harder rather
> than easier for the last several years.
>
> I am saying the project needs a non-hackenstien solution like it had
> before.
>
> So I say again someone better than me needs to write a basesnap utility
> and the server infrastructure needs to be put in place.  This should be
> pretty trivial considering it will be basically a clone of portsnap!

Let's keep the tone a bit more pleasant, everyone.

g



From nycbug at wynn.com  Fri Feb 15 01:35:50 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Fri, 15 Feb 2013 01:35:50 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
Message-ID: <20130215013550.7b4fb18e@ivory.wynn.com>

On Thu, 14 Feb 2013 18:28:20 -0500
Marc Spitzer <mspitzer at gmail.com> wrote:

> On Thu, Feb 14, 2013 at 3:26 PM, Brian Callahan <bcallah at devio.us>
> wrote:

SNIP SNIP
> 
> How is source access not easy?  just install svn and you have it.  Is
> there a hard step I am missing?

Let's start with....maybe you do not have /usr/ports or
want /usr/ports.  It is not a good practice to force the user to
install ports to be able to update base.  I had the same complaint with
cvs years ago because you had to install ports to get either full  cvs
or cvsup before you could update base.  cvs and cvsup were also both
largish compared to just base, especially when ports gets tossed in
just to get it. 

This issue was solved with the introduction of csup in base.  

> 

> again install svn and done, ie cd /usr/ports/devel/subversion && make
> && make install is the hard way.  To be honest I have small hope of
> anyone becoming a developer if svn is a show stopper.

I will buy you a clue here:


wynkoop at beaglebone:~ % uname -a
FreeBSD beaglebone 10.0-CURRENT FreeBSD 10.0-CURRENT #3: Mon Feb 11 18:24:59 EST 2013     root at beaglebone:/sys/arm/compile/BEAGLEBONE-DEBUG  arm

wynkoop at beaglebone:~ % df -h
Filesystem        Size    Used   Avail Capacity  Mounted on
/dev/mmcsd0s2a    7.2G    3.4G    3.2G    51%    /
devfs             1.0k    1.0k      0B   100%    /dev
devfs             1.0k    1.0k      0B   100%    /var/named/dev
wynkoop at beaglebone:~ % 


I want to take up a huge bunch of space on my disk so I can do updates
to base?  I do not think so.

> 
> 
> 
> >
> > If even one person (/business/manufacturer/**whatever) skips out on
> > fbsd because of this svn debacle, it should be considered a massive
> > failure in the eyes of the project and the devs, because who knows
> > - that one person could have gone on to become the next core team
> > member.
> >

He is right.  There is no telling what lost opportunities and talent
will come from the core developers not thinking of the outer layers of
the community as broadly as they should.

> 
> one person is not a failure, let alone a debacle.  I think freebsd
> made the right decision for freebsd from the available choices as I
> understand it. the community of freebsd devs made the decision for
> valid reasons as believed by the people who get to decide and some
> people are not happy about it and blowing it out of proportion, well
> ok life goes on.

No one is arguing that svn is the wrong tool for full developer
access.  The point is that for non-core developers that need source for
updates it should not require almost as much disk space as the core of
the OS and it should not require the installation of anything NOT
INCLUDED IN BASE!  It should also be easy to use.


> to be honest I think that the binary updates are vastly superior to
> source updates, I get check sums, reinstall/clone is faster and when
> build tools show up in my web server I just may have a problem.
> 
> marc

So marc are you saying you are going to start doing daily builds of head
for ARM including each different board?  In the ARM world there is no
such thing as a GENERIC kernel that will run on all arm boards.  The Pi
can not boot the Bone kernel and it is more complex than just making
sure the kernel is driver overloaded!

How do I access the binary updates you are building for ARM?

-Brett


-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

April 19, 1775 An English attempt to confiscate guns from Americans
triggered a successful revolution......

     Dear Congress, that's a hint.


From nycbug at wynn.com  Fri Feb 15 01:42:19 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Fri, 15 Feb 2013 01:42:19 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e6tSpZiW8sHPx_03HPF3TQ76yzns1YuUtbu00LXLenqbw@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <511D99B1.40501@devio.us>
 <CA+qr9e6tSpZiW8sHPx_03HPF3TQ76yzns1YuUtbu00LXLenqbw@mail.gmail.com>
Message-ID: <20130215014219.7fc9ac47@ivory.wynn.com>

On Thu, 14 Feb 2013 22:00:52 -0500
Marc Spitzer <mspitzer at gmail.com> wrote:

> Since binary patching is how things are getting done for the os
> patching, why should there be a tool in base to grab source?  and you
> can download the system source code using wget or curl, unless they
> are also too big to install on your system.  Source is not as
> critical to operational management of your system as it used to be,
> things change.
> 

Please show me the binary patches for ARM.  I am sure there are other
processors in the same boat as ARM where binary updates not really
there.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

"The strongest reason for the people to retain the right to keep 
and bear arms is, as a last resort, to protect themselves against 
tyranny in government" - Thomas Jefferson. 



From mwlucas at blackhelicopters.org  Fri Feb 15 08:39:29 2013
From: mwlucas at blackhelicopters.org (Michael W. Lucas)
Date: Fri, 15 Feb 2013 08:39:29 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <20130215014219.7fc9ac47@ivory.wynn.com>
References: <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <511D99B1.40501@devio.us>
 <CA+qr9e6tSpZiW8sHPx_03HPF3TQ76yzns1YuUtbu00LXLenqbw@mail.gmail.com>
 <20130215014219.7fc9ac47@ivory.wynn.com>
Message-ID: <20130215133929.GB33094@bewilderbeast.blackhelicopters.org>

I might have missed this, so apologies if it's a repeat:

What's wrong with 'freebsd-update src'? Supported, in base, the first
download is rather slow but after that it's quite fast.

Admittedly, there is no freebsd-update for arm yet. But arm is a Tier
2 platform
(http://www.freebsd.org/doc/en/articles/committers-guide/archs.html). Tier
2 platforms require the use of development tools, as they are still
under development. Before arm can be Tier 1, it must have
freebsd-update, binary patches, etc. arm users are in the same boat as
powerpc and sparc64 users.

For over a decade, FreeBSD used cvsup instead of csup. Written in
Modula-3. If you think svn is bad, just remember building Modula-3 on
a 25MHz CPU...

==ml


-- 
Michael W. Lucas  -  mwlucas at michaelwlucas.com, Twitter @mwlauthor 
http://www.MichaelWLucas.com/, http://blather.MichaelWLucas.com/
Latest book: Absolute OpenBSD 2/e - http://www.nostarch.com/openbsd2e
coupon code "ILUVMICHAEL" gets you 30% off & helps me.


From mspitzer at gmail.com  Fri Feb 15 11:15:45 2013
From: mspitzer at gmail.com (Marc Spitzer)
Date: Fri, 15 Feb 2013 11:15:45 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130215013550.7b4fb18e@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <20130215013550.7b4fb18e@ivory.wynn.com>
Message-ID: <CA+qr9e66_gMh1tK29fHzdF0X=F+zZ7UtdsukuJJoKBDW-tCWcg@mail.gmail.com>

On Fri, Feb 15, 2013 at 1:35 AM, Brett Wynkoop <nycbug at wynn.com> wrote:

> On Thu, 14 Feb 2013 18:28:20 -0500
> Marc Spitzer <mspitzer at gmail.com> wrote:
>
> > On Thu, Feb 14, 2013 at 3:26 PM, Brian Callahan <bcallah at devio.us>
> > wrote:
>
> SNIP SNIP
> >
> > How is source access not easy?  just install svn and you have it.  Is
> > there a hard step I am missing?
>
> Let's start with....maybe you do not have /usr/ports or
> want /usr/ports.  It is not a good practice to force the user to
> install ports to be able to update base.  I had the same complaint with
> cvs years ago because you had to install ports to get either full  cvs
> or cvsup before you could update base.  cvs and cvsup were also both
> largish compared to just base, especially when ports gets tossed in
> just to get it.


> This issue was solved with the introduction of csup in base.
>

Well csup was a very new thing in base, it used to be in ports as cvsup and
written in modula-3.  And I agree with you that it is not a good thing to
not force users to install ports to upgrade their system nor in the general
to actually expect them to always compile their system from scratch, that
is why freebsd-update and binary patching are a big win for
the community because it makes the system more maintainable and much
more scaleable in production.  Comparing those big wins to you are not
happy, and a few other people may not be happy, I am fine with you are not
happy.


>
> >
>
> > again install svn and done, ie cd /usr/ports/devel/subversion && make
> > && make install is the hard way.  To be honest I have small hope of
> > anyone becoming a developer if svn is a show stopper.
>
> I will buy you a clue here:
>
>
> wynkoop at beaglebone:~ % uname -a
> FreeBSD beaglebone 10.0-CURRENT FreeBSD 10.0-CURRENT #3: Mon Feb 11
> 18:24:59 EST 2013     root at beaglebone:/sys/arm/compile/BEAGLEBONE-DEBUG
>  arm
>
> wynkoop at beaglebone:~ % df -h
> Filesystem        Size    Used   Avail Capacity  Mounted on
> /dev/mmcsd0s2a    7.2G    3.4G    3.2G    51%    /
> devfs             1.0k    1.0k      0B   100%    /dev
> devfs             1.0k    1.0k      0B   100%    /var/named/dev
> wynkoop at beaglebone:~ %
>
>
clue for clue: running head is not operations, it is not production, it is
development.  now how to actually fix your problem:  Put svn on a box that
these other boxes that you can put dev tools, you need only one.  Then
update that box using svn and scp over a src tarball, untar it and build
your world.


>
> I want to take up a huge bunch of space on my disk so I can do updates
> to base?  I do not think so.
>

see above simple solution that does not even require csup, now that there
is a solution that meets your criteria of only using the base system on
these tiny ARM boxes you are happy and can do exactly what you are saying
you want to do.  Glad to be of service


>
> >
> >
> >
> > >
> > > If even one person (/business/manufacturer/**whatever) skips out on
> > > fbsd because of this svn debacle, it should be considered a massive
> > > failure in the eyes of the project and the devs, because who knows
> > > - that one person could have gone on to become the next core team
> > > member.
> > >
>
> He is right.  There is no telling what lost opportunities and talent
> will come from the core developers not thinking of the outer layers of
> the community as broadly as they should.
>

Well there is absolutely no telling either way, how many people/businesses
will walk away from the project or never use it because I have to muck
around with source on all my servers?


>
> >
> > one person is not a failure, let alone a debacle.  I think freebsd
> > made the right decision for freebsd from the available choices as I
> > understand it. the community of freebsd devs made the decision for
> > valid reasons as believed by the people who get to decide and some
> > people are not happy about it and blowing it out of proportion, well
> > ok life goes on.
>
> No one is arguing that svn is the wrong tool for full developer
> access.  The point is that for non-core developers that need source for
> updates it should not require almost as much disk space as the core of
> the OS and it should not require the installation of anything NOT
> INCLUDED IN BASE!  It should also be easy to use.
>

we disagree, you are fully entitled to your opinion.  The thing is I
just don't see anything other then "I want it" backing it up.  Now some
other people may agree with you, but I have not seen a case made by you
except "I want it" with the occasional argument to authority "I have been
doing it for 30 years" and people who agree with you are smart
and conversely people who disagree with you are clueless.  To put it simply
this is not a convincing or compelling argument to change anyone's mind.
 If you really want the change the way things are you need to develop a
better argument.



>
>
> > to be honest I think that the binary updates are vastly superior to
> > source updates, I get check sums, reinstall/clone is faster and when
> > build tools show up in my web server I just may have a problem.
> >
> > marc
>
> So marc are you saying you are going to start doing daily builds of head
> for ARM including each different board?  In the ARM world there is no
> such thing as a GENERIC kernel that will run on all arm boards.  The Pi
> can not boot the Bone kernel and it is more complex than just making
> sure the kernel is driver overloaded!
>
> How do I access the binary updates you are building for ARM?
>
>
To be honest at the present time I could care less about ARM boxes, they
are still in alpha near as I can tell from a real use point of view.
 People who enjoy playing with them should have fun with them and if you
are doing long term research for work great enjoy the ride.

now people have posted several reasonable solutions to your choice to not
put dev tools on your dev boxes, because you choose not to.  You have
posted no compelling arguments to support your point and you seem to be
unwilling to listen to other peoples POV or actually solve your problem by
using any of the solutions posted.  So why should I care about what you
want?

marc

-- 
Freedom is nothing but a chance to be better.
--Albert Camus

The inherent vice of capitalism is the unequal sharing of blessings; the
inherent virtue of socialism is the equal sharing of miseries.
-- Winston Churchill

Do the arithmetic or be doomed to talk nonsense.
--John McCarthy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130215/003b8b1e/attachment.htm>

From akosela at andykosela.com  Fri Feb 15 12:23:11 2013
From: akosela at andykosela.com (Andy Kosela)
Date: Fri, 15 Feb 2013 11:23:11 -0600
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130215010933.551e994b@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us>
 <20130214134902.2e532e5d@ivory.wynn.com>
 <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>
 <20130215010933.551e994b@ivory.wynn.com>
Message-ID: <CALMnNGgV2FatW_Q_MhYFX1D+KNhXwtSCu_ze4De8aCfMKE_L=A@mail.gmail.com>

On Fri, Feb 15, 2013 at 12:09 AM, Brett Wynkoop <nycbug at wynn.com> wrote:
> So I say again someone better than me needs to write a basesnap utility
> and the server infrastructure needs to be put in place.  This should be
> pretty trivial considering it will be basically a clone of portsnap!

There is a saying in the Open Source world and definetly you know it
too.  You want something done the way you see it?  You better write it
yourself...

svnup in the base would be nice though.

--Andy


From george at ceetonetechnology.com  Fri Feb 15 12:24:10 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 15 Feb 2013 12:24:10 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CA+qr9e66_gMh1tK29fHzdF0X=F+zZ7UtdsukuJJoKBDW-tCWcg@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CA+qr9e4o8TEWcQXwX50SoW2a1CqgR+9y+OnmG92zvqNzCYMmNQ@mail.gmail.com>
 <20130214140225.480aa649@ivory.wynn.com>
 <CA+qr9e6QraiyZBWt4aWT8m+5jKPhRGvJ83bkEiGVeXEkJdFm0Q@mail.gmail.com>
 <511D488D.1030209@devio.us>
 <CA+qr9e4AQc8ThYSS5a1ZC+VH7k8f8hHdaSSUdLkYv98WPJykFg@mail.gmail.com>
 <20130215013550.7b4fb18e@ivory.wynn.com>
 <CA+qr9e66_gMh1tK29fHzdF0X=F+zZ7UtdsukuJJoKBDW-tCWcg@mail.gmail.com>
Message-ID: <511E6F3A.6030506@ceetonetechnology.com>

On 02/15/13 11:15, Marc Spitzer wrote:
> On Fri, Feb 15, 2013 at 1:35 AM, Brett Wynkoop <nycbug at wynn.com> wrote:
>
>> On Thu, 14 Feb 2013 18:28:20 -0500
>> Marc Spitzer <mspitzer at gmail.com> wrote:
>>
>>> On Thu, Feb 14, 2013 at 3:26 PM, Brian Callahan <bcallah at devio.us>
>>> wrote:
>>

<large snip>

So I had a long response sitting in drafts yesterday when this 
discussion came up, but as I was a bit tied up, it became less and less 
relevant as the thread developed.

First thing, there's no reason to have a nasty tone about this.  This 
applies to everyone in the discussion.  Don't be condescending.  We 
don't normally have that tone on our list, and I was tempted to just 
start booting a few people off the list last night to make a point.

I have a hard time believing most of you would be so nasty face-to-face. 
  If you can't discuss in a public forum like this, then keep your mouth 
shut.

Now, to the issue, without directly addressing any particular comment.

I am very much on the frustrated side with the changes to FreeBSD with 
dropping a little base utility called csup and replacing with 
subversion.  Yes, the benefits of subversion, blah blah blah, yes, 
freebsd-update, blah blah blah, but FreeBSD is going through massive 
changes, and it gets a bit uneasy for those of us set in their ways.

I remember building my first Linux box way way back, and was horrified 
that somehow I had Sendmail and Apache running.  Was I running an open 
relay?  This was how Linux "intimidated" me into BSD-land, after getting 
my feet wet with Sun.

I installed OpenBSD and top was so empty, that it was cleanly viewable 
on a 12" CRT.  I was overjoyed, to say the least.

Now FreeBSD is in a situation where three branches are officially 
supported (7.x, 8.x, 9.x) and I'd imagine most devs are focused on 
-CURRENT, aka, 10.x.

Now read the FBSD Handbook.  gmirror, for instance, went from six or so 
steps through 8.x, now it's a mess of steps in 9.x  But the Handbook, 
AFAIK, only shows the 9.x directions.  What used to take a few moments 
has become convoluted, and the documentation doesn't even apply to 2 of 
the supported branches.  Enter gpart land.  WTF.

So lots of devs have replied to this calmly, and it's been a regular 
topic of discussion in various forms.  People are upset, since I think 
we're all a bit 'conservative' in terms of technology.  I've yet to meet 
anyone in NYC*BUG who's enthralled by 'the cloud' marketing crap.  We 
know open source didn't spawn from someone's brain in the 1980s.  That 
Unix time might have started in 1970, but the decades before were part 
of the overall continuum.  BSD Unix is not about flash and splash.

If all these changes were happening with two supported branches, and it 
was all well-documented, then I think there would be a lot less stress 
all around.

Think about the massive changes:  pkgng, gpart/gpt, cvs dying for ports 
and source, the installer overhauled, the first real involvement in 
other architectures for many people (eg, ARM).  That's a lot, and we are 
all not only facing a learning curve, but also have to rework our tried 
and trusted methods.

Now Ike asked Kirk about this at that Axial meeting in a general way, 
and Kirk made it clear that a bunch of businesses were pumping more code 
upstream than in the past.  That's understandable, and probably a good 
thing.  Right?  It was the best explanation I'd heard so far.

So to wrap up this ramble, those 'defending' or 'explaining' the changes 
need to understand the real frustration many people are experiencing. 
We all have our ways to do things.  All sides should understand that. 
But I also think the FreeBSD project needs to slow the process down a 
bit, clear up the documentation and understand enormous "revolutionary" 
changes are not the norm 'culturally' for us.

For us 'end-users' we should probably spend more time in watching the 
changes, documenting them including by posting on this list, subscribe 
to freebsd-announce, and go from there.

The changes are inevitable, but there does need to better interaction 
between the user base (esp in groups like NYC*BUG) and the project.  And 
both 'sides' have responsibility in resolving it.

g

PS.  I didn't mean to sound like a diplomat.  I must be aging quicker 
than I thought.



From mcevoy.pat at gmail.com  Fri Feb 15 12:25:19 2013
From: mcevoy.pat at gmail.com (Pat McEvoy)
Date: Fri, 15 Feb 2013 12:25:19 -0500
Subject: [nycbug-talk] RE BSDTV
Message-ID: <CABPwdMX6_P3h05kjcY+TEHNksi-41SLzGogRN6A4vx5dmxQS0Q@mail.gmail.com>

I wanted to ask the group some advice on an idea I have to grow the BSDTV
project. I have been getting some good results on the hosted videos on
YouTube. Our last video from Eitan already has over 500 views and it is
barely up a week and the Twitter account is up over 666 followers. I wanted
to start a blog website for the project and wanted to ask your advice.

Goal:
Have a small informative website promoting the BSDTV videos and other
content garnered from NYCBug meetings and NYCBSDCons.
(If anyone has additional content ideas, please let me know. Jim and I
spoke of "HowTo" and "How do you use BSD" segments. If anyone would like to
volunteer to allow me to make a video of their work flows/setups, I would
be very interested. )


Using a small CMS powered website I am hoping to strike the balance
between the sometimes mutually exclusive ease of use and secure blogging
website. I would like it to have a RSS system that is configured to
generate a RSS entry with each new entry to the website. I would then
like to use that feed to populate the Facebook and Twitter feeds making
management of the BSDTV info as easy as possible. This will allow me to
keep the project going when things get busy. I would like to hear any
experienced input as to the best system to use for this project. I plan
to keep hosting the videos on YouTube and possibly explore the Mozilla
Popcorn project to make the presentations as informative as possible.


https://popcorn.webmaker.org/

Thank you.
P
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130215/d9b98e48/attachment.htm>

From bcallah at devio.us  Fri Feb 15 13:13:07 2013
From: bcallah at devio.us (Brian Callahan)
Date: Fri, 15 Feb 2013 13:13:07 -0500
Subject: [nycbug-talk] RE BSDTV
In-Reply-To: <CABPwdMX6_P3h05kjcY+TEHNksi-41SLzGogRN6A4vx5dmxQS0Q@mail.gmail.com>
References: <CABPwdMX6_P3h05kjcY+TEHNksi-41SLzGogRN6A4vx5dmxQS0Q@mail.gmail.com>
Message-ID: <511E7AB3.1060302@devio.us>

On 2/15/2013 12:25 PM, Pat McEvoy wrote:
> I wanted to ask the group some advice on an idea I have to grow the
> BSDTV project. I have been getting some good results on the hosted
> videos on YouTube. Our last video from Eitan already has over 500 views
> and it is barely up a week and the Twitter account is up over 666
> followers. I wanted to start a blog website for the project and wanted
> to ask your advice.
>
> Goal:
> Have a small informative website promoting the BSDTV videos and other
> content garnered from NYCBug meetings and NYCBSDCons.
> (If anyone has additional content ideas, please let me know. Jim and I
> spoke of "HowTo" and "How do you use BSD" segments. If anyone would like
> to volunteer to allow me to make a video of their work flows/setups, I
> would be very interested. )
>

Re: Content ideas. Are you talking about something like this?
http://www.youtube.com/user/bsdtutorial (see the earlier videos from 1-2 
years ago)

I could see stuff like this being refreshed and possibly made more 
"lively" (though I have no idea how to achieve the "lively" - I'll come 
clean and admit that video production is way over my head).

I think "how do you use BSD" is a great idea.

~Brian

>
> Using a small CMS powered website I am hoping to strike the balance
> between the sometimes mutually exclusive ease of use and secure blogging
> website. I would like it to have a RSS system that is configured to
> generate a RSS entry with each new entry to the website. I would then
> like to use that feed to populate the Facebook and Twitter feeds making
> management of the BSDTV info as easy as possible. This will allow me to
> keep the project going when things get busy. I would like to hear any
> experienced input as to the best system to use for this project. I plan
> to keep hosting the videos on YouTube and possibly explore the Mozilla
> Popcorn project to make the presentations as informative as possible.
>
>
> https://popcorn.webmaker.org/
>
> Thank you.
> P
>
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>



From nycbug at wynn.com  Fri Feb 15 14:19:58 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Fri, 15 Feb 2013 14:19:58 -0500
Subject: [nycbug-talk] Request for Review,
 Summary of FreeBSD src fetching problems
In-Reply-To: <CALMnNGgV2FatW_Q_MhYFX1D+KNhXwtSCu_ze4De8aCfMKE_L=A@mail.gmail.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us>
 <20130214134902.2e532e5d@ivory.wynn.com>
 <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>
 <20130215010933.551e994b@ivory.wynn.com>
 <CALMnNGgV2FatW_Q_MhYFX1D+KNhXwtSCu_ze4De8aCfMKE_L=A@mail.gmail.com>
Message-ID: <20130215141958.1fa8cd11@ivory.wynn.com>

On Fri, 15 Feb 2013 11:23:11 -0600
Andy Kosela <akosela at andykosela.com> wrote:

> On Fri, Feb 15, 2013 at 12:09 AM, Brett Wynkoop <nycbug at wynn.com>
> wrote:
> > So I say again someone better than me needs to write a basesnap
> > utility and the server infrastructure needs to be put in place.
> > This should be pretty trivial considering it will be basically a
> > clone of portsnap!
> 
> There is a saying in the Open Source world and definetly you know it
> too.  You want something done the way you see it?  You better write it
> yourself...
> 
> svnup in the base would be nice though.

Andy-

I could probably hack portsnap to become basesnap in a couple of days
(being realistic for testing and fixing bugs I am sure to introduce),
but doing such a utility without buy in from those who decide on how
stuff is put up on the http/ftp servers would be useless.  I believe,
but I have not had the time to look at it, portsnap requires specially
generated files daily on the hosting servers.

I am open to suggestions on how to get buy in at that level.  

-Brett


-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

I would never invade the United States.  There would be a gun behind
every blade of grass.  --Isoroku Yamamoto



From akosela at andykosela.com  Fri Feb 15 14:36:56 2013
From: akosela at andykosela.com (Andy Kosela)
Date: Fri, 15 Feb 2013 13:36:56 -0600
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
In-Reply-To: <20130215141958.1fa8cd11@ivory.wynn.com>
References: <1359316681-142543.691364919.fr0RJvVul024357@rs149.luxsci.com>
 <20130127225158.GQ1423@glenbarber.us>
 <1359329162-1994236.5580628.fr0RNPUGr003018@rs149.luxsci.com>
 <20130214104053.37716dea@ivory.wynn.com>
 <CAJU5ZBP2omnmg2xpMuhaPmZBSs=s-gK0LRhVP4JvHD6sNOjSRA@mail.gmail.com>
 <511D24FC.9040300@devio.us>
 <20130214134902.2e532e5d@ivory.wynn.com>
 <CAJU5ZBOT3vtkvnSPs-wuLd7PGhKC+sWm4Xs51Qpp6nVucU+MFQ@mail.gmail.com>
 <20130215010933.551e994b@ivory.wynn.com>
 <CALMnNGgV2FatW_Q_MhYFX1D+KNhXwtSCu_ze4De8aCfMKE_L=A@mail.gmail.com>
 <20130215141958.1fa8cd11@ivory.wynn.com>
Message-ID: <CALMnNGg1TrW2JepeTNgKaX47XLaSqHnYGy0vCgeiK_5cA-KELQ@mail.gmail.com>

On Fri, Feb 15, 2013 at 1:19 PM, Brett Wynkoop <nycbug at wynn.com> wrote:
> On Fri, 15 Feb 2013 11:23:11 -0600
> Andy Kosela <akosela at andykosela.com> wrote:
>
>> On Fri, Feb 15, 2013 at 12:09 AM, Brett Wynkoop <nycbug at wynn.com>
>> wrote:
>> > So I say again someone better than me needs to write a basesnap
>> > utility and the server infrastructure needs to be put in place.
>> > This should be pretty trivial considering it will be basically a
>> > clone of portsnap!
>>
>> There is a saying in the Open Source world and definetly you know it
>> too.  You want something done the way you see it?  You better write it
>> yourself...
>>
>> svnup in the base would be nice though.
>
> I could probably hack portsnap to become basesnap in a couple of days
> (being realistic for testing and fixing bugs I am sure to introduce),
> but doing such a utility without buy in from those who decide on how
> stuff is put up on the http/ftp servers would be useless.  I believe,
> but I have not had the time to look at it, portsnap requires specially
> generated files daily on the hosting servers.
>
> I am open to suggestions on how to get buy in at that level.

Would it be not more beneficial to implement something similar to
csup(1) but working with subversion instead of going portsnap way?
That way you don't need servers infrastructure.  It's all already set
up in place.  My understanding is that the frustration comes from
multiple levels of dependencies for subversion.  A lightweight svnup
would be an ideal solution for that.

--Andy


From gjb at FreeBSD.org  Fri Feb 15 14:38:44 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Fri, 15 Feb 2013 14:38:44 -0500
Subject: [nycbug-talk] Request for Review,
	Summary of FreeBSD src fetching problems
Message-ID: <7af2c8b6-cd60-48e9-8285-b464031be22d@email.android.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256



Now with proper from address...

Brett Wynkoop <nycbug at wynn.com> wrote:

>On Fri, 15 Feb 2013 11:23:11 -0600
>Andy Kosela <akosela at andykosela.com> wrote:
>
>> On Fri, Feb 15, 2013 at 12:09 AM, Brett Wynkoop <nycbug at wynn.com>
>> wrote:
>I could probably hack portsnap to become basesnap in a couple of days

Why not look at DES's svnup code in svn before reinventing GID=0?

>(being realistic for testing and fixing bugs I am sure to introduce),
>but doing such a utility without buy in from those who decide on how
>stuff is put up on the http/ftp servers would be useless.  I believe,
>but I have not had the time to look at it, portsnap requires specially
>generated files daily on the hosting servers.
>
>I am open to suggestions on how to get buy in at that level.
>

If you can provide a working test case, I will personally pull necessary strings to generate upstream deltas.

Glen

- --
Glen
Sent from my Android phone.
-----BEGIN PGP SIGNATURE-----
Version: APG v1.0.8

iQFDBAEBCAAtBQJRHo7EJhxHbGVuIEJhcmJlciA8Z2xlbi5qLmJhcmJlckBnbWFp
bC5jb20+AAoJEFJPDDeguUajDi0H/2Q/41ZSv/ecPxzoEZzvlh348PSmdJ6MEGON
+T+0tGNX8N5wIHpv+bg5af38MH7VBOBr/p+6vI0zHEyBECJzVpHDy7lbsuPaDre/
11P+LyoJworf0UJAEzAV2ZXaSpCQOa5qHhyGd/vnQSq7tuKJKehLXTxaYo0zId9a
clOe+5piUCoPocFYtNioo75AlSa1DKk81xzqGa3ch5+xr+SLbW5hT5MjjqW9pnGs
IdnjND1c7HKuyaDtblDDU7sssT/LHXK9zWH/wXhfHmXU8rqfNppfkCKJ1ZVAI+py
1do17y1qVLsM4KL18FkWnDtp0HLGf7lVQfWxkl5lN6jybYGmLZ0=
=ftpH
-----END PGP SIGNATURE-----



From george at ceetonetechnology.com  Wed Feb 20 23:41:24 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Wed, 20 Feb 2013 23:41:24 -0500
Subject: [nycbug-talk] Another BeagleBone release
Message-ID: <5125A574.1090601@ceetonetechnology.com>

Good thing we didn't put in that order yet.

http://beagleboard.org/unzipped/

g


From spork at bway.net  Thu Feb 21 00:26:34 2013
From: spork at bway.net (Charles Sprickman)
Date: Thu, 21 Feb 2013 00:26:34 -0500
Subject: [nycbug-talk] Another BeagleBone release
In-Reply-To: <5125A574.1090601@ceetonetechnology.com>
References: <5125A574.1090601@ceetonetechnology.com>
Message-ID: <8C3AAAD2-5636-400E-A130-511713B8D5F3@bway.net>


On Feb 20, 2013, at 11:41 PM, George Rosamond wrote:

> Good thing we didn't put in that order yet.
> 
> http://beagleboard.org/unzipped/

This is interesting:

http://downloads.element14.com/beaglebone/

My Pi arrived the other day (rev. B, only delay was the shipping) and the box it was in had Element 14 branding all over it.  Kind of sounds like the same folks manufacturing the Pi will be making the "unzipped" beaglebone.

I haven't slapped anything on the Pi yet mainly because I can't find either of the SD card readers I own.

C

> g
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk




From nycbug at wynn.com  Thu Feb 21 01:21:12 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 21 Feb 2013 01:21:12 -0500
Subject: [nycbug-talk] jabber server suggestion
Message-ID: <20130221012112.05d9ddc6@ivory.wynn.com>

Greeting-

I have decided to put a jabber server up on my BeagleBone.  I
discovered the other night that iChat on the Mac can use jabber, so the
mission of this server is to allow my wife and I to video chat via
iChat when one of us is traveling and the other is home.  

Those of you who know me well will understand why suggesting skype, or
yahoo, or name-your-favorite-big-player is not how I want my traffic
handled.

So I am looking for a suggestion of a jabber server that will go up
nice, easy, and clean without needing weeks of study to figure out the
config file settings needed.  Any ideas?  

I probably want to avoid java as I have no clue if it can even be built
on ARM.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

Gun Control: The theory that a woman found dead in an alley, raped and
strangled with her own pantyhose, is somehow morally superior to a
woman explaining to police how her attacker got that fatal bullet wound



From spork at bway.net  Thu Feb 21 01:39:29 2013
From: spork at bway.net (Charles Sprickman)
Date: Thu, 21 Feb 2013 01:39:29 -0500
Subject: [nycbug-talk] jabber server suggestion
In-Reply-To: <20130221012112.05d9ddc6@ivory.wynn.com>
References: <20130221012112.05d9ddc6@ivory.wynn.com>
Message-ID: <8558F077-023D-4A59-95D7-9BAD22F5195B@bway.net>

On Feb 21, 2013, at 1:21 AM, Brett Wynkoop wrote:

> Greeting-
> 
> I have decided to put a jabber server up on my BeagleBone.  I
> discovered the other night that iChat on the Mac can use jabber, so the
> mission of this server is to allow my wife and I to video chat via
> iChat when one of us is traveling and the other is home.
> 
> Those of you who know me well will understand why suggesting skype, or
> yahoo, or name-your-favorite-big-player is not how I want my traffic
> handled.
> 
> So I am looking for a suggestion of a jabber server that will go up
> nice, easy, and clean without needing weeks of study to figure out the
> config file settings needed.  Any ideas?  

I didn't find anything that was really simple when looking, but I'm running this internally for a client and it felt like the least bad option at the time:

http://www.freshports.org/net-im/ejabberd
http://www.ejabberd.im/

It requires erlang.  I'm running it on a fairly old crapbox:

CPU: AMD Sempron(tm) Processor 3000+ (1799.81-MHz 686-class CPU)

The box does have 2GB of RAM, but the jabber process is not too much of a pig:

543      1438  0.0  3.3 40876 32472  ??  SJ   26Dec12   4:54.18 [beam]

All I can say about it really is that it works, and the config was not too tough.  Never got the "push contact lists to clients" thing working, but that could be client-side issues.  Have not tried any audio/video chat features, just plain old text.  I've had zero issues since installing it though, it tends to "just work".

> I probably want to avoid java as I have no clue if it can even be built
> on ARM.

I ran some other java based jabber server for a bit.  Even if you could build java, the thing was using a ton of memory from the get-go and only got worse as time went on.

Charles


> 
> -Brett
> 
> -- 
> 
> wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
> 917-642-6925
> 718-717-5435
> 
> Gun Control: The theory that a woman found dead in an alley, raped and
> strangled with her own pantyhose, is somehow morally superior to a
> woman explaining to police how her attacker got that fatal bullet wound
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk




From pete at nomadlogic.org  Thu Feb 21 13:20:59 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Thu, 21 Feb 2013 10:20:59 -0800
Subject: [nycbug-talk] jabber server suggestion
In-Reply-To: <8558F077-023D-4A59-95D7-9BAD22F5195B@bway.net>
References: <20130221012112.05d9ddc6@ivory.wynn.com>
 <8558F077-023D-4A59-95D7-9BAD22F5195B@bway.net>
Message-ID: <5126658B.2060202@nomadlogic.org>

On 02/20/13 22:39, Charles Sprickman wrote:
> On Feb 21, 2013, at 1:21 AM, Brett Wynkoop wrote:
>
>> Greeting-
>>
>> I have decided to put a jabber server up on my BeagleBone.  I
>> discovered the other night that iChat on the Mac can use jabber, so the
>> mission of this server is to allow my wife and I to video chat via
>> iChat when one of us is traveling and the other is home.
>>
>> Those of you who know me well will understand why suggesting skype, or
>> yahoo, or name-your-favorite-big-player is not how I want my traffic
>> handled.
>>
>> So I am looking for a suggestion of a jabber server that will go up
>> nice, easy, and clean without needing weeks of study to figure out the
>> config file settings needed.  Any ideas?
> I didn't find anything that was really simple when looking, but I'm running this internally for a client and it felt like the least bad option at the time:
>
> http://www.freshports.org/net-im/ejabberd
> http://www.ejabberd.im/
>
> It requires erlang.  I'm running it on a fairly old crapbox:
>
> CPU: AMD Sempron(tm) Processor 3000+ (1799.81-MHz 686-class CPU)
>
> The box does have 2GB of RAM, but the jabber process is not too much of a pig:
>
> 543      1438  0.0  3.3 40876 32472  ??  SJ   26Dec12   4:54.18 [beam]
>
> All I can say about it really is that it works, and the config was not too tough.  Never got the "push contact lists to clients" thing working, but that could be client-side issues.  Have not tried any audio/video chat features, just plain old text.  I've had zero issues since installing it though, it tends to "just work".

+1 on ejabberd especially on more constrained systems.  configs are well 
documented, and the webUI is pretty intuitive as well.  i just built 
erlang-15 yesterday for some rabbitMQ hacking so I have the size of the 
exploded pkg handy:

 > pkg info -s erlang
erlang-15.b.03.1,3 size is: 266 MB

>> I probably want to avoid java as I have no clue if it can even be built
>> on ARM.
> I ran some other java based jabber server for a bit.  Even if you could build java, the thing was using a ton of memory from the get-go and only got worse as time went on.

+1 on that too :)

-pete

-- 
Pete Wright
pete at nomadlogic.org
twitter =>  @nomadlogicLA




From mike at myownsoho.net  Thu Feb 21 13:40:02 2013
From: mike at myownsoho.net (Mike N.)
Date: Thu, 21 Feb 2013 13:40:02 -0500
Subject: [nycbug-talk] Another BeagleBone release
In-Reply-To: <5125A574.1090601@ceetonetechnology.com>
References: <5125A574.1090601@ceetonetechnology.com>
Message-ID: <7f11bc92e4810196c20b5aab2db510ba@myownsoho.net>

 

put me down for the next group-purchase 

mike-- 

On 2013-02-20 23:41, George Rosamond wrote: 

> Good thing we didn't put in that order yet.
> 
> http://beagleboard.org/unzipped/ [1]
> 
> g
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk [2]

-- 
Mike Nichols
mike at myownsoho.net
o. 212 2022194
c. 347 7251661
http://myownsoho.com
 

Links:
------
[1] http://beagleboard.org/unzipped/
[2] http://lists.nycbug.org/mailman/listinfo/talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130221/d918e217/attachment.htm>

From nycbug at wynn.com  Thu Feb 21 15:54:42 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 21 Feb 2013 15:54:42 -0500
Subject: [nycbug-talk] Fw: BIND10 1.0.0 is now available
Message-ID: <20130221155442.66b8380a@ivory.wynn.com>

FYI folks

-Brett

Begin forwarded message:

Date: Thu, 21 Feb 2013 13:27:15 -0600 (CST)
From: "Jeremy C. Reed" <jreed at isc.org>
To: undisclosed
Subject: BIND10 1.0.0 is now available


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


Welcome to the 1.0.0 release of BIND 10, the new DNS and DHCP
software from Internet Systems Consortium (ISC).  BIND 10 provides
a C++ library for DNS (with Python wrappers) and several cooperating
daemons for providing authoritative DNS service, dynamic DNS, zone
transfers, and experimental forwarding and recursive name service.
Supplementary components are included for statistics collection
and reporting and remote configuration and control. The current
DNS features include:

 * IXFR and AXFR zone transfers in and out
 * In-memory backend for zone data for faster performance
 * SQLite3 backend for quick startup or huge amounts of zone data
 * DNSSEC: serving signed zones by authoritative server
   - Either in-memory or from DB backend
   - Including NSEC3
 * DDNS (Dynamic Updates) support
 * Access Control (for DDNS and outbound zone transfers)
 * TSIG support:
   - Part of general ACLs (DDNS uses it)
   - Authoritative server can verify and sign DNS messages with TSIG
   - Zone transfers can have signed transfers

BIND 10 allows enabling and disabling extra features via different
runable modules.  It provides a semi-interactive client to conveniently
look at and set some configuration settings.  It offers a detailed
log message system which provides unique log ids with corresponding
paragraph explanations.  Some statistics are available via the
command-line or in XML over HTTP.

It also provides a RESTful interface over HTTPS for configurations
and control, detailed documentation for the implementation/public
API, and meticulous test cases to help understand and extend the
code.  Developers may use the Python scripting language to extend
or tweak BIND 10 and replacement modules may be used to customize
BIND 10 or to offer other services.

DOCUMENTATION

BIND 10 provides a guide with installation instructions
and usage tutorials, a messages document with the unique logging
explanations, and manual pages.  The documentation for BIND 10 is
available with the installation or via http://bind10.isc.org/docs/
in HTML, plain text, or PDF formats.

DOWNLOAD & INSTALLATION

The bind10-1.0.0 source may be downloaded from:

  ftp://ftp.isc.org/isc/bind10/1.0.0/bind10-1.0.0.tar.gz

A PGP signature of the distribution is at

  ftp://ftp.isc.org/isc/bind10/1.0.0/bind10-1.0.0.tar.gz.sha512.asc

The signature was generated with the ISC code signing key which is
available at https://www.isc.org/about/openpgp

Installation details are documented in the BIND 10 Guide.  In
addition, installation suggestions for various operating systems
are available via our wiki http://bind10.isc.org/wiki/InstallStartPage

SUPPORT

While it is not feature complete, BIND 10 1.0.0 is fully supported.
If you have any questions or comments about working with this
release, you may post them to the BIND 10 Users List
https://lists.isc.org/mailman/listinfo/bind10-users
or to the BIND 10 Jabber room.

Bugs and feature requests may also be submitted via the ticket
tracking system at http://bind10.isc.org/wiki

Paid support, training, and consulting options are available; see
http://www.isc.org/support for information.

BIND 10 DHCP

The version of BIND 10 DHCP included in BIND 1.0.0 is an engineering
snapshot and intended for experimental use only.  The DHCP servers
include the following features:

 * Ability to allocate and renew addresses, and handle lease expiration
   and releases.
 * Supports a subset of clients:
   - DHCPv4 clients connected to the server via a relay.
   - DHCPv6 clients on the same LAN as the server.
 * Ability to configure values for standard options returned to a
   client, either globally or on a per-subnet basis.
 * Ability to define new options and configure them in the same way as
   standard options.
 * Lease storage in a MySQL Database.
 * Configuration, logging and process control uses the same mechanisms
   as the BIND 10 DNS server.

Limitations and known issues with this release can be found at
http://bind10.isc.org/wiki/KeaKnownIssues.  Please consult this
page if you encounter problems.

We would also appreciate feedback on usability and reports of any
DHCP bugs.  Please contribute your experiences to the BIND 10 DHCP
mailing list: https://lists.isc.org/mailman/listinfo/bind10-dhcp

SPONSORS

BIND 10 is a sponsored development project, and would not be possible
without the generous support of the sponsors.  JPRS and CIRA are
Patron Level sponsors.  AFNIC, CNNIC, CZ.NIC, DENIC eG, Google,
RIPE NCC, Registro.br, .nz Registry Services, and Technical Center
of Internet are current sponsors.  Afilias, IIS.SE, Nominet, and
SIDN were founding sponsors of the project.  Support for BIND 10
development of the DHCPv4 and DHCPv6 components is provided by
Comcast.

If you would like to contribute to ISC to assist us in continuing to
make quality open source software, please visit our donations page at
http://www.isc.org/supportisc

Thanks again to those who contributed bug reports, code, and reviews.

Jeremy C. Reed
ISC Release Engineering

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (NetBSD)

iEYEARECAAYFAlEmdPwACgkQs9Bv5D4YwC1qDQCgywlJeasA63nWc2/jNXzhMqPy
IwEAn2uv7o7QEihbRTfdWO0IOtAA6922
=CsJY
-----END PGP SIGNATURE-----


-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

Gun Control: The theory that a woman found dead in an alley, raped and
strangled with her own pantyhose, is somehow morally superior to a
woman explaining to police how her attacker got that fatal bullet wound



From okan at demirmen.com  Thu Feb 21 16:22:45 2013
From: okan at demirmen.com (Okan Demirmen)
Date: Thu, 21 Feb 2013 16:22:45 -0500
Subject: [nycbug-talk] Fw: BIND10 1.0.0 is now available
In-Reply-To: <20130221155442.66b8380a@ivory.wynn.com>
References: <20130221155442.66b8380a@ivory.wynn.com>
Message-ID: <20130221212244.GA15815@clam.khaoz.org>

is there a reason why these are being forwarded here?

On Thu 2013.02.21 at 15:54 -0500, Brett Wynkoop wrote:
> FYI folks
> 
> -Brett
> 
> Begin forwarded message:
> 
> Date: Thu, 21 Feb 2013 13:27:15 -0600 (CST)
> From: "Jeremy C. Reed" <jreed at isc.org>
> To: undisclosed
> Subject: BIND10 1.0.0 is now available
> 
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> 
> Welcome to the 1.0.0 release of BIND 10, the new DNS and DHCP
> software from Internet Systems Consortium (ISC).  BIND 10 provides
> a C++ library for DNS (with Python wrappers) and several cooperating
> daemons for providing authoritative DNS service, dynamic DNS, zone
> transfers, and experimental forwarding and recursive name service.
> Supplementary components are included for statistics collection
> and reporting and remote configuration and control. The current
> DNS features include:
> 
>  * IXFR and AXFR zone transfers in and out
>  * In-memory backend for zone data for faster performance
>  * SQLite3 backend for quick startup or huge amounts of zone data
>  * DNSSEC: serving signed zones by authoritative server
>    - Either in-memory or from DB backend
>    - Including NSEC3
>  * DDNS (Dynamic Updates) support
>  * Access Control (for DDNS and outbound zone transfers)
>  * TSIG support:
>    - Part of general ACLs (DDNS uses it)
>    - Authoritative server can verify and sign DNS messages with TSIG
>    - Zone transfers can have signed transfers
> 
> BIND 10 allows enabling and disabling extra features via different
> runable modules.  It provides a semi-interactive client to conveniently
> look at and set some configuration settings.  It offers a detailed
> log message system which provides unique log ids with corresponding
> paragraph explanations.  Some statistics are available via the
> command-line or in XML over HTTP.
> 
> It also provides a RESTful interface over HTTPS for configurations
> and control, detailed documentation for the implementation/public
> API, and meticulous test cases to help understand and extend the
> code.  Developers may use the Python scripting language to extend
> or tweak BIND 10 and replacement modules may be used to customize
> BIND 10 or to offer other services.
> 
> DOCUMENTATION
> 
> BIND 10 provides a guide with installation instructions
> and usage tutorials, a messages document with the unique logging
> explanations, and manual pages.  The documentation for BIND 10 is
> available with the installation or via http://bind10.isc.org/docs/
> in HTML, plain text, or PDF formats.
> 
> DOWNLOAD & INSTALLATION
> 
> The bind10-1.0.0 source may be downloaded from:
> 
>   ftp://ftp.isc.org/isc/bind10/1.0.0/bind10-1.0.0.tar.gz
> 
> A PGP signature of the distribution is at
> 
>   ftp://ftp.isc.org/isc/bind10/1.0.0/bind10-1.0.0.tar.gz.sha512.asc
> 
> The signature was generated with the ISC code signing key which is
> available at https://www.isc.org/about/openpgp
> 
> Installation details are documented in the BIND 10 Guide.  In
> addition, installation suggestions for various operating systems
> are available via our wiki http://bind10.isc.org/wiki/InstallStartPage
> 
> SUPPORT
> 
> While it is not feature complete, BIND 10 1.0.0 is fully supported.
> If you have any questions or comments about working with this
> release, you may post them to the BIND 10 Users List
> https://lists.isc.org/mailman/listinfo/bind10-users
> or to the BIND 10 Jabber room.
> 
> Bugs and feature requests may also be submitted via the ticket
> tracking system at http://bind10.isc.org/wiki
> 
> Paid support, training, and consulting options are available; see
> http://www.isc.org/support for information.
> 
> BIND 10 DHCP
> 
> The version of BIND 10 DHCP included in BIND 1.0.0 is an engineering
> snapshot and intended for experimental use only.  The DHCP servers
> include the following features:
> 
>  * Ability to allocate and renew addresses, and handle lease expiration
>    and releases.
>  * Supports a subset of clients:
>    - DHCPv4 clients connected to the server via a relay.
>    - DHCPv6 clients on the same LAN as the server.
>  * Ability to configure values for standard options returned to a
>    client, either globally or on a per-subnet basis.
>  * Ability to define new options and configure them in the same way as
>    standard options.
>  * Lease storage in a MySQL Database.
>  * Configuration, logging and process control uses the same mechanisms
>    as the BIND 10 DNS server.
> 
> Limitations and known issues with this release can be found at
> http://bind10.isc.org/wiki/KeaKnownIssues.  Please consult this
> page if you encounter problems.
> 
> We would also appreciate feedback on usability and reports of any
> DHCP bugs.  Please contribute your experiences to the BIND 10 DHCP
> mailing list: https://lists.isc.org/mailman/listinfo/bind10-dhcp
> 
> SPONSORS
> 
> BIND 10 is a sponsored development project, and would not be possible
> without the generous support of the sponsors.  JPRS and CIRA are
> Patron Level sponsors.  AFNIC, CNNIC, CZ.NIC, DENIC eG, Google,
> RIPE NCC, Registro.br, .nz Registry Services, and Technical Center
> of Internet are current sponsors.  Afilias, IIS.SE, Nominet, and
> SIDN were founding sponsors of the project.  Support for BIND 10
> development of the DHCPv4 and DHCPv6 components is provided by
> Comcast.
> 
> If you would like to contribute to ISC to assist us in continuing to
> make quality open source software, please visit our donations page at
> http://www.isc.org/supportisc
> 
> Thanks again to those who contributed bug reports, code, and reviews.
> 
> Jeremy C. Reed
> ISC Release Engineering
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.12 (NetBSD)
> 
> iEYEARECAAYFAlEmdPwACgkQs9Bv5D4YwC1qDQCgywlJeasA63nWc2/jNXzhMqPy
> IwEAn2uv7o7QEihbRTfdWO0IOtAA6922
> =CsJY
> -----END PGP SIGNATURE-----
> 
> 
> -- 
> 
> wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
> 917-642-6925
> 718-717-5435
> 
> Gun Control: The theory that a woman found dead in an alley, raped and
> strangled with her own pantyhose, is somehow morally superior to a
> woman explaining to police how her attacker got that fatal bullet wound
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk


From nycbug at wynn.com  Thu Feb 21 17:20:56 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 21 Feb 2013 17:20:56 -0500
Subject: [nycbug-talk] Fw: BIND10 1.0.0 is now available
In-Reply-To: <20130221212244.GA15815@clam.khaoz.org>
References: <20130221155442.66b8380a@ivory.wynn.com>
 <20130221212244.GA15815@clam.khaoz.org>
Message-ID: <20130221172056.75ac989b@ivory.wynn.com>

On Thu, 21 Feb 2013 16:22:45 -0500
Okan Demirmen <okan at demirmen.com> wrote:

> is there a reason why these are being forwarded here?

Some number of NYCBUG folks > 0 are system administrators
Some number of NYCBUG system admins > 0 maintain DNS servers
Some number of NYCBUG system admins who maintain DNS servers > 0 use \
     bind


BIND 9 is in the base of all the *BSD releases AFIK.

To my belief and knowledge there are members of the group who are
interested in new technologies that apply to *BSD.

Therefor I forward the announcements.

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

April 19, 1775 An English attempt to confiscate guns from Americans
triggered a successful revolution......

     Dear Congress, that's a hint.


From okan at demirmen.com  Thu Feb 21 17:51:43 2013
From: okan at demirmen.com (Okan Demirmen)
Date: Thu, 21 Feb 2013 17:51:43 -0500
Subject: [nycbug-talk] Fw: BIND10 1.0.0 is now available
In-Reply-To: <20130221172056.75ac989b@ivory.wynn.com>
References: <20130221155442.66b8380a@ivory.wynn.com>
 <20130221212244.GA15815@clam.khaoz.org>
 <20130221172056.75ac989b@ivory.wynn.com>
Message-ID: <20130221225142.GA30879@clam.khaoz.org>

On Thu 2013.02.21 at 17:20 -0500, Brett Wynkoop wrote:
> On Thu, 21 Feb 2013 16:22:45 -0500
> Okan Demirmen <okan at demirmen.com> wrote:
> 
> > is there a reason why these are being forwarded here?
> 
> Some number of NYCBUG folks > 0 are system administrators
> Some number of NYCBUG system admins > 0 maintain DNS servers
> Some number of NYCBUG system admins who maintain DNS servers > 0 use \
>      bind
> 
> BIND 9 is in the base of all the *BSD releases AFIK.
> 
> To my belief and knowledge there are members of the group who are
> interested in new technologies that apply to *BSD.
> 
> Therefor I forward the announcements.

The point is if these users are interested, then shouldn't they be on
the respective product/tool/gizmo lists?

NYCBUG is not supposed to be a news aggregator.


From nycbug at wynn.com  Thu Feb 21 18:27:29 2013
From: nycbug at wynn.com (Brett Wynkoop)
Date: Thu, 21 Feb 2013 18:27:29 -0500
Subject: [nycbug-talk] Fw: BIND10 1.0.0 is now available
In-Reply-To: <20130221225142.GA30879@clam.khaoz.org>
References: <20130221155442.66b8380a@ivory.wynn.com>
 <20130221212244.GA15815@clam.khaoz.org>
 <20130221172056.75ac989b@ivory.wynn.com>
 <20130221225142.GA30879@clam.khaoz.org>
Message-ID: <20130221182729.1f07f3d3@ivory.wynn.com>

On Thu, 21 Feb 2013 17:51:43 -0500
Okan Demirmen <okan at demirmen.com> wrote:

> The point is if these users are interested, then shouldn't they be on
> the respective product/tool/gizmo lists?
> 
> NYCBUG is not supposed to be a news aggregator.


So you also objected to the recent email about the new version of the
BeagleBone?

-Brett

-- 

wynkoop at wynn.com               http://prd4.wynn.com/wynkoop/pgp-keys.txt
917-642-6925
718-717-5435

Gun Control: The theory that a woman found dead in an alley, raped and
strangled with her own pantyhose, is somehow morally superior to a
woman explaining to police how her attacker got that fatal bullet wound



From george at ceetonetechnology.com  Thu Feb 21 18:45:30 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Thu, 21 Feb 2013 18:45:30 -0500
Subject: [nycbug-talk] Fw: BIND10 1.0.0 is now available
In-Reply-To: <20130221182729.1f07f3d3@ivory.wynn.com>
References: <20130221155442.66b8380a@ivory.wynn.com>
 <20130221212244.GA15815@clam.khaoz.org>
 <20130221172056.75ac989b@ivory.wynn.com>
 <20130221225142.GA30879@clam.khaoz.org>
 <20130221182729.1f07f3d3@ivory.wynn.com>
Message-ID: <5126B19A.4040902@ceetonetechnology.com>

On 02/21/13 18:27, Brett Wynkoop wrote:
> On Thu, 21 Feb 2013 17:51:43 -0500
> Okan Demirmen <okan at demirmen.com> wrote:
> 
>> The point is if these users are interested, then shouldn't they be on
>> the respective product/tool/gizmo lists?
>>
>> NYCBUG is not supposed to be a news aggregator.
> 
> 
> So you also objected to the recent email about the new version of the
> BeagleBone?

There is a significant difference between noting the release of a new
product of interest to a number of us and forwarding several announces
as a 'news aggregator'.  We don't swarm to post every vaguely relevant
slashdot post either.

If there is some particularly interesting on occasion, it's fine, IMHO.
 But those who are following BIND details should be plugged into  the
relevant place.  Talk@ is not the place for it.

g


From ike at blackskyresearch.net  Sat Feb 23 15:06:39 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Sat, 23 Feb 2013 15:06:39 -0500
Subject: [nycbug-talk] Sourcing Supermicro Drive Sleds?
Message-ID: <1361650022-932602.848237284.fr1NK6fh3030585@rs149.luxsci.com>

Hi All,

Does anyone know a good place to find supermicro drive sleds?
I'm shooting for inexpensive? (hoping for <$15 each)

I'm not having luck searching the net...

Best,
.ike





From gnn at neville-neil.com  Sat Feb 23 15:29:06 2013
From: gnn at neville-neil.com (George Neville-Neil)
Date: Sat, 23 Feb 2013 15:29:06 -0500
Subject: [nycbug-talk] Sourcing Supermicro Drive Sleds?
In-Reply-To: <1361650022-932602.848237284.fr1NK6fh3030585@rs149.luxsci.com>
References: <1361650022-932602.848237284.fr1NK6fh3030585@rs149.luxsci.com>
Message-ID: <B2D19A18-58D6-40E5-8549-C63060019E5A@neville-neil.com>


On Feb 23, 2013, at 15:06 , "Isaac (.ike) Levy" <ike at blackskyresearch.net> wrote:

> Hi All,
> 
> Does anyone know a good place to find supermicro drive sleds?
> I'm shooting for inexpensive? (hoping for <$15 each)
> 
> I'm not having luck searching the net...
> 

Talk to the folks at iX systems. They carry SM boards and the like.

Best,
George





From ike at blackskyresearch.net  Sat Feb 23 16:10:01 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Sat, 23 Feb 2013 16:10:01 -0500
Subject: [nycbug-talk] Sourcing Supermicro Drive Sleds?
In-Reply-To: <B2D19A18-58D6-40E5-8549-C63060019E5A@neville-neil.com>
References: <1361650022-932602.848237284.fr1NK6fh3030585@rs149.luxsci.com>
 <B2D19A18-58D6-40E5-8549-C63060019E5A@neville-neil.com>
Message-ID: <1361653862-1482481.93486694.fr1NLA2jD003126@rs149.luxsci.com>

On Feb 23, 2013, at 3:29 PM, George Neville-Neil wrote:

> 
> On Feb 23, 2013, at 15:06 , "Isaac (.ike) Levy" <ike at blackskyresearch.net> wrote:
> 
>> Hi All,
>> 
>> Does anyone know a good place to find supermicro drive sleds?
>> I'm shooting for inexpensive? (hoping for <$15 each)
>> 
>> I'm not having luck searching the net...
>> 
> 
> Talk to the folks at iX systems. They carry SM boards and the like.
> 
> Best,
> George

Thanks- just fired em' a note?

Rocket-
.ike





From spork at bway.net  Sat Feb 23 17:13:05 2013
From: spork at bway.net (Charles Sprickman)
Date: Sat, 23 Feb 2013 17:13:05 -0500
Subject: [nycbug-talk] Sourcing Supermicro Drive Sleds?
In-Reply-To: <1361650022-932602.848237284.fr1NK6fh3030585@rs149.luxsci.com>
References: <1361650022-932602.848237284.fr1NK6fh3030585@rs149.luxsci.com>
Message-ID: <DC0C1035-2621-46AD-8D35-A15F7F18CBEB@bway.net>

On Feb 23, 2013, at 3:06 PM, Isaac (.ike) Levy wrote:

> Hi All,
> 
> Does anyone know a good place to find supermicro drive sleds?
> I'm shooting for inexpensive? (hoping for <$15 each)

Insight seems to always have a bunch of weird Supermicro stuff available:

https://www.insight.com/insightweb/search#ProdInfoMtrId=MCP-220-00024-0V#licenseContractIds=~fromSearch?

That may not be in-stock, but we've had decent results with them on stranger parts that were special order (like the cable assembly for the front panel display, drive backplanes and really dated PSUs).  In the old days, these would be Li orders?

C

> 
> I'm not having luck searching the net...
> 
> Best,
> .ike
> 
> 
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk




From brian.gupta at gmail.com  Thu Feb 28 11:58:37 2013
From: brian.gupta at gmail.com (Brian Gupta)
Date: Thu, 28 Feb 2013 11:58:37 -0500
Subject: [nycbug-talk] CISPA 2.0's invasion of privacy is a huge and urgent
	problem
Message-ID: <CAAGunsS1ZRz-D0hTajuh2jx4zsGM317h+ZKbo9UJanO9KQPung@mail.gmail.com>

My apologies for this intrusion, but my understanding is that many in
this group share a similar set of sensibilities when it comes to
privacy rights.

Very recently, I've personally come to the conclusion, that CISPA,
even in it's never "friendlier" incarnation, is a vaguely worded,
government sanctioned, invasion of individual privacy with no real
recourse for overreach or
abuse. In light of recent episodes of Prosecutorial overreach, and
stretching of laws to hound Aaron Swartz, we can't allow this to pass.
It's going to be hard, as I feel like we aren't going to get much
industry support, as this isn't a direct attack on internet
businesses, like SOPA was.

Please read this FAQ, it makes the flaws and issues clear, and helps
lay out action steps, if you want to get involved in fighting it.
https://www.eff.org/cybersecurity-bill-faq

So far I've done the following, but want to hear more ideas:

1) I've written to Senator Kirsten Gillibrand's office and got a
largely non-committal answer in response. (I think lots of people need
to write.)
2) I've tweeted about it:
https://twitter.com/HelixOne/status/307145540305580032 (This feeds
into my Facebook page.)
3) I've shared on G+ (from both my accounts):
   https://plus.google.com/u/0/103030499292048040964/posts/SaGdAJTQWgY?cfem=1
   https://plus.google.com/u/0/112857626163792726537/posts/QCgAXueMJSf?cfem=1
4) I signed this petition: http://act.demandprogress.org/act/cispa_is_back/
5) I've started writing emails.

I think we as a group should brainstorm and start reaching out to
other like minded groups as well.

Any other ideas? (Also, If you are acting or know of groups that are,
please consider sharing.)

Thanks,
Brian


From ike at blackskyresearch.net  Thu Feb 28 12:47:34 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Thu, 28 Feb 2013 12:47:34 -0500
Subject: [nycbug-talk] CISPA 2.0's invasion of privacy is a huge and
	urgent problem
In-Reply-To: <CAAGunsS1ZRz-D0hTajuh2jx4zsGM317h+ZKbo9UJanO9KQPung@mail.gmail.com>
References: <CAAGunsS1ZRz-D0hTajuh2jx4zsGM317h+ZKbo9UJanO9KQPung@mail.gmail.com>
Message-ID: <1362073683-4464003.73978418.fr1SHlYwL025039@rs149.luxsci.com>

Hi Brian,

On Feb 28, 2013, at 11:58 AM, Brian Gupta wrote:

> My apologies for this intrusion, but my understanding is that many in
> this group share a similar set of sensibilities when it comes to
> privacy rights.
> 
> Very recently, I've personally come to the conclusion, that CISPA,
> even in it's never "friendlier" incarnation, is a vaguely worded,
> government sanctioned, invasion of individual privacy with no real
> recourse for overreach or
> abuse. In light of recent episodes of Prosecutorial overreach, and
> stretching of laws to hound Aaron Swartz, we can't allow this to pass.
> It's going to be hard, as I feel like we aren't going to get much
> industry support, as this isn't a direct attack on internet
> businesses, like SOPA was.
> 
> Please read this FAQ, it makes the flaws and issues clear, and helps
> lay out action steps, if you want to get involved in fighting it.
> https://www.eff.org/cybersecurity-bill-faq
> 
> So far I've done the following, but want to hear more ideas:
> 
> 1) I've written to Senator Kirsten Gillibrand's office and got a
> largely non-committal answer in response. (I think lots of people need
> to write.)
> 2) I've tweeted about it:
> https://twitter.com/HelixOne/status/307145540305580032 (This feeds
> into my Facebook page.)
> 3) I've shared on G+ (from both my accounts):
>   https://plus.google.com/u/0/103030499292048040964/posts/SaGdAJTQWgY?cfem=1
>   https://plus.google.com/u/0/112857626163792726537/posts/QCgAXueMJSf?cfem=1
> 4) I signed this petition: http://act.demandprogress.org/act/cispa_is_back/
> 5) I've started writing emails.
> 
> I think we as a group should brainstorm and start reaching out to
> other like minded groups as well.
> 
> Any other ideas? (Also, If you are acting or know of groups that are,
> please consider sharing.)
> 
> Thanks,
> Brian


This is a pretty cool note, although people may not reply very loudly on list regarding privacy concerns, I believe most people who've been part of internet infrastructure for as long as most of us have, are pretty like minded on these issues.

I think the EFF Link you shared is terrific, 
> https://www.eff.org/cybersecurity-bill-faq

it has a quick form to easily email your Congress/Senate representatives.

I think all the other ideas you pasted are great- personally, I'm continuing to share the EFF link with friends and colleagues.

Best,
.ike





From gnn at neville-neil.com  Thu Feb 28 22:04:40 2013
From: gnn at neville-neil.com (George Neville-Neil)
Date: Thu, 28 Feb 2013 22:04:40 -0500
Subject: [nycbug-talk] Efika laptop running X11...
Message-ID: <B6B6A216-BA00-4875-8D68-DAAF3153C373@neville-neil.com>

Yup, that project is done.  Now to get some more useful stuff on this.

ray@ did the work, and I did the testing.

Best,
George

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130228/d098eb1c/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IMG_0651.jpeg
Type: image/jpg
Size: 83041 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130228/d098eb1c/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 203 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130228/d098eb1c/attachment.bin>

From bcallah at devio.us  Thu Feb 28 22:10:31 2013
From: bcallah at devio.us (Brian Callahan)
Date: Thu, 28 Feb 2013 22:10:31 -0500
Subject: [nycbug-talk] Efika laptop running X11...
In-Reply-To: <B6B6A216-BA00-4875-8D68-DAAF3153C373@neville-neil.com>
References: <B6B6A216-BA00-4875-8D68-DAAF3153C373@neville-neil.com>
Message-ID: <51301C27.70903@devio.us>

On 2/28/2013 10:04 PM, George Neville-Neil wrote:
> Yup, that project is done.  Now to get some more useful stuff on this.
>
> ray@ did the work, and I did the testing.
>

Cool! Is it an accelerated X?

~Brian



From gnn at neville-neil.com  Thu Feb 28 22:16:08 2013
From: gnn at neville-neil.com (George Neville-Neil)
Date: Thu, 28 Feb 2013 22:16:08 -0500
Subject: [nycbug-talk] Efika laptop running X11...
In-Reply-To: <51301C27.70903@devio.us>
References: <B6B6A216-BA00-4875-8D68-DAAF3153C373@neville-neil.com>
 <51301C27.70903@devio.us>
Message-ID: <C987B3D4-7FE8-418F-9FCB-3AB1A5DCB5E4@neville-neil.com>


On Feb 28, 2013, at 22:10 , Brian Callahan <bcallah at devio.us> wrote:

> On 2/28/2013 10:04 PM, George Neville-Neil wrote:
>> Yup, that project is done.  Now to get some more useful stuff on this.
>> 
>> ray@ did the work, and I did the testing.
>> 
> 
> Cool! Is it an accelerated X?

I rather doubt that, this is a little ARM box.

Lots more info here: http://raybsd.blogspot.com

The next thing I need to do is start building more packages for it.

Ray's blog has all the instructions on how to get one of these running.

They're cheap, and yes, they feel cheap too, but damned amusing.

http://www.genesi-tech.com/products/smartbook

Best,
George