From ike at blackskyresearch.net  Sun Mar  3 21:14:46 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Sun, 3 Mar 2013 21:14:46 -0500
Subject: [nycbug-talk] interesting read (old pacemaker thread)
In-Reply-To: <201108291308.p7TD82Ab002791@rs134.luxsci.com>
References: <201108291308.p7TD82Ab002791@rs134.luxsci.com>
Message-ID: <1362363303-7816302.25606558.fr242ElHk003512@rs149.luxsci.com>

Hi All,

Ripping open a hot 8 year old thread, does anyone on list remember "the pacemaker scenario"?

On Aug 29, 2011, at 9:08 AM, Isaac Levy wrote:
On Sat, 21 May 2005 alex at pilosoft.com wrote:
>>>> Let's keep in mind that the trustworthiness of a life-critical
>>>> application has everything to do with how that program was written
> 
> http://lists.nycbug.org/pipermail/talk/2005-May/005497.html


I thought I'd pass along this lucid presentation I recently watched:

  "Beyond the War on General Purpose Computing: What's Inside the Box?"
  Cory Docotorow
http://defcon.org/html/links/dc-archives/dc-20-archive.html#Doctorow

  Here's a direct URL to the Video:
https://media.defcon.org/dc-20/video/DEF%20CON%2020%20Hacking%20Conference%20Presentation%20By%20Cory%20Docotorow%20-%20Beyond%20the%20War%20on%20General%20Purpose%20Computing%20Whats%20Inside%20the%20Box%20-%20Slides.m4v

--
Keeping things focused on *BSD matters, I'm really curious to know what folks here think about it?

 - TPM modules
 - data ownership
  - property rights, pitted against human rights
 - contemporary redefinition of user/owner
 - repo-man scenarios for pacemakers, etc...
 - dips into some interesting "Free/Open" arguments
   (which I believe are a bit shallow, in the presentation)

Best,
.ike





From jpb at jimby.name  Mon Mar  4 21:36:29 2013
From: jpb at jimby.name (Jimmy B.)
Date: Mon, 4 Mar 2013 21:36:29 -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: <20130305023629.GA28728@jimby.name>

* Brett Wynkoop <nycbug at wynn.com> [2013-02-21 15:55]:
> 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
> 

Hi Brett,

Thanks for posting this.  I know you took some minor flack for it,
but I appreciated it.

Cheers,
Jim B.



From george at ceetonetechnology.com  Mon Mar  4 21:47:01 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 04 Mar 2013 21:47:01 -0500
Subject: [nycbug-talk] NYC*BUG: This Week
Message-ID: <51355CA5.10406@ceetonetechnology.com>

The next two meetings focus on architectures beyond the regular
i386/amd64 world.  This month's meeting is about FreeBSD porting to the
Beagle Bone running on ARM, and April's meeting will feature a talk on
porting OpenBSD to MIPS.

Note that BSDCan (.org) is coming up in May.  A number of people from
NYC*BUG will be there, and we strongly encourage you to attend.

Yes, we're still cross-posting to talk@, but get on announce at for
these announces in the future.

Note that we launched a Tor-BSD list recently.  For anyone developing or
using Tor on a BSD, join up and let's get the discussion moving.

*  *  *

March 6 @ 18:45 - Location: Suspenders

BeagleBone with FreeBSD, Brett Wynkoop

About the speaker:
Brett Wynkoop fell in love with computers while a freshman at the United
States Merchant Marine Academy, where he almost flunked out his first
term by spending too much time playing with Dartmouth Time Sharing on a
model 33 teletype at 110 baud, instead of studying marine engineering
and navigation.

His first Unix job was administering an AT&T Dimension PBX which used
tape for random access....ls took a long time! His first BSD experience
was on a PDP 11/70 and he has been a BSD lover ever since.

His once wrote a web server in /bin/sh, just because he could
http://prd4.wynn.com:8080/.

He was a member of the technical staff at BSDI and is currently a
systems engineer with the Internet Systems Consortium and is working on
the BIND 10 project.


From george at ceetonetechnology.com  Mon Mar  4 23:33:23 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 04 Mar 2013 23:33:23 -0500
Subject: [nycbug-talk] announce list
Message-ID: <51357593.5080102@ceetonetechnology.com>

So we don't like cross-posting announce@ to talk at .  Really.

So we won't be doing it anymore.  If you want announces, join the
announce list.

It seems the footer disappeared from talk@ indicating that you should
join announce@ for announces.  We will re-add it, and cease cross-posting.

So join announce, or we just won't see you at meetings.

g


From george at ceetonetechnology.com  Wed Mar  6 10:19:23 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Wed, 06 Mar 2013 10:19:23 -0500
Subject: [nycbug-talk] another cool ARM board
Message-ID: <51375E7B.9070801@ceetonetechnology.com>

www.marsboard.com

The hits just keep on coming....

Also, armed.nycbug.org is up, but not fully operational.  Stick getting
an error on building RPi images, which hopefully will resolve when the
build box itself is updated from 8.3 to -CURRENT.

The BeagleBone images should be fine.

Feedback welcome.

g


From chsnyder at gmail.com  Wed Mar  6 10:32:12 2013
From: chsnyder at gmail.com (Chris Snyder)
Date: Wed, 6 Mar 2013 10:32:12 -0500
Subject: [nycbug-talk] another cool ARM board
In-Reply-To: <51375E7B.9070801@ceetonetechnology.com>
References: <51375E7B.9070801@ceetonetechnology.com>
Message-ID: <CAPKKYS1yCwpQG4V1uQ7_C2psPC0PEfG=m6uU_8sHi_hj+oHiyQ@mail.gmail.com>

On Wed, Mar 6, 2013 at 10:19 AM, George Rosamond <
george at ceetonetechnology.com> wrote:

> www.marsboard.com
>
> The hits just keep on coming....


"MarsBoard A10 Dev Board
An ARM GUN/Linux box"

There WAS a lot of hype a couple months ago about Linux-powered auto aim
for rifles...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130306/86150d5c/attachment.htm>

From george at ceetonetechnology.com  Thu Mar  7 14:37:15 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Thu, 07 Mar 2013 14:37:15 -0500
Subject: [nycbug-talk] list/event policy
Message-ID: <5138EC6B.3000004@ceetonetechnology.com>

We really didn't want to have to state the obvious, but due to a former
list member's sig on this list, we have created a simple policy for
NYC*BUG lists and events.

<policy>
The New York City *BSD User Group strives to create an environment
welcoming to all.

This means we do not tolerate racist, sexist or homophobic actions or
comments anywhere on our mailing lists and at all of our events.

We hope that each and every NYC*BUG participant views this
code of conduct as a reminder to help maintain a respectful
environment and create a welcoming community.
</policy>

Admin@ wasn't aware of the sig in question, but someone on this list
pinged us.  It was greatly appreciated, and we strongly believe that
everyone should jump on the responsibility.

I like to think this policy is already in the guts of most people on
this list, and that merely stating it is a formality.  Nevertheless, it
*is* being stated and will be enforced as it is already.

g


From mike at myownsoho.net  Thu Mar  7 15:07:48 2013
From: mike at myownsoho.net (Mike N.)
Date: Thu, 07 Mar 2013 15:07:48 -0500
Subject: [nycbug-talk] Mt. Kisco computer store aides CIA to combat computer
	virus
Message-ID: <da4d75aa07bc8f6a73276355ffe530c4@myownsoho.net>

 

ok guys, this made my day. There's a computer store in Mt. Kisco
(Westchester County) near, what once was, Borders. It's super
over-priced and probably won't be in business much longer. 

I just had to share this with you. 

Here's the article that i lold my face off at:
http://www.nytimes.com/2010/11/09/nyregion/09fraud.html?_r=0 [1] 

... after hearing about this:
http://www.lohud.com/article/20130305/NEWS02/303050046/Chappaqua-man-sentenced-20-million-swindle
[2] 

from my dad who knows Mr. Davidson personally. 

C-R-A-Z-Y 

I hope this brings a smile to someone's face after George's downer. 

t/c all of you, and please keep your system's security up to date.....
or the priests of Honduras will come to get you. 

-- 
Mike Nichols
c. 347 7251661
 

Links:
------
[1] http://www.nytimes.com/2010/11/09/nyregion/09fraud.html?_r=0
[2]
http://www.lohud.com/article/20130305/NEWS02/303050046/Chappaqua-man-sentenced-20-million-swindle
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130307/fa4ee24e/attachment.htm>

From george at ceetonetechnology.com  Fri Mar  8 20:14:58 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 08 Mar 2013 20:14:58 -0500
Subject: [nycbug-talk] svnup for FreeBSD
Message-ID: <513A8D12.1060205@ceetonetechnology.com>

So it's here and it's hot.

net/svnup

Just started playing with it... and according to Eitan, it may/will be
part of base at some point.

Correct me if you lied to me on IRC Eitan...

;)

g


From akosela at andykosela.com  Fri Mar  8 21:35:04 2013
From: akosela at andykosela.com (Andy Kosela)
Date: Fri, 8 Mar 2013 19:35:04 -0700
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <513A8D12.1060205@ceetonetechnology.com>
References: <513A8D12.1060205@ceetonetechnology.com>
Message-ID: <CALMnNGj7-fTDwW1q_ATkuc5guJqvuw=jY-H0mJWQ0q3+8jKKUQ@mail.gmail.com>

On Fri, Mar 8, 2013 at 6:14 PM, George Rosamond
<george at ceetonetechnology.com> wrote:
> So it's here and it's hot.
>
> net/svnup
>
> Just started playing with it... and according to Eitan, it may/will be
> part of base at some point.
>
> Correct me if you lied to me on IRC Eitan...

Wow.
That's exactly what people wanted.  We got some core kernel developers
on this list too like John Baldwin.  John, any chance this could be a
part of the base in near future?  Seems very reasonable to be so.

--Andy


From pete at nomadlogic.org  Sat Mar  9 00:50:24 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Fri, 08 Mar 2013 21:50:24 -0800
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <513A8D12.1060205@ceetonetechnology.com>
References: <513A8D12.1060205@ceetonetechnology.com>
Message-ID: <513ACDA0.3010606@nomadlogic.org>

On 03/08/2013 05:14 PM, George Rosamond wrote:
> So it's here and it's hot.
>
> net/svnup

awesome - just finished building a pkg for it and testing it now :)

-p

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



From george at ceetonetechnology.com  Sat Mar  9 13:07:29 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sat, 09 Mar 2013 13:07:29 -0500
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <CALMnNGj7-fTDwW1q_ATkuc5guJqvuw=jY-H0mJWQ0q3+8jKKUQ@mail.gmail.com>
References: <513A8D12.1060205@ceetonetechnology.com>
 <CALMnNGj7-fTDwW1q_ATkuc5guJqvuw=jY-H0mJWQ0q3+8jKKUQ@mail.gmail.com>
Message-ID: <513B7A61.6040703@ceetonetechnology.com>

On 03/08/13 21:35, Andy Kosela wrote:
> On Fri, Mar 8, 2013 at 6:14 PM, George Rosamond
> <george at ceetonetechnology.com> wrote:
>> So it's here and it's hot.
>>
>> net/svnup
>>
>> Just started playing with it... and according to Eitan, it may/will be
>> part of base at some point.
>>
>> Correct me if you lied to me on IRC Eitan...
> 
> Wow.
> That's exactly what people wanted.  We got some core kernel developers
> on this list too like John Baldwin.  John, any chance this could be a
> part of the base in near future?  Seems very reasonable to be so.

Well, I think hats off to Eitan too.  He is the youngest person with a
commit bit on this list, AFAIK.  He is one of "our kids" in the BSDs,
and if he's the future, I'm happy.  The 30/40-somethings are too old for
the future.

He listens to the talk@ whining!

g



From okan at demirmen.com  Sat Mar  9 14:03:06 2013
From: okan at demirmen.com (Okan Demirmen)
Date: Sat, 9 Mar 2013 14:03:06 -0500
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <513B7A61.6040703@ceetonetechnology.com>
References: <513A8D12.1060205@ceetonetechnology.com>
 <CALMnNGj7-fTDwW1q_ATkuc5guJqvuw=jY-H0mJWQ0q3+8jKKUQ@mail.gmail.com>
 <513B7A61.6040703@ceetonetechnology.com>
Message-ID: <CACs0eVQh0ep2=N68zf_xbWwF6DDuF+m-JyTZVK9GNANR-gWeTw@mail.gmail.com>

On Sat, Mar 9, 2013 at 1:07 PM, George Rosamond
<george at ceetonetechnology.com> wrote:
> On 03/08/13 21:35, Andy Kosela wrote:
>> On Fri, Mar 8, 2013 at 6:14 PM, George Rosamond
>> <george at ceetonetechnology.com> wrote:
>>> So it's here and it's hot.
>>>
>>> net/svnup
>>>
>>> Just started playing with it... and according to Eitan, it may/will be
>>> part of base at some point.
>>>
>>> Correct me if you lied to me on IRC Eitan...
>>
>> Wow.
>> That's exactly what people wanted.  We got some core kernel developers
>> on this list too like John Baldwin.  John, any chance this could be a
>> part of the base in near future?  Seems very reasonable to be so.
>
> Well, I think hats off to Eitan too.  He is the youngest person with a
> commit bit on this list, AFAIK.  He is one of "our kids" in the BSDs,
> and if he's the future, I'm happy.  The 30/40-somethings are too old for
> the future.

whoa there buddy! age has nothing to do with it :)  dare we go over what
"kids" created that are now on the intergarbage and rate them? :)  I'm
just saying!

No offense Eitan!  ...this George guy is running his mouth with
generalizations..

> He listens to the talk@ whining!

heh, that's part of the fun!

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


From bcallah at devio.us  Sat Mar  9 16:47:05 2013
From: bcallah at devio.us (Brian Callahan)
Date: Sat, 9 Mar 2013 16:47:05 -0500 (EST)
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <513B7A61.6040703@ceetonetechnology.com>
References: <513A8D12.1060205@ceetonetechnology.com>
 <CALMnNGj7-fTDwW1q_ATkuc5guJqvuw=jY-H0mJWQ0q3+8jKKUQ@mail.gmail.com>
 <513B7A61.6040703@ceetonetechnology.com>
Message-ID: <alpine.BSO.2.00.1303091640420.18716@wolfman.devio.us>



On Sat, 9 Mar 2013, George Rosamond wrote:

> On 03/08/13 21:35, Andy Kosela wrote:
>> On Fri, Mar 8, 2013 at 6:14 PM, George Rosamond
>> <george at ceetonetechnology.com> wrote:
>>> So it's here and it's hot.
>>>
>>> net/svnup
>>>
>>> Just started playing with it... and according to Eitan, it may/will be
>>> part of base at some point.
>>>
>>> Correct me if you lied to me on IRC Eitan...
>>
>> Wow.
>> That's exactly what people wanted.  We got some core kernel developers
>> on this list too like John Baldwin.  John, any chance this could be a
>> part of the base in near future?  Seems very reasonable to be so.
>

Cool stuff.
One minor nit: in svnup.1, .Dt svnup(1) should be .Dt svnup 1 (or .Dt 
SVNUP 1), no?
Otherwise, you get svnup(1)(1) at the top of the manpage.

~Brian

> Well, I think hats off to Eitan too.  He is the youngest person with a
> commit bit on this list, AFAIK.  He is one of "our kids" in the BSDs,
> and if he's the future, I'm happy.  The 30/40-somethings are too old for
> the future.
>
> He listens to the talk@ whining!
>
> g


From george at ceetonetechnology.com  Sat Mar  9 17:00:14 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sat, 09 Mar 2013 17:00:14 -0500
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <CACs0eVQh0ep2=N68zf_xbWwF6DDuF+m-JyTZVK9GNANR-gWeTw@mail.gmail.com>
References: <513A8D12.1060205@ceetonetechnology.com>
 <CALMnNGj7-fTDwW1q_ATkuc5guJqvuw=jY-H0mJWQ0q3+8jKKUQ@mail.gmail.com>
 <513B7A61.6040703@ceetonetechnology.com>
 <CACs0eVQh0ep2=N68zf_xbWwF6DDuF+m-JyTZVK9GNANR-gWeTw@mail.gmail.com>
Message-ID: <513BB0EE.6000701@ceetonetechnology.com>

On 03/09/13 14:03, Okan Demirmen wrote:
> On Sat, Mar 9, 2013 at 1:07 PM, George Rosamond
> <george at ceetonetechnology.com> wrote:
>> On 03/08/13 21:35, Andy Kosela wrote:
>>> On Fri, Mar 8, 2013 at 6:14 PM, George Rosamond
>>> <george at ceetonetechnology.com> wrote:
>>>> So it's here and it's hot.
>>>>
>>>> net/svnup
>>>>
>>>> Just started playing with it... and according to Eitan, it may/will be
>>>> part of base at some point.
>>>>
>>>> Correct me if you lied to me on IRC Eitan...
>>>
>>> Wow.
>>> That's exactly what people wanted.  We got some core kernel developers
>>> on this list too like John Baldwin.  John, any chance this could be a
>>> part of the base in near future?  Seems very reasonable to be so.
>>
>> Well, I think hats off to Eitan too.  He is the youngest person with a
>> commit bit on this list, AFAIK.  He is one of "our kids" in the BSDs,
>> and if he's the future, I'm happy.  The 30/40-somethings are too old for
>> the future.
> 
> whoa there buddy! age has nothing to do with it :)  dare we go over what
> "kids" created that are now on the intergarbage and rate them? :)  I'm
> just saying!
>

of course your point is valid, but it misses the thrust of what I was
saying.

The younger people in BSD land are not living in unstructured hackster
land.  I think they are better positioned to recognize the beauty of
cathedrals.

> No offense Eitan!  ...this George guy is running his mouth with
> generalizations..
> 
>> He listens to the talk@ whining!
> 
> heh, that's part of the fun!

:)

g



From ike at blackskyresearch.net  Sun Mar 10 20:31:41 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Sun, 10 Mar 2013 20:31:41 -0400
Subject: [nycbug-talk] svnup for FreeBSD
In-Reply-To: <513ACDA0.3010606@nomadlogic.org>
References: <513A8D12.1060205@ceetonetechnology.com>
 <513ACDA0.3010606@nomadlogic.org>
Message-ID: <1362961922-9262463.70088477.fr2B0Vghb025999@rs149.luxsci.com>

Hi All,

On Mar 9, 2013, at 12:50 AM, Pete Wright wrote:
> On 03/08/2013 05:14 PM, George Rosamond wrote:
>> So it's here and it's hot.
>> 
>> net/svnup
> 
> awesome - just finished building a pkg for it and testing it now :)
> 
> -p


So rad!

I've now both built the *tiny* original source tarball for sunup(1), as well as built the port- it's the best 9.3k tarball I've touched in a year :)

--
If anyone has any info about sunup(1) making it into base, please tell me so I can kill this page:

https://wiki.freebsd.org/UsersFetchingSource

(After trying to catch up on the relevant lists, I've updated that 'status page' on the FreeBSD wiki.)


Rocket-
.ike





From ike at blackskyresearch.net  Mon Mar 11 09:54:57 2013
From: ike at blackskyresearch.net (Isaac Levy (.ike))
Date: Mon, 11 Mar 2013 09:54:57 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
Message-ID: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>

Hi All,

>From prior threads, the FreeBSD svnup(1) utility is alive, all buttoned up, man page and all!

It's currently in ports, but it's not over- I believe it would be EXTREMELY advantageous for this utility to make it's way into the base system (in time for 10x?).  If anyone has any info (or sees any commits), please shout back in this thread!

--
If you believe svnup(1) in base is a good idea, and want to help, I wanted to ask folks for some user testing?  Break it.  Read it's short source, (1310 lines- one file).  Anything you can think of to make it tight as it can be, (worthy of base).

--
My observations so far:

svnup(1) speed "feel"

Without opening up any cans of worms on SCM tools, I believe the 'git effect' creates some unrealistic expectations for the utility.  (Git's impact is so huge on development, it's become so popular, it affects/warps perception of other tools.)  git(1) has spectacular indexing/hashing, it's implementation is really thoughtful compared to svn(1).  So, remote fetching of deltas, on massive codebases, is extremely fast.
svn(1) itself- (and svnup(1)), is not quite as slick in this area.  To this end, svnup(1) *feels* surprisingly "slow" when fetching deltas.  Yet, for what it's doing, (comparing every file), it's pretty darned fast.

Because it appears to walk every file, I'm curious to know how much impact (if any) svnup(1) has when many clients are simultaneously hitting a single repo?  Does anyone have local subversion mirrors setup who can test this?

I'm wondering if anyone can point out where the bottlenecks are, so "best practices" can emerge quickly for setting up local svn repos?

Rocket-
.ike





From george at ceetonetechnology.com  Mon Mar 11 10:17:42 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Mar 2013 10:17:42 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
Message-ID: <513DE786.7070702@ceetonetechnology.com>

On 03/11/13 09:54, Isaac (.ike) Levy wrote:
> Hi All,
> 
> From prior threads, the FreeBSD svnup(1) utility is alive, all
> buttoned up, man page and all!
> 
> It's currently in ports, but it's not over- I believe it would be
> EXTREMELY advantageous for this utility to make it's way into the
> base system (in time for 10x?).  If anyone has any info (or sees any
> commits), please shout back in this thread!

and backported into base all the way to 8.x would be nice too.

> 
> -- If you believe svnup(1) in base is a good idea, and want to help,
> I wanted to ask folks for some user testing?  Break it.  Read it's
> short source, (1310 lines- one file).  Anything you can think of to
> make it tight as it can be, (worthy of base).
> 
> -- My observations so far:
> 
> svnup(1) speed "feel"
> 
> Without opening up any cans of worms on SCM tools, I believe the 'git
> effect' creates some unrealistic expectations for the utility.
> (Git's impact is so huge on development, it's become so popular, it
> affects/warps perception of other tools.)  git(1) has spectacular
> indexing/hashing, it's implementation is really thoughtful compared
> to svn(1).  So, remote fetching of deltas, on massive codebases, is
> extremely fast. svn(1) itself- (and svnup(1)), is not quite as slick
> in this area.  To this end, svnup(1) *feels* surprisingly "slow" when
> fetching deltas.  Yet, for what it's doing, (comparing every file),
> it's pretty darned fast.

So I don't "feel" it's slower than csup/cvsup.. but haven't tested.

> 
> Because it appears to walk every file, I'm curious to know how much
> impact (if any) svnup(1) has when many clients are simultaneously
> hitting a single repo?  Does anyone have local subversion mirrors
> setup who can test this?
> 
> I'm wondering if anyone can point out where the bottlenecks are, so
> "best practices" can emerge quickly for setting up local svn repos?

Breaking would be interesting.  So I didn't blow away the svn created
/usr/ports when I ran it, and nothing broke, and of course the .svn
directory is gone.

g


From ike at blackskyresearch.net  Mon Mar 11 12:12:40 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Mon, 11 Mar 2013 12:12:40 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <513DE786.7070702@ceetonetechnology.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <513DE786.7070702@ceetonetechnology.com>
Message-ID: <1363018385-6584648.44620218.fr2BGCfPl005339@rs149.luxsci.com>

On Mar 11, 2013, at 10:17 AM, George Rosamond wrote:
> On 03/11/13 09:54, Isaac (.ike) Levy wrote:
>> Hi All,
>> 
>> From prior threads, the FreeBSD svnup(1) utility is alive, all
>> buttoned up, man page and all!
>> 
>> It's currently in ports, but it's not over- I believe it would be
>> EXTREMELY advantageous for this utility to make it's way into the
>> base system (in time for 10x?).  If anyone has any info (or sees any
>> commits), please shout back in this thread!
> 
> and backported into base all the way to 8.x would be nice too.

Agreed, but personally I'd rather see the near future sured up.

> 
>> 
>> -- If you believe svnup(1) in base is a good idea, and want to help,
>> I wanted to ask folks for some user testing?  Break it.  Read it's
>> short source, (1310 lines- one file).  Anything you can think of to
>> make it tight as it can be, (worthy of base).
>> 
>> -- My observations so far:
>> 
>> svnup(1) speed "feel"
>> 
>> Without opening up any cans of worms on SCM tools, I believe the 'git
>> effect' creates some unrealistic expectations for the utility.
>> (Git's impact is so huge on development, it's become so popular, it
>> affects/warps perception of other tools.)  git(1) has spectacular
>> indexing/hashing, it's implementation is really thoughtful compared
>> to svn(1).  So, remote fetching of deltas, on massive codebases, is
>> extremely fast. svn(1) itself- (and svnup(1)), is not quite as slick
>> in this area.  To this end, svnup(1) *feels* surprisingly "slow" when
>> fetching deltas.  Yet, for what it's doing, (comparing every file),
>> it's pretty darned fast.
> 
> So I don't "feel" it's slower than csup/cvsup.. but haven't tested.

Compared to csup/cvsup, it does indeed feel similar- (which is totally acceptable to me).

>> Because it appears to walk every file, I'm curious to know how much
>> impact (if any) svnup(1) has when many clients are simultaneously
>> hitting a single repo?  Does anyone have local subversion mirrors
>> setup who can test this?
>> 
>> I'm wondering if anyone can point out where the bottlenecks are, so
>> "best practices" can emerge quickly for setting up local svn repos?
> 
> Breaking would be interesting.  

Are there any NYC*BUG resources to try testing this?  (e.g. set up an SVN source mirror and hammer it like crazy)  If not, I may be able to get hold of a box to test from?  Not sure if it's worth the effort(?)

> So I didn't blow away the svn created
> /usr/ports when I ran it, and nothing broke, and of course the .svn
> directory is gone.
> 
> g

Yeah- the thing certainly behaves within the expectations of c[v]sup!  Really exciting, really simple.

Rocket-
.ike





From freebsd-listen at fabiankeil.de  Mon Mar 11 11:31:11 2013
From: freebsd-listen at fabiankeil.de (Fabian Keil)
Date: Mon, 11 Mar 2013 16:31:11 +0100
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
Message-ID: <20130311163111.56ef0e02@fabiankeil.de>

Isaac (.ike) Levy <ike at blackskyresearch.net> wrote:

> --
> My observations so far:
> 
> svnup(1) speed "feel"
> 
> Without opening up any cans of worms on SCM tools, I believe the 'git effect' creates some unrealistic expectations for the utility.  (Git's impact is so huge on development, it's become so popular, it affects/warps perception of other tools.)  git(1) has spectacular indexing/hashing, it's implementation is really thoughtful compared to svn(1).  So, remote fetching of deltas, on massive codebases, is extremely fast.

But actually checking out the fetched deltas scales with
the repository size.

> svn(1) itself- (and svnup(1)), is not quite as slick in this area.  To this end, svnup(1) *feels* surprisingly "slow" when fetching deltas.  Yet, for what it's doing, (comparing every file), it's pretty darned fast.

Did you use git for base, ports or other projects with similar size?

I'm not claiming that git isn't faster than svnup (which I never used),
but a lot of people seem to only use git for small projects and expect
it to perform similar well with larger projects.

Once upon a time I did that too, but then I started using git for both
ports and base on my admittedly old laptop.

Until the ARC is warm some operations are depressingly slow because git
lstat()s every monitored file (and even when you know better you can't
trivially tell git not to). To give you an example:

fk at r500 /usr/ports $time git fetch
remote: Counting objects: 1433, done.
remote: Compressing objects: 100% (508/508), done.
remote: Total 1014 (delta 599), reused 906 (delta 502)
Receiving objects: 100% (1014/1014), 936.97 KiB | 168 KiB/s, done.
Resolving deltas: 100% (599/599), completed with 230 local objects.
From git://github.com/freebsd/freebsd-ports
   a486bc5..e355038  master     -> origin/master
   10e61e6..c4e0181  svn_head   -> origin/svn_head

real	0m48.982s
user	0m1.064s
sys	0m0.699s
fk at r500 /usr/ports $time git checkout master
Switched to branch 'master'
Your branch is behind 'origin/master' by 131 commits, and can be fast-forwarded.
  (use "git pull" to update your local branch)

real	4m51.360s
user	0m2.090s
sys	0m8.008s
fk at r500 /usr/ports $time git rebase origin/master
First, rewinding head to replay your work on top of it...
Fast-forwarded master to origin/master.

real	6m0.862s
user	0m2.308s
sys	0m12.356s

Fast-forwarding 131 commits in 6 minutes probably isn't the
kind of spectacular you were referring to.

It's reasonable once the ARC is warm:

fk at r500 /usr/ports $time git pull
remote: Counting objects: 59, done.
remote: Compressing objects: 100% (17/17), done.
remote: Total 39 (delta 26), reused 35 (delta 22)
Unpacking objects: 100% (39/39), done.
From git://github.com/freebsd/freebsd-ports
   e355038..dc0fb02  master     -> origin/master
   c4e0181..532d005  svn_head   -> origin/svn_head
Updating e355038..dc0fb02
Fast-forward
 databases/rubygem-familia/pkg-descr | 1 +
 databases/rubygem-redis/Makefile    | 2 +-
 databases/rubygem-redis/distinfo    | 4 ++--
 devel/rubygem-stella/pkg-descr      | 1 +
 graphics/ruby-gdal/Makefile         | 5 ++---
 www/rubygem-em-websocket/Makefile   | 2 +-
 www/rubygem-em-websocket/distinfo   | 4 ++--
 www/wgetpaste/Makefile              | 8 ++------
 www/wgetpaste/distinfo              | 4 ++--
 9 files changed, 14 insertions(+), 17 deletions(-)

real	0m16.813s
user	0m1.823s
sys	0m3.183s

But then again, I reboot the laptop at least once a day
and the ARC isn't persistent (yet) ...

Fabian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130311/300fdaea/attachment.bin>

From ike at blackskyresearch.net  Mon Mar 11 12:58:34 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Mon, 11 Mar 2013 12:58:34 -0400
Subject: [nycbug-talk] Git Tangent: [Was: svnup(1) - worthy of promotion to
	base?]
In-Reply-To: <20130311163111.56ef0e02@fabiankeil.de>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
Message-ID: <1363021143-8239092.49411088.fr2BGwYLV028858@rs149.luxsci.com>

On Mar 11, 2013, at 11:31 AM, Fabian Keil wrote:
> Isaac (.ike) Levy <ike at blackskyresearch.net> wrote:
> 
>> --
>> My observations so far:
>> 
>> svnup(1) speed "feel"
>> 
>> Without opening up any cans of worms on SCM tools, I believe the 'git effect' creates some unrealistic expectations for the utility.  (Git's impact is so huge on development, it's become so popular, it affects/warps perception of other tools.)  git(1) has spectacular indexing/hashing, it's implementation is really thoughtful compared to svn(1).  So, remote fetching of deltas, on massive codebases, is extremely fast.
> 
> But actually checking out the fetched deltas scales with
> the repository size.

Absolutely- files read remote, files over network, files local write/remove/resolve- doesn't change from SCM to SCM.

> 
>> svn(1) itself- (and svnup(1)), is not quite as slick in this area.  To this end, svnup(1) *feels* surprisingly "slow" when fetching deltas.  Yet, for what it's doing, (comparing every file), it's pretty darned fast.
> 
> Did you use git for base, ports or other projects with similar size?

Only toyed with it for base/ports.

I've used git in other codebases, mush larger than FreeBSD.  This year I've been living in a git repo with 38M+ lines of various software packages in < 3Gb, including gnarly bad-form binaries stuff in there, and other general repo abuses.

This repo is *not* fast to work with, but using it, it's at least fast enough to be workable.  Anecdotally:

- Fetching the entire repo over fast network: 20-30 minutes. (expected)
- Running 'fetch' to check deltas against remote repo: typically < 2-8 seconds. (the awesomeness)

The fetch time is of course dependent on how many changes there have been upstream, how many branches added, etc?  But for a codebase this large, the 'index as it goes' approach of git yields pretty impressive results when the deltas themselves are minimal.

This is not the case for SVN, apparently, based on the design of the protocol and underlying data storage.  In essence, git is comparing hash strings of directory indexes- svn is comparing files (and metadata) individually.

--
On a different git backed codebase I work in: 2.8m lines (which grows an average of 3 branches a day- fast-foreward-only workflow), it's still mere seconds to check/fetch deltas.

Etc?  Git is really quite impressive on all these fronts- to bad it's getting so obese in the feature implementation department (build it from source one day, you'll see what I mean :)

> I'm not claiming that git isn't faster than svnup (which I never used),
> but a lot of people seem to only use git for small projects and expect
> it to perform similar well with larger projects.

Absolutely agree, in particular as github is the new internet black?  and most repos are (thankfully) quite small.

> 
> Once upon a time I did that too, but then I started using git for both
> ports and base on my admittedly old laptop.
> 
> Until the ARC is warm some operations are depressingly slow because git
> lstat()s every monitored file (and even when you know better you can't
> trivially tell git not to). To give you an example:
> 
> fk at r500 /usr/ports $time git fetch
> remote: Counting objects: 1433, done.
> remote: Compressing objects: 100% (508/508), done.
> remote: Total 1014 (delta 599), reused 906 (delta 502)
> Receiving objects: 100% (1014/1014), 936.97 KiB | 168 KiB/s, done.
> Resolving deltas: 100% (599/599), completed with 230 local objects.
> From git://github.com/freebsd/freebsd-ports
>   a486bc5..e355038  master     -> origin/master
>   10e61e6..c4e0181  svn_head   -> origin/svn_head
> 
> real	0m48.982s
> user	0m1.064s
> sys	0m0.699s
> fk at r500 /usr/ports $time git checkout master
> Switched to branch 'master'
> Your branch is behind 'origin/master' by 131 commits, and can be fast-forwarded.
>  (use "git pull" to update your local branch)
> 
> real	4m51.360s
> user	0m2.090s
> sys	0m8.008s
> fk at r500 /usr/ports $time git rebase origin/master
> First, rewinding head to replay your work on top of it...
> Fast-forwarded master to origin/master.
> 
> real	6m0.862s
> user	0m2.308s
> sys	0m12.356s
> 
> Fast-forwarding 131 commits in 6 minutes probably isn't the
> kind of spectacular you were referring to.

Ha- no :)

> 
> It's reasonable once the ARC is warm:
> 
> fk at r500 /usr/ports $time git pull
> remote: Counting objects: 59, done.
> remote: Compressing objects: 100% (17/17), done.
> remote: Total 39 (delta 26), reused 35 (delta 22)
> Unpacking objects: 100% (39/39), done.
> From git://github.com/freebsd/freebsd-ports
>   e355038..dc0fb02  master     -> origin/master
>   c4e0181..532d005  svn_head   -> origin/svn_head
> Updating e355038..dc0fb02
> Fast-forward
> databases/rubygem-familia/pkg-descr | 1 +
> databases/rubygem-redis/Makefile    | 2 +-
> databases/rubygem-redis/distinfo    | 4 ++--
> devel/rubygem-stella/pkg-descr      | 1 +
> graphics/ruby-gdal/Makefile         | 5 ++---
> www/rubygem-em-websocket/Makefile   | 2 +-
> www/rubygem-em-websocket/distinfo   | 4 ++--
> www/wgetpaste/Makefile              | 8 ++------
> www/wgetpaste/distinfo              | 4 ++--
> 9 files changed, 14 insertions(+), 17 deletions(-)
> 
> real	0m16.813s
> user	0m1.823s
> sys	0m3.183s
> 
> But then again, I reboot the laptop at least once a day
> and the ARC isn't persistent (yet) ?

And also since ARC is for ZFS, how much physical memory do you have on that box?

Just curious- (mostly because I'm amazed to see how much more feasable using ZFS has become with smaller memory footprints).

Could you perhaps share a dmesg, out of curiosity?
http://www.nycbug.org/?action=dmesgd

--
Realistically, my point is that the git internals help it blow away other open source SCM tools in delta sync- and for many new/future FreeBSD users, svnup(1) may be perceived as slow.
So, anything we can do in advance to identify and resolve any speed issues which are real- and within reach to resolve, will make this important tool stand up much better- IMHO.

I'm not meaning to compare- but I do want svnup(1) to stand solid.

Rocket-
.ike





From mark.saad at ymail.com  Mon Mar 11 14:45:19 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Mon, 11 Mar 2013 14:45:19 -0400
Subject: [nycbug-talk] NYC*BUG Now has a pkgng repo for 9.1-RELEASE via
	PC-BSD
Message-ID: <CAMXt9Na3xt1ribAm4XbdZk4iyw5kGofpUDkx+emq=Cv1hy6FWQ@mail.gmail.com>

All
 I am pleased to announce we now have a pkgng repo for 9.1-RELEASE via
our PC-BSD Mirror.
We are mirroring the main PC-BSD repository so please refer to their
site for the update intervals.

To use this on FreeBSD 9.1 Please set the following in your pkg.conf

packagesite: http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/{amd64,i386}
# One of the two options i386 or amd64 not the {}'s

PUBKEY: /usr/local/etc/pkg-pubkey.cert
PKG_CACHEDIR: /usr/local/tmp


You will also need the pc-bsd pub key found here
http://trac.pcbsd.org/export/21629/pcbsd/current/src-sh/pc-extractoverlay/desktop-overlay/usr/local/etc/pkg-pubkey.cert



-- 

Mark Saad | mark.saad at ymail.com


From pete at nomadlogic.org  Mon Mar 11 15:02:25 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Mon, 11 Mar 2013 12:02:25 -0700
Subject: [nycbug-talk] NYC*BUG Now has a pkgng repo for 9.1-RELEASE via
 PC-BSD
In-Reply-To: <CAMXt9Na3xt1ribAm4XbdZk4iyw5kGofpUDkx+emq=Cv1hy6FWQ@mail.gmail.com>
References: <CAMXt9Na3xt1ribAm4XbdZk4iyw5kGofpUDkx+emq=Cv1hy6FWQ@mail.gmail.com>
Message-ID: <513E2A41.7030004@nomadlogic.org>

On 03/11/13 11:45, Mark Saad wrote:
> All
>   I am pleased to announce we now have a pkgng repo for 9.1-RELEASE via
> our PC-BSD Mirror.
> We are mirroring the main PC-BSD repository so please refer to their
> site for the update intervals.
>
> To use this on FreeBSD 9.1 Please set the following in your pkg.conf
>
> packagesite: http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/{amd64,i386}
> # One of the two options i386 or amd64 not the {}'s
>
> PUBKEY: /usr/local/etc/pkg-pubkey.cert
> PKG_CACHEDIR: /usr/local/tmp
>
>
> You will also need the pc-bsd pub key found here
> http://trac.pcbsd.org/export/21629/pcbsd/current/src-sh/pc-extractoverlay/desktop-overlay/usr/local/etc/pkg-pubkey.cert
>
>

this is awesome - nice one guys!

-pete

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



From pete at nomadlogic.org  Mon Mar 11 15:06:30 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Mon, 11 Mar 2013 12:06:30 -0700
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <20130311163111.56ef0e02@fabiankeil.de>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
Message-ID: <513E2B36.8040906@nomadlogic.org>

On 03/11/13 08:31, Fabian Keil wrote:
> Isaac (.ike) Levy <ike at blackskyresearch.net> wrote:
>
>> --
>> My observations so far:
>>
>> svnup(1) speed "feel"
>>
>> Without opening up any cans of worms on SCM tools, I believe the 'git effect' creates some unrealistic expectations for the utility.  (Git's impact is so huge on development, it's become so popular, it affects/warps perception of other tools.)  git(1) has spectacular indexing/hashing, it's implementation is really thoughtful compared to svn(1).  So, remote fetching of deltas, on massive codebases, is extremely fast.
> But actually checking out the fetched deltas scales with
> the repository size.

i can attest to this unfortunately - been struggling with a poorly 
thought out production system where we use git to replicate data to all 
our POPs.  lets just say when re-calculating your repo's hashes takes 
close to an hour you've made some terrible decisions :)

but back to your point, calculating checksums would most likely be a 
non-trivial task for the entire FreeBSD repository 
(usr+work+ports+gnats+www etc...).

-pete





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



From freebsd-listen at fabiankeil.de  Mon Mar 11 15:09:29 2013
From: freebsd-listen at fabiankeil.de (Fabian Keil)
Date: Mon, 11 Mar 2013 20:09:29 +0100
Subject: [nycbug-talk] Git Tangent: [Was: svnup(1) - worthy of promotion
 to base?]
In-Reply-To: <1363021143-8239092.49411088.fr2BGwYLV028858@rs149.luxsci.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
 <1363021143-8239092.49411088.fr2BGwYLV028858@rs149.luxsci.com>
Message-ID: <20130311200929.4c6a4d48@fabiankeil.de>

"Isaac (.ike) Levy" <ike at blackskyresearch.net> wrote:

> On Mar 11, 2013, at 11:31 AM, Fabian Keil wrote:
> > Isaac (.ike) Levy <ike at blackskyresearch.net> wrote:

> >> svn(1) itself- (and svnup(1)), is not quite as slick in this area.  To this end, svnup(1) *feels* surprisingly "slow" when fetching deltas.  Yet, for what it's doing, (comparing every file), it's pretty darned fast.
> > 
> > Did you use git for base, ports or other projects with similar size?
> 
> Only toyed with it for base/ports.
[...]
> Etc?  Git is really quite impressive on all these fronts- to bad it's getting so obese in the feature implementation department (build it from source one day, you'll see what I mean :)

Without the svn support most of the obesity seems to be gone.
And building everything including clang and Firefox from source
helps to keeps things in perspective ...

> > It's reasonable once the ARC is warm:
[...]
> > But then again, I reboot the laptop at least once a day
> > and the ARC isn't persistent (yet) ?
> 
> And also since ARC is for ZFS, how much physical memory do you have on that box?

2 GB. For this workload I mainly blame the slow disk, though.

> Just curious- (mostly because I'm amazed to see how much more feasable using ZFS has become with smaller memory footprints).

After a bit of tuning it always worked for me.

> Could you perhaps share a dmesg, out of curiosity?
> http://www.nycbug.org/?action=dmesgd

Sure.
http://www.nycbug.org/?action=dmesgd&dmesgid=2449

> Realistically, my point is that the git internals help it blow away other open source SCM tools in delta sync- and for many new/future FreeBSD users, svnup(1) may be perceived as slow.
> So, anything we can do in advance to identify and resolve any speed issues which are real- and within reach to resolve, will make this important tool stand up much better- IMHO.

Makes sense.

Fabian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130311/035026d5/attachment.bin>

From ike at blackskyresearch.net  Mon Mar 11 15:33:13 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Mon, 11 Mar 2013 15:33:13 -0400
Subject: [nycbug-talk] NYC*BUG Now has a pkgng repo for 9.1-RELEASE via
	PC-BSD
In-Reply-To: <513E2A41.7030004@nomadlogic.org>
References: <CAMXt9Na3xt1ribAm4XbdZk4iyw5kGofpUDkx+emq=Cv1hy6FWQ@mail.gmail.com>
 <513E2A41.7030004@nomadlogic.org>
Message-ID: <1363030444-1796219.66511905.fr2BJXDlv011460@rs149.luxsci.com>

On Mar 11, 2013, at 3:02 PM, Pete Wright wrote:
> On 03/11/13 11:45, Mark Saad wrote:
>> All
>>  I am pleased to announce we now have a pkgng repo for 9.1-RELEASE via
>> our PC-BSD Mirror.
>> We are mirroring the main PC-BSD repository so please refer to their
>> site for the update intervals.
>> 
>> To use this on FreeBSD 9.1 Please set the following in your pkg.conf
>> 
>> packagesite: http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/{amd64,i386}
>> # One of the two options i386 or amd64 not the {}'s
>> 
>> PUBKEY: /usr/local/etc/pkg-pubkey.cert
>> PKG_CACHEDIR: /usr/local/tmp
>> 
>> 
>> You will also need the pc-bsd pub key found here
>> http://trac.pcbsd.org/export/21629/pcbsd/current/src-sh/pc-extractoverlay/desktop-overlay/usr/local/etc/pkg-pubkey.cert
>> 
>> 
> 
> this is awesome - nice one guys!
> 
> -pete

I'll second that!  Really awesome...

Rocket-
.ike





From ike at blackskyresearch.net  Mon Mar 11 15:42:39 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Mon, 11 Mar 2013 15:42:39 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <513E2B36.8040906@nomadlogic.org>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de> <513E2B36.8040906@nomadlogic.org>
Message-ID: <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>

On Mar 11, 2013, at 3:06 PM, Pete Wright wrote:
> On 03/11/13 08:31, Fabian Keil wrote:
>> Isaac (.ike) Levy <ike at blackskyresearch.net> wrote:
>> 
>>> --
>>> My observations so far:
>>> 
>>> svnup(1) speed "feel"
>>> 
>>> Without opening up any cans of worms on SCM tools, I believe the 'git effect' creates some unrealistic expectations for the utility.  (Git's impact is so huge on development, it's become so popular, it affects/warps perception of other tools.)  git(1) has spectacular indexing/hashing, it's implementation is really thoughtful compared to svn(1).  So, remote fetching of deltas, on massive codebases, is extremely fast.
>> But actually checking out the fetched deltas scales with
>> the repository size.
> 
> i can attest to this unfortunately - been struggling with a poorly thought out production system where we use git to replicate data to all our POPs.  lets just say when re-calculating your repo's hashes takes close to an hour you've made some terrible decisions :)

Ohboy, the allure of solving problems with a source code repo bites again.
I feel you on that one man...

> 
> but back to your point, calculating checksums would most likely be a non-trivial task for the entire FreeBSD repository (usr+work+ports+gnats+www etc...).
> 
> -pete


Just in case you want to play with the 'feel', TOTALLY NOT an official repo, but public pull access if you want to hack on the FreeBSD src for size and 'feel' comparisons:
https://github.com/freebsd/freebsd

Rocket-
.ike





From ike at blackskyresearch.net  Mon Mar 11 15:47:13 2013
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Mon, 11 Mar 2013 15:47:13 -0400
Subject: [nycbug-talk] Git Tangent: [Was: svnup(1) - worthy of promotion
	to base?]
In-Reply-To: <20130311200929.4c6a4d48@fabiankeil.de>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
 <1363021143-8239092.49411088.fr2BGwYLV028858@rs149.luxsci.com>
 <20130311200929.4c6a4d48@fabiankeil.de>
Message-ID: <1363031285-9612595.53050213.fr2BJlD4D006250@rs149.luxsci.com>

On Mar 11, 2013, at 3:09 PM, Fabian Keil wrote:
> "Isaac (.ike) Levy" <ike at blackskyresearch.net> wrote:
>> On Mar 11, 2013, at 11:31 AM, Fabian Keil wrote:
>>> Git is really quite impressive on all these fronts- to bad it's getting so obese in the feature implementation department (build it from source one day, you'll see what I mean :)
> 
> Without the svn support most of the obesity seems to be gone.
> And building everything including clang and Firefox from source
> helps to keeps things in perspective ?

Well, perhaps clang is large- but I don't believe it requires both python and perl to build 'feature-expect complete' :)

Git was pretty awesome until 2010, it was a few MB, all C, compiled from source nearly everywhere- (original git was small but not so natively portable).

That kind of simplicity made me like it- but alas...

> 
>>> It's reasonable once the ARC is warm:
> [...]
>>> But then again, I reboot the laptop at least once a day
>>> and the ARC isn't persistent (yet) ?
>> 
>> And also since ARC is for ZFS, how much physical memory do you have on that box?
> 
> 2 GB. For this workload I mainly blame the slow disk, though.

Cool :)

> 
>> Just curious- (mostly because I'm amazed to see how much more feasable using ZFS has become with smaller memory footprints).
> 
> After a bit of tuning it always worked for me.

That's pretty awesome- do you have, or could you recommend some notes?

I've tried ZFS on boxes with <4gb memory, and failed miserably at making them run reasonably...

> 
>> Could you perhaps share a dmesg, out of curiosity?
>> http://www.nycbug.org/?action=dmesgd
> 
> Sure.
> http://www.nycbug.org/?action=dmesgd&dmesgid=2449

Cool?  Thanks!!!  That's a respectable dmesg on several levels...

Rocket-
.ike





From mspitzer at gmail.com  Mon Mar 11 17:23:33 2013
From: mspitzer at gmail.com (Marc Spitzer)
Date: Mon, 11 Mar 2013 17:23:33 -0400
Subject: [nycbug-talk] NYC*BUG Now has a pkgng repo for 9.1-RELEASE via
	PC-BSD
In-Reply-To: <1363030444-1796219.66511905.fr2BJXDlv011460@rs149.luxsci.com>
References: <CAMXt9Na3xt1ribAm4XbdZk4iyw5kGofpUDkx+emq=Cv1hy6FWQ@mail.gmail.com>
 <513E2A41.7030004@nomadlogic.org>
 <1363030444-1796219.66511905.fr2BJXDlv011460@rs149.luxsci.com>
Message-ID: <CA+qr9e4E7u-CV2DS4wEp522Kbxw0foiRRoe_4SwH4wNKuy8dRA@mail.gmail.com>

I third thanks for putting this up.

marc


On Mon, Mar 11, 2013 at 3:33 PM, Isaac (.ike) Levy <ike at blackskyresearch.net
> wrote:

> On Mar 11, 2013, at 3:02 PM, Pete Wright wrote:
> > On 03/11/13 11:45, Mark Saad wrote:
> >> All
> >>  I am pleased to announce we now have a pkgng repo for 9.1-RELEASE via
> >> our PC-BSD Mirror.
> >> We are mirroring the main PC-BSD repository so please refer to their
> >> site for the update intervals.
> >>
> >> To use this on FreeBSD 9.1 Please set the following in your pkg.conf
> >>
> >> packagesite:
> http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/{amd64,i386}
> >> # One of the two options i386 or amd64 not the {}'s
> >>
> >> PUBKEY: /usr/local/etc/pkg-pubkey.cert
> >> PKG_CACHEDIR: /usr/local/tmp
> >>
> >>
> >> You will also need the pc-bsd pub key found here
> >>
> http://trac.pcbsd.org/export/21629/pcbsd/current/src-sh/pc-extractoverlay/desktop-overlay/usr/local/etc/pkg-pubkey.cert
> >>
> >>
> >
> > this is awesome - nice one guys!
> >
> > -pete
>
> I'll second that!  Really awesome...
>
> Rocket-
> .ike
>
>
>
> _______________________________________________
> 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/20130311/8966fd7c/attachment.htm>

From george at ceetonetechnology.com  Mon Mar 11 21:02:45 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Mar 2013 21:02:45 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de> <513E2B36.8040906@nomadlogic.org>
 <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>
Message-ID: <513E7EB5.3070909@ceetonetechnology.com>

On 03/11/13 15:42, Isaac (.ike) Levy wrote:
> On Mar 11, 2013, at 3:06 PM, Pete Wright wrote:
>> On 03/11/13 08:31, Fabian Keil wrote:
>>> Isaac (.ike) Levy <ike at blackskyresearch.net> wrote:
>>>
>>>> --
>>>> My observations so far:
>>>>
>>>> svnup(1) speed "feel"
>>>>
>>>> Without opening up any cans of worms on SCM tools, I believe the 'git effect' creates some unrealistic expectations for the utility.  (Git's impact is so huge on development, it's become so popular, it affects/warps perception of other tools.)  git(1) has spectacular indexing/hashing, it's implementation is really thoughtful compared to svn(1).  So, remote fetching of deltas, on massive codebases, is extremely fast.
>>> But actually checking out the fetched deltas scales with
>>> the repository size.
>>
>> i can attest to this unfortunately - been struggling with a poorly thought out production system where we use git to replicate data to all our POPs.  lets just say when re-calculating your repo's hashes takes close to an hour you've made some terrible decisions :)
> 


Jumping back into svnup(1)... quick and undigested observation.

So ran it on a few boxes for the past few days, but found a nasty quirk
on... a laptop.

Libreoffice died during building previously, leaving its enormous work/
full of, well, work.

svnup of course wants to clean this up.

So not so great with preexisting messes?

Hmmm.  make clean now a prerequisite?

g



From gjb at FreeBSD.org  Mon Mar 11 21:08:27 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Mon, 11 Mar 2013 21:08:27 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <513E7EB5.3070909@ceetonetechnology.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
 <513E2B36.8040906@nomadlogic.org>
 <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>
 <513E7EB5.3070909@ceetonetechnology.com>
Message-ID: <20130312010827.GC1543@glenbarber.us>

On Mon, Mar 11, 2013 at 09:02:45PM -0400, George Rosamond wrote:
> Libreoffice died during building previously, leaving its enormous work/
> full of, well, work.
> 
> svnup of course wants to clean this up.
> 
> So not so great with preexisting messes?
> 
> Hmmm.  make clean now a prerequisite?
> 

You can use WRKDIRPREFIX in make.conf(5) to keep your tree pristine.
I do this just for that reason.

    %% make.conf %%
    WRKDIRPREFIX?=	/dumpster/scratch
    NOCLEANDEPENDS=	yes
    PACKAGES=		/dumpster/packages/All
    DISTDIR=		/dumpster/distfiles

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/20130311/f45c67a1/attachment.bin>

From george at ceetonetechnology.com  Mon Mar 11 21:15:05 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Mar 2013 21:15:05 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <20130312010827.GC1543@glenbarber.us>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de> <513E2B36.8040906@nomadlogic.org>
 <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>
 <513E7EB5.3070909@ceetonetechnology.com>
 <20130312010827.GC1543@glenbarber.us>
Message-ID: <513E8199.7040604@ceetonetechnology.com>

On 03/11/13 21:08, Glen Barber wrote:
> On Mon, Mar 11, 2013 at 09:02:45PM -0400, George Rosamond wrote:
>> Libreoffice died during building previously, leaving its enormous work/
>> full of, well, work.
>>
>> svnup of course wants to clean this up.
>>
>> So not so great with preexisting messes?
>>
>> Hmmm.  make clean now a prerequisite?
>>
> 
> You can use WRKDIRPREFIX in make.conf(5) to keep your tree pristine.
> I do this just for that reason.
> 
>     %% make.conf %%
>     WRKDIRPREFIX?=	/dumpster/scratch
>     NOCLEANDEPENDS=	yes
>     PACKAGES=		/dumpster/packages/All
>     DISTDIR=		/dumpster/distfiles

Nice.

But it's a noted default difference from csup.

g



From gjb at FreeBSD.org  Mon Mar 11 21:20:16 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Mon, 11 Mar 2013 21:20:16 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <513E8199.7040604@ceetonetechnology.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
 <513E2B36.8040906@nomadlogic.org>
 <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>
 <513E7EB5.3070909@ceetonetechnology.com>
 <20130312010827.GC1543@glenbarber.us>
 <513E8199.7040604@ceetonetechnology.com>
Message-ID: <20130312012016.GD1543@glenbarber.us>

On Mon, Mar 11, 2013 at 09:15:05PM -0400, George Rosamond wrote:
> On 03/11/13 21:08, Glen Barber wrote:
> > On Mon, Mar 11, 2013 at 09:02:45PM -0400, George Rosamond wrote:
> >> Libreoffice died during building previously, leaving its enormous work/
> >> full of, well, work.
> >>
> >> svnup of course wants to clean this up.
> >>
> >> So not so great with preexisting messes?
> >>
> >> Hmmm.  make clean now a prerequisite?
> >>
> > 
> > You can use WRKDIRPREFIX in make.conf(5) to keep your tree pristine.
> > I do this just for that reason.
> > 
> >     %% make.conf %%
> >     WRKDIRPREFIX?=	/dumpster/scratch
> >     NOCLEANDEPENDS=	yes
> >     PACKAGES=		/dumpster/packages/All
> >     DISTDIR=		/dumpster/distfiles
> 
> Nice.
> 
> But it's a noted default difference from csup.
> 

Sure.  Not saying it's a "fix", but a workaround for those just as
pedantic as me. :-)

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/20130311/536d1a50/attachment.bin>

From george at ceetonetechnology.com  Mon Mar 11 21:23:08 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Mon, 11 Mar 2013 21:23:08 -0400
Subject: [nycbug-talk] svnup(1) - worthy of promotion to base?
In-Reply-To: <20130312012016.GD1543@glenbarber.us>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de> <513E2B36.8040906@nomadlogic.org>
 <1363030985-751439.338438689.fr2BJgdd4031441@rs149.luxsci.com>
 <513E7EB5.3070909@ceetonetechnology.com>
 <20130312010827.GC1543@glenbarber.us>
 <513E8199.7040604@ceetonetechnology.com>
 <20130312012016.GD1543@glenbarber.us>
Message-ID: <513E837C.2090904@ceetonetechnology.com>

On 03/11/13 21:20, Glen Barber wrote:
> On Mon, Mar 11, 2013 at 09:15:05PM -0400, George Rosamond wrote:
>> On 03/11/13 21:08, Glen Barber wrote:
>>> On Mon, Mar 11, 2013 at 09:02:45PM -0400, George Rosamond wrote:
>>>> Libreoffice died during building previously, leaving its enormous work/
>>>> full of, well, work.
>>>>
>>>> svnup of course wants to clean this up.
>>>>
>>>> So not so great with preexisting messes?
>>>>
>>>> Hmmm.  make clean now a prerequisite?
>>>>
>>>
>>> You can use WRKDIRPREFIX in make.conf(5) to keep your tree pristine.
>>> I do this just for that reason.
>>>
>>>     %% make.conf %%
>>>     WRKDIRPREFIX?=	/dumpster/scratch
>>>     NOCLEANDEPENDS=	yes
>>>     PACKAGES=		/dumpster/packages/All
>>>     DISTDIR=		/dumpster/distfiles
>>
>> Nice.
>>
>> But it's a noted default difference from csup.
>>
> 
> Sure.  Not saying it's a "fix", but a workaround for those just as
> pedantic as me. :-)

you're "pedantic"?  Did you witness the rants about FBSD changes on
talk@?  We are renaming list to pedantic-uptight-nyc at lists.nycbug.org,
or alias from WoodyAllenBSD@

In all seriousness, Ike was looking for *any* notable comments, and I
threw that in the mix.

g



From mark.saad at ymail.com  Tue Mar 12 11:15:02 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Tue, 12 Mar 2013 08:15:02 -0700 (PDT)
Subject: [nycbug-talk] I am at datacenter dynamics anyone else here ?
Message-ID: <1363101302.53127.YahooMailMobile@web122204.mail.ne1.yahoo.com>

Hi talk ,
  Anyone else at datacenter dynamics today ? 

---
Mark saad | mark.saad at ymail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130312/f623c490/attachment.htm>

From mark.saad at ymail.com  Wed Mar 13 12:43:05 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Wed, 13 Mar 2013 12:43:05 -0400
Subject: [nycbug-talk] QMail DNS Any lookup issues
Message-ID: <CAMXt9Nao_hqMrCvjanHPQKaNNypx=bF4dhMh_6h885AJHkAXwA@mail.gmail.com>

All
 I need to quickly fix a qmail dns lookup bug.  Basic issue is the dns
lookup for type any no longer works for hosts using ultradns . For
whatever reason they will not respond for queries of type any.

There is a quick discussion here

http://fanf.livejournal.com/122220.html


So does anyone have a working solution to this , other then gut qmail
and install something else ?

-- 

Mark Saad | mark.saad at ymail.com


From freebsd-listen at fabiankeil.de  Wed Mar 13 13:25:48 2013
From: freebsd-listen at fabiankeil.de (Fabian Keil)
Date: Wed, 13 Mar 2013 18:25:48 +0100
Subject: [nycbug-talk] Git Tangent: [Was: svnup(1) - worthy of promotion
 to base?]
In-Reply-To: <1363031285-9612595.53050213.fr2BJlD4D006250@rs149.luxsci.com>
References: <1363010104-2555463.7380363.fr2BDsvtX019627@rs149.luxsci.com>
 <20130311163111.56ef0e02@fabiankeil.de>
 <1363021143-8239092.49411088.fr2BGwYLV028858@rs149.luxsci.com>
 <20130311200929.4c6a4d48@fabiankeil.de>
 <1363031285-9612595.53050213.fr2BJlD4D006250@rs149.luxsci.com>
Message-ID: <20130313182548.5ecdedea@fabiankeil.de>

"Isaac (.ike) Levy" <ike at blackskyresearch.net> wrote:

> On Mar 11, 2013, at 3:09 PM, Fabian Keil wrote:
> > "Isaac (.ike) Levy" <ike at blackskyresearch.net> wrote:
> >> On Mar 11, 2013, at 11:31 AM, Fabian Keil wrote:
> >>> Git is really quite impressive on all these fronts- to bad it's getting so obese in the feature implementation department (build it from source one day, you'll see what I mean :)
> > 
> > Without the svn support most of the obesity seems to be gone.
> > And building everything including clang and Firefox from source
> > helps to keeps things in perspective ?
> 
> Well, perhaps clang is large- but I don't believe it requires both python and perl to build 'feature-expect complete' :)

I guess it depends on your feature expectations.

The "stripped-down" clang in base doesn't, but the port
depends on both python and perl.

> >> Just curious- (mostly because I'm amazed to see how much more feasable using ZFS has become with smaller memory footprints).
> > 
> > After a bit of tuning it always worked for me.
> 
> That's pretty awesome- do you have, or could you recommend some notes?

On this laptop I merely set:

vm.kmem_size="3G"
vfs.zfs.write_limit_override=300000000
vfs.zfs.arc_max="1200M"
vfs.zfs.arc_meta_limit="600M"

With the exception of the last one they are all mentioned
in both the handbook and the Wiki.

I still occasionally experiment with other values and sysctls
and wouldn't recommend applying any of them without testing,
but they (mostly) work for me and the remaining issues aren't
bad enough to consider going back to UFS.

> I've tried ZFS on boxes with <4gb memory, and failed miserably at making them run reasonably...

Are you referring to reliability issues or "just" performance?

Fabian
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130313/7e6115bc/attachment.bin>

From mcevoy.pat at gmail.com  Fri Mar 15 10:39:40 2013
From: mcevoy.pat at gmail.com (Patrick McEvoy)
Date: Fri, 15 Mar 2013 10:39:40 -0400
Subject: [nycbug-talk] OT: Help question / office space
Message-ID: <514332AC.7010203@gmail.com>

Hello,
I am helping a small client through a small office move and need a 
reliable "Man with Van" guy. It is for probably one load to stuff on a 
week day and would probably be about a full day of work. The reason why 
I am posting here instead of just picking someone off Craigslist is that 
I would like to work with someone who has been vouched for. This is of 
course a normal paying gig, the only favor I ask is the vouch.
Thank you.

On a related note, if anyone needs office space around w56street BTW 
Bway & 7th ave, I know of a room for you.

P



From pete at nomadlogic.org  Fri Mar 15 14:24:04 2013
From: pete at nomadlogic.org (Pete Wright)
Date: Fri, 15 Mar 2013 11:24:04 -0700
Subject: [nycbug-talk] FreeBSD moving to py-distribute
Message-ID: <51436744.3010701@nomadlogic.org>

Don't know if anyone else caught this in their UPDATING ports file (i 
know there are a few guys who may have to deal with this on talk@):

"date 2013-03-05

affects users of devel/py-setuptools (i.e you)


devel/py-setuptools was replaced with devel/py-distribute. Please do
the following according to package manager used. py-setuptools port
will be removed shortly."


There are further instructions on migrating over to py-distribute.  This 
looks like a generally good thing IMHO - i know i've built a whole 
system around py-setuptools and felt it was not getting much love lately.

Also thought I'd share this pretty awesome tool:

http://updating.braincomb.com/

-pete

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



From george at ceetonetechnology.com  Fri Mar 15 18:16:56 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 15 Mar 2013 18:16:56 -0400
Subject: [nycbug-talk] NSLs unconstitutional
Message-ID: <51439DD8.3080108@ceetonetechnology.com>

Not OT IMHO since so many people are around the colo scene here...

http://www.wired.com/threatlevel/2013/03/nsl-found-unconstitutional

It's a huge deal for providers.

g


From nick at hackermonkey.com  Mon Mar 18 20:27:22 2013
From: nick at hackermonkey.com (Nick Danger)
Date: Mon, 18 Mar 2013 20:27:22 -0400
Subject: [nycbug-talk] Nas4Free
Message-ID: <5147B0EA.6050309@hackermonkey.com>

Is anyone running nas4free and mounting via NFS? I just set up a box and 
I am having horrible write performance over NFS. Write only, reads are 
fine. And same partition mounted via CIFS works great reads and writes 
(to the same linux host even, just mount cifs instead of nfs).   I 
haven't been able to gleen any advice via google, and most of the NFS 
tuning advice is more like "how to eek the last few bits of speed" where 
as when I say horrible, I mean instead of 45 MB/s, I am getting 0.6 MB/s 
write performance.

Any advice?

Server: Nas4Free v9.1.0.1, 4 disks in a ZFS pool (raid1Z).  Client: 
Fedora 18, mounts 2 other NFS servers (working fine) and can mount the 
NAS4Free as CIFS or NFS. CIFS works, NFS is SLOOOOOOOOOOW. :-)
More details on request.

Nick



From spork at bway.net  Mon Mar 18 22:17:56 2013
From: spork at bway.net (Charles Sprickman)
Date: Mon, 18 Mar 2013 22:17:56 -0400
Subject: [nycbug-talk] Nas4Free
In-Reply-To: <5147B0EA.6050309@hackermonkey.com>
References: <5147B0EA.6050309@hackermonkey.com>
Message-ID: <67193651-8971-4147-B847-E90A0FBC84A6@bway.net>

On Mar 18, 2013, at 8:27 PM, Nick Danger wrote:

> Is anyone running nas4free and mounting via NFS? I just set up a box and I am having horrible write performance over NFS. Write only, reads are fine. And same partition mounted via CIFS works great reads and writes (to the same linux host even, just mount cifs instead of nfs).   I haven't been able to gleen any advice via google, and most of the NFS tuning advice is more like "how to eek the last few bits of speed" where as when I say horrible, I mean instead of 45 MB/s, I am getting 0.6 MB/s write performance.
> 
> Any advice?

Not a solution, but something to test.

If you turn off sync writes on whatever fs you're exporting, does the write speed on nfs start to match smb?

ie:  zfs set sync=disabled yourpool/yourfs

Don't do that with important data, just use it for testing.

Charles

> 
> Server: Nas4Free v9.1.0.1, 4 disks in a ZFS pool (raid1Z).  Client: Fedora 18, mounts 2 other NFS servers (working fine) and can mount the NAS4Free as CIFS or NFS. CIFS works, NFS is SLOOOOOOOOOOW. :-)
> More details on request.
> 
> Nick
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk




From bonsaime at gmail.com  Mon Mar 18 23:26:33 2013
From: bonsaime at gmail.com (Jesse Callaway)
Date: Mon, 18 Mar 2013 23:26:33 -0400
Subject: [nycbug-talk] Nas4Free
In-Reply-To: <67193651-8971-4147-B847-E90A0FBC84A6@bway.net>
References: <5147B0EA.6050309@hackermonkey.com>
 <67193651-8971-4147-B847-E90A0FBC84A6@bway.net>
Message-ID: <CAJU5ZBOFrUKMB1AG1CucTvj47_kKRVecf_Mr78Vos4OQikBSFQ@mail.gmail.com>

This is some good advice. Also, before tearing your hair out at the cli on
this appliance see if there's an NFS option for going async <-- on the
client and the server

I can't tell you how long I've spent poring over numbers and doing large
writes vs many small writes and changing the buffers and playing with udp
vs tcp... meh. It can be a real pain getting the expected performance out
of NFS. Sometimes it's just not the right tool for the job, although I'll
have to say I'm not the most expert implementer.

If your tests involve contentious access then maybe see if there's any
improvement when splitting out the exports among the directories.


On Mon, Mar 18, 2013 at 10:17 PM, Charles Sprickman <spork at bway.net> wrote:

> On Mar 18, 2013, at 8:27 PM, Nick Danger wrote:
>
> > Is anyone running nas4free and mounting via NFS? I just set up a box and
> I am having horrible write performance over NFS. Write only, reads are
> fine. And same partition mounted via CIFS works great reads and writes (to
> the same linux host even, just mount cifs instead of nfs).   I haven't been
> able to gleen any advice via google, and most of the NFS tuning advice is
> more like "how to eek the last few bits of speed" where as when I say
> horrible, I mean instead of 45 MB/s, I am getting 0.6 MB/s write
> performance.
> >
> > Any advice?
>
> Not a solution, but something to test.
>
> If you turn off sync writes on whatever fs you're exporting, does the
> write speed on nfs start to match smb?
>
> ie:  zfs set sync=disabled yourpool/yourfs
>
> Don't do that with important data, just use it for testing.
>
> Charles
>
> >
> > Server: Nas4Free v9.1.0.1, 4 disks in a ZFS pool (raid1Z).  Client:
> Fedora 18, mounts 2 other NFS servers (working fine) and can mount the
> NAS4Free as CIFS or NFS. CIFS works, NFS is SLOOOOOOOOOOW. :-)
> > More details on request.
> >
> > Nick
> >
> > _______________________________________________
> > 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
>



-- 
-jesse
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130318/86e98469/attachment.htm>

From mark.saad at ymail.com  Tue Mar 19 09:52:44 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Tue, 19 Mar 2013 09:52:44 -0400
Subject: [nycbug-talk] pkgng question
Message-ID: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>

All
 I am wondering if there is a way to prevent a pkg from being updated.
For example, i want to build a port with custom options that the pkgng
repos will not support, and therefore
when I run pkg update && pkg upgrade I do not want it to upgrade my
custom package.  I cant find any way to do this. Any ideas ?



-- 

Mark Saad | mark.saad at ymail.com


From mark.saad at ymail.com  Tue Mar 19 09:54:18 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Tue, 19 Mar 2013 09:54:18 -0400
Subject: [nycbug-talk] Nas4Free
In-Reply-To: <5147B0EA.6050309@hackermonkey.com>
References: <5147B0EA.6050309@hackermonkey.com>
Message-ID: <CAMXt9Na4y=EY_7Z2wZKNNZS9aU1cQOpX21BFinGYzzR-DdNJ9Q@mail.gmail.com>

On Mon, Mar 18, 2013 at 8:27 PM, Nick Danger <nick at hackermonkey.com> wrote:
> Is anyone running nas4free and mounting via NFS? I just set up a box and I
> am having horrible write performance over NFS. Write only, reads are fine.
> And same partition mounted via CIFS works great reads and writes (to the
> same linux host even, just mount cifs instead of nfs).   I haven't been able
> to gleen any advice via google, and most of the NFS tuning advice is more
> like "how to eek the last few bits of speed" where as when I say horrible, I
> mean instead of 45 MB/s, I am getting 0.6 MB/s write performance.
>
> Any advice?
>
> Server: Nas4Free v9.1.0.1, 4 disks in a ZFS pool (raid1Z).  Client: Fedora
> 18, mounts 2 other NFS servers (working fine) and can mount the NAS4Free as
> CIFS or NFS. CIFS works, NFS is SLOOOOOOOOOOW. :-)
> More details on request.
>
> Nick
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk

Nick
  What kinds of hardware are you using ? What is the Nas box, what
type of switch , and what are the Linux boxen? Can you send the list
the details and maybe the output from Lspci -lv and  pciconf -v would
be a good start.  Also can you confirm the nas has the correct
speed/duplex settings . This sounds like a network issue .  IMHO.


-- 

Mark Saad | mark.saad at ymail.com


From gjb at FreeBSD.org  Tue Mar 19 09:59:01 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Tue, 19 Mar 2013 09:59:01 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
Message-ID: <20130319135901.GB1556@glenbarber.us>

On Tue, Mar 19, 2013 at 09:52:44AM -0400, Mark Saad wrote:
> All
>  I am wondering if there is a way to prevent a pkg from being updated.
> For example, i want to build a port with custom options that the pkgng
> repos will not support, and therefore
> when I run pkg update && pkg upgrade I do not want it to upgrade my
> custom package.  I cant find any way to do this. Any ideas ?
> 

Tricky question to answer without more info.  What is the port you
customize?

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/20130319/2ff94306/attachment.bin>

From mark.saad at ymail.com  Tue Mar 19 10:08:47 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Tue, 19 Mar 2013 10:08:47 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <20130319135901.GB1556@glenbarber.us>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
Message-ID: <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>

On Tue, Mar 19, 2013 at 9:59 AM, Glen Barber <gjb at freebsd.org> wrote:
> On Tue, Mar 19, 2013 at 09:52:44AM -0400, Mark Saad wrote:
>> All
>>  I am wondering if there is a way to prevent a pkg from being updated.
>> For example, i want to build a port with custom options that the pkgng
>> repos will not support, and therefore
>> when I run pkg update && pkg upgrade I do not want it to upgrade my
>> custom package.  I cant find any way to do this. Any ideas ?
>>
>
> Tricky question to answer without more info.  What is the port you
> customize?
>
> Glen
>

Glen in my case I use x11/slim build from ports. The current pkg in
http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/amd64/ has
some issue with intel video using KMS. Slim starts but when it trys to
switch to the wm it hangs up and restarts. I am not sure what the
issue is and I would like to rebuild and test out some things. If I
remove slim and go back to "startx" as my user there is no issue
starting X with my wm.

But as a general use case I would like to use this for  devel/boost as
we use some non-default options for boost.


-- 

Mark Saad | mark.saad at ymail.com


From gjb at FreeBSD.org  Tue Mar 19 10:28:21 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Tue, 19 Mar 2013 10:28:21 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
 <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
Message-ID: <20130319142821.GC1556@glenbarber.us>

On Tue, Mar 19, 2013 at 10:08:47AM -0400, Mark Saad wrote:
> On Tue, Mar 19, 2013 at 9:59 AM, Glen Barber <gjb at freebsd.org> wrote:
> > On Tue, Mar 19, 2013 at 09:52:44AM -0400, Mark Saad wrote:
> >> All
> >>  I am wondering if there is a way to prevent a pkg from being updated.
> >> For example, i want to build a port with custom options that the pkgng
> >> repos will not support, and therefore
> >> when I run pkg update && pkg upgrade I do not want it to upgrade my
> >> custom package.  I cant find any way to do this. Any ideas ?
> >>
> >
> > Tricky question to answer without more info.  What is the port you
> > customize?
> >
> > Glen
> >
> 
> Glen in my case I use x11/slim build from ports. The current pkg in
> http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/amd64/ has
> some issue with intel video using KMS. Slim starts but when it trys to
> switch to the wm it hangs up and restarts. I am not sure what the
> issue is and I would like to rebuild and test out some things. If I
> remove slim and go back to "startx" as my user there is no issue
> starting X with my wm.
> 

Hmm.  I use x11/slim with Intel i915kms no problem.  I'd really like to
see the "real" problem fixed in this case then.

For KMS to work, iirc, you need WITH_NEW_XORG and WITH_KMS in make.conf;
do you have these?

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/20130319/2d674b19/attachment.bin>

From mark.saad at ymail.com  Tue Mar 19 10:33:12 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Tue, 19 Mar 2013 10:33:12 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <20130319142821.GC1556@glenbarber.us>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
 <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
 <20130319142821.GC1556@glenbarber.us>
Message-ID: <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>

On Tue, Mar 19, 2013 at 10:28 AM, Glen Barber <gjb at freebsd.org> wrote:
> On Tue, Mar 19, 2013 at 10:08:47AM -0400, Mark Saad wrote:
>> On Tue, Mar 19, 2013 at 9:59 AM, Glen Barber <gjb at freebsd.org> wrote:
>> > On Tue, Mar 19, 2013 at 09:52:44AM -0400, Mark Saad wrote:
>> >> All
>> >>  I am wondering if there is a way to prevent a pkg from being updated.
>> >> For example, i want to build a port with custom options that the pkgng
>> >> repos will not support, and therefore
>> >> when I run pkg update && pkg upgrade I do not want it to upgrade my
>> >> custom package.  I cant find any way to do this. Any ideas ?
>> >>
>> >
>> > Tricky question to answer without more info.  What is the port you
>> > customize?
>> >
>> > Glen
>> >
>>
>> Glen in my case I use x11/slim build from ports. The current pkg in
>> http://mirrors.nycbug.org/pub/PC-BSD/packages/9.1-RELEASE/amd64/ has
>> some issue with intel video using KMS. Slim starts but when it trys to
>> switch to the wm it hangs up and restarts. I am not sure what the
>> issue is and I would like to rebuild and test out some things. If I
>> remove slim and go back to "startx" as my user there is no issue
>> starting X with my wm.
>>
>
> Hmm.  I use x11/slim with Intel i915kms no problem.  I'd really like to
> see the "real" problem fixed in this case then.
>

I agree but its still a good question, can a package be blacklisted or
held at a particular version . What about the case of an in house
package
like marks_custom_thing.tbz . Say I have a newer version in the repo
but I do not want to upgrade to it on some servers.


> For KMS to work, iirc, you need WITH_NEW_XORG and WITH_KMS in make.conf;
> do you have these?
>
> Glen
>

Glen , I hate to say it but I am not using ports to build this stuff.
I migrated to using all binary packages on one box. I am just pulling
down what PC-BSD has up on their pkgng repo. I would guess they are
using the with kms knobs to build their pkgs, since I am loading all
of the kms modules.



-- 

Mark Saad | mark.saad at ymail.com


From nick at hackermonkey.com  Tue Mar 19 10:53:04 2013
From: nick at hackermonkey.com (Nick Danger)
Date: Tue, 19 Mar 2013 10:53:04 -0400
Subject: [nycbug-talk] Nas4Free
In-Reply-To: <CAMXt9Na4y=EY_7Z2wZKNNZS9aU1cQOpX21BFinGYzzR-DdNJ9Q@mail.gmail.com>
References: <5147B0EA.6050309@hackermonkey.com>
 <CAMXt9Na4y=EY_7Z2wZKNNZS9aU1cQOpX21BFinGYzzR-DdNJ9Q@mail.gmail.com>
Message-ID: <51487BD0.5070403@hackermonkey.com>

On 03/19/2013 09:54 AM, Mark Saad wrote:
>
> Nick
>    What kinds of hardware are you using ? What is the Nas box, what
> type of switch , and what are the Linux boxen? Can you send the list
> the details and maybe the output from Lspci -lv and  pciconf -v would
> be a good start.  Also can you confirm the nas has the correct
> speed/duplex settings . This sounds like a network issue .  IMHO.
>
>

I am not near the hardware at the moment, so some of this is from 
memory. The NAS box is running NAS4Free v9.1.0.1. The hardware is a 
MiniITX board (dual core Atom) 4GB memory, Intel board if I remember 
correctly. Drives are Seagate 500GBs. I can't send the outputs as I 
can't get into the box right now :-)   The switch is a ProCurve SOHO, 12 
ports. Linux box is a Dell Optiplex i3 something or other, Fedora 18. I 
think it was a Broadcom chip. Again, I'll get those stats tonight when I 
get back home. I sort of doubt its a network/hardware issue as I can get 
the throughput on the same hardware just mounting via CIFS/SMB rather 
than NFS. I won't rule it out as I really have no good ideas.

Some results of messing around last night you can see the NFS throughput 
is horrible on writes, but good on reads. CIFS is good on both.

http://pastebin.com/u4V7TGSB

I'll check the async status also when I get home tonight. And run some 
more tests.

Nick



From nick at hackermonkey.com  Tue Mar 19 18:45:47 2013
From: nick at hackermonkey.com (Nick Danger)
Date: Tue, 19 Mar 2013 18:45:47 -0400
Subject: [nycbug-talk] Nas4Free
In-Reply-To: <51487BD0.5070403@hackermonkey.com>
References: <5147B0EA.6050309@hackermonkey.com>
 <CAMXt9Na4y=EY_7Z2wZKNNZS9aU1cQOpX21BFinGYzzR-DdNJ9Q@mail.gmail.com>
 <51487BD0.5070403@hackermonkey.com>
Message-ID: <5148EA9B.7000403@hackermonkey.com>

On 03/19/2013 10:53 AM, Nick Danger wrote:
> On 03/19/2013 09:54 AM, Mark Saad wrote:
>>
>> Nick
>>    What kinds of hardware are you using ? What is the Nas box, what
>> type of switch , and what are the Linux boxen? Can you send the list
>> the details and maybe the output from Lspci -lv and  pciconf -v would
>> be a good start.  Also can you confirm the nas has the correct
>> speed/duplex settings . This sounds like a network issue . IMHO.
>>
>>

Ok, what I am finding online is that loads of people are having similar 
issues. NFS and FreeBSD and ZFS are not happy all together. Which seems 
odd.


http://christopher-technicalmusings.blogspot.com/2010/09/zfs-and-nfs-performance-with-zil.html


Most of the complaints seem to center around ZFS/NFS/Esxi or KVM, but no 
matter the last bit, the first two are the same. And it shouldn't matter 
of its ESX, KVM or just mounting some disks for space, it should work. I 
have not figured out how to change to nfs v4 and turn off async mostly 
because nas4free overwrites things you change in rc.conf when you 
reboot. How nice of it. I suppose thats what I get for installing a 
"take care of everything and give me a pretty gui" software, and then 
expecting it to let me much around in the bowels.  I'll poke around on 
the forums for nas4free tonight, but if I don't find a decent answer, I 
might just blow the thing away and redo it without ZFS and cheating 
'NAS' software distros :-) Or I could always switch to iSCSI since its 
only a single client mounting that data.

Thanks,

Nick




From gjb at FreeBSD.org  Tue Mar 19 20:39:10 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Tue, 19 Mar 2013 20:39:10 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
 <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
 <20130319142821.GC1556@glenbarber.us>
 <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>
Message-ID: <20130320003910.GB1566@glenbarber.us>

On Tue, Mar 19, 2013 at 10:33:12AM -0400, Mark Saad wrote:
> > Hmm.  I use x11/slim with Intel i915kms no problem.  I'd really like to
> > see the "real" problem fixed in this case then.
> >
> 
> I agree but its still a good question, can a package be blacklisted or
> held at a particular version . What about the case of an in house
> package
> like marks_custom_thing.tbz . Say I have a newer version in the repo
> but I do not want to upgrade to it on some servers.

In such a case, you would have a port that pkg(8) is not aware of.

What I do for my machines is create a misc/$(hostname -s) port, which
I then use as a meta-port for software I want installed on a particular
machine.

> > For KMS to work, iirc, you need WITH_NEW_XORG and WITH_KMS in make.conf;
> > do you have these?
> >
> 
> Glen , I hate to say it but I am not using ports to build this stuff.
> I migrated to using all binary packages on one box. I am just pulling
> down what PC-BSD has up on their pkgng repo. I would guess they are
> using the with kms knobs to build their pkgs, since I am loading all
> of the kms modules.
> 

No, because WITH_NEW_XORG is not default, and as far as I recall,
neither is WITH_KMS.

So, here's the problem with your x11/slim situation - an unfortunate
fact is that binary packages are not "one size fits all."  An excellent
example is lang/php5, where if you want the apache module included, you
need to build the port.

This is the primary reason I roll my own packages, using
ports-mgmt/tinderbox.

This way, I can be certain all of the options I want are enabled, and
options I do not want or do not need are disabled.

Going back to your original question, my 'misc/$(hostname -s)' solution
was going to be my suggestion.  But, since boost-libs is a dependency of
other ports (not a dependent port), this is much more difficult.

To worsen your boost-libs situation, you will eventually run into
a situation where a 'pkg upgrade' will blow up because your
locally-patched version is not up-to-date.  I'm not saying this _can_
happen - I am saying this _will_ happen.  Unfortunately, I do not think
my solution will work for you.

It is never advisable to mix ports and packages.  This was the case with
pkg_install(1), and remains true with pkg(8).

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/20130319/3ca1256b/attachment.bin>

From lists at eitanadler.com  Tue Mar 19 21:25:46 2013
From: lists at eitanadler.com (Eitan Adler)
Date: Tue, 19 Mar 2013 21:25:46 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
 <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
 <20130319142821.GC1556@glenbarber.us>
 <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>
Message-ID: <CAF6rxgn+4KZY9JTJFvmNO0FQQdw2COjCarz6irkAsB9x6WibuA@mail.gmail.com>

On 19 March 2013 10:33, Mark Saad <mark.saad at ymail.com> wrote:
> On Tue, Mar 19, 2013 at 10:28 AM, Glen Barber <gjb at freebsd.org> wrote:
>> On Tue, Mar 19, 2013 at 10:08:47AM -0400, Mark Saad wrote:
>>> On Tue, Mar 19, 2013 at 9:59 AM, Glen Barber <gjb at freebsd.org> wrote:
>>> > On Tue, Mar 19, 2013 at 09:52:44AM -0400, Mark Saad wrote:
>>> >> All
>>> >>  I am wondering if there is a way to prevent a pkg from being updated.
>>> >> For example, i want to build a port with custom options that the pkgng
>>> >> repos will not support, and therefore
>>> >> when I run pkg update && pkg upgrade I do not want it to upgrade my
>>> >> custom package.  I cant find any way to do this. Any ideas ?

It is not possible to do this.  There is an open bug report for this.
It is planned in the future.

:(

-- 
Eitan Adler


From mark.saad at ymail.com  Wed Mar 20 10:18:51 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Wed, 20 Mar 2013 10:18:51 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <20130320003910.GB1566@glenbarber.us>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
 <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
 <20130319142821.GC1556@glenbarber.us>
 <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>
 <20130320003910.GB1566@glenbarber.us>
Message-ID: <CAMXt9NbTiWhVo9ZsNgUH-tm+D8km+j6a_hadg258NBnXcVOOmQ@mail.gmail.com>

On Tue, Mar 19, 2013 at 8:39 PM, Glen Barber <gjb at freebsd.org> wrote:
> On Tue, Mar 19, 2013 at 10:33:12AM -0400, Mark Saad wrote:
>> > Hmm.  I use x11/slim with Intel i915kms no problem.  I'd really like to
>> > see the "real" problem fixed in this case then.
>> >
>>
>> I agree but its still a good question, can a package be blacklisted or
>> held at a particular version . What about the case of an in house
>> package
>> like marks_custom_thing.tbz . Say I have a newer version in the repo
>> but I do not want to upgrade to it on some servers.
>
> In such a case, you would have a port that pkg(8) is not aware of.
>
> What I do for my machines is create a misc/$(hostname -s) port, which
> I then use as a meta-port for software I want installed on a particular
> machine.

Glen I worked at a shop that used this method ; are there any
guide/handbook/wiki pages describing how to make your own ports ?




>
>> > For KMS to work, iirc, you need WITH_NEW_XORG and WITH_KMS in make.conf;
>> > do you have these?
>> >
>>
>> Glen , I hate to say it but I am not using ports to build this stuff.
>> I migrated to using all binary packages on one box. I am just pulling
>> down what PC-BSD has up on their pkgng repo. I would guess they are
>> using the with kms knobs to build their pkgs, since I am loading all
>> of the kms modules.
>>
>
> No, because WITH_NEW_XORG is not default, and as far as I recall,
> neither is WITH_KMS.
>
> So, here's the problem with your x11/slim situation - an unfortunate
> fact is that binary packages are not "one size fits all."  An excellent
> example is lang/php5, where if you want the apache module included, you
> need to build the port.
>
> This is the primary reason I roll my own packages, using
> ports-mgmt/tinderbox.
>
> This way, I can be certain all of the options I want are enabled, and
> options I do not want or do not need are disabled.
>
> Going back to your original question, my 'misc/$(hostname -s)' solution
> was going to be my suggestion.  But, since boost-libs is a dependency of
> other ports (not a dependent port), this is much more difficult.
>
> To worsen your boost-libs situation, you will eventually run into
> a situation where a 'pkg upgrade' will blow up because your
> locally-patched version is not up-to-date.  I'm not saying this _can_
> happen - I am saying this _will_ happen.  Unfortunately, I do not think
> my solution will work for you.
>
> It is never advisable to mix ports and packages.  This was the case with
> pkg_install(1), and remains true with pkg(8).
>
> Glen

>

What about using two repos . Repo 1 is the public FreeBSD repo , Repo
2 is a private repo with local versions of the packages we nee
d customized . Yum currently has support for this , as far as I can
tell Pkgng doesn't


-- 

Mark Saad | mark.saad at ymail.com


From gjb at FreeBSD.org  Wed Mar 20 10:34:19 2013
From: gjb at FreeBSD.org (Glen Barber)
Date: Wed, 20 Mar 2013 10:34:19 -0400
Subject: [nycbug-talk] pkgng question
In-Reply-To: <CAMXt9NbTiWhVo9ZsNgUH-tm+D8km+j6a_hadg258NBnXcVOOmQ@mail.gmail.com>
References: <CAMXt9NbNW3FOnKFsqb09VRX0JfByYaM8h_F32HBj=rypwE1pqg@mail.gmail.com>
 <20130319135901.GB1556@glenbarber.us>
 <CAMXt9NaztfaXqEQ2QLpqvWRwKUpsLm6qCqAcKu-GF0qF_3zmvQ@mail.gmail.com>
 <20130319142821.GC1556@glenbarber.us>
 <CAMXt9NaG6=LjuNbnwPz=6ceyoiCbB1ZTYiOxbUU=nKSfoJR8=w@mail.gmail.com>
 <20130320003910.GB1566@glenbarber.us>
 <CAMXt9NbTiWhVo9ZsNgUH-tm+D8km+j6a_hadg258NBnXcVOOmQ@mail.gmail.com>
Message-ID: <20130320143419.GC1629@glenbarber.us>

On Wed, Mar 20, 2013 at 10:18:51AM -0400, Mark Saad wrote:
> > What I do for my machines is create a misc/$(hostname -s) port, which
> > I then use as a meta-port for software I want installed on a particular
> > machine.
> 
> Glen I worked at a shop that used this method ; are there any
> guide/handbook/wiki pages describing how to make your own ports ?
> 

I have a sample called "laptop-port" here:

    http://people.freebsd.org/~gjb/laptop-port/

> What about using two repos . Repo 1 is the public FreeBSD repo , Repo
> 2 is a private repo with local versions of the packages we nee
> d customized . Yum currently has support for this , as far as I can
> tell Pkgng doesn't

This is from the pkg.conf(5) manual:

     PKG_MULTIREPOS: boolean	This option when enabled will tell
				pkg(1) to work in multiple repositories mode.
				For example repositories definitions, please
				have a look at the sample configuration file.

I do not know how well it works.

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/20130320/12daab03/attachment.bin>

From edlinuxguru at gmail.com  Thu Mar 21 09:53:55 2013
From: edlinuxguru at gmail.com (Edward Capriolo)
Date: Thu, 21 Mar 2013 09:53:55 -0400
Subject: [nycbug-talk] Wonker Parody World of System Automation
Message-ID: <CAENxBwxe6SMVE8S3ZCeJrYW_91F6KzGpfjk8nFPZ2XG3rabUBA@mail.gmail.com>

My buddy Nathan opened up a talk with this.
http://www.youtube.com/watch?v=LThNLVkTUE0

It will be stuck in your head for weeks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20130321/fad938fd/attachment.htm>

From george at ceetonetechnology.com  Sat Mar 23 13:14:06 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sat, 23 Mar 2013 13:14:06 -0400
Subject: [nycbug-talk] armed.nycbug.org update
Message-ID: <514DE2DE.6030209@ceetonetechnology.com>

This box was setup to create FreeBSD images for BeagleBones and
Raspberry Pis to make life easy for those without build machines.

The box has been updated to 9.1.  It ends up the issues we were having
with Gonzo's script was due to FBSD 8.x's make(1) not supporting
MAKESYSPATH.

So we should have workable images up and regular builds in the near future.

g


From matt at atopia.net  Sat Mar 23 19:07:50 2013
From: matt at atopia.net (Matt Juszczak)
Date: Sat, 23 Mar 2013 23:07:50 +0000 (UTC)
Subject: [nycbug-talk] Modern Mailing List as a Service?  Or software?
Message-ID: <alpine.BSF.2.00.1303232302100.16643@endeavour.bitlancer.net>

Hi folks,

We have a client using Monarch for mailing list archiving with ezmlm for 
the actual mailing lists, hosted internally in their own data center. 
It's part of a bigger puzzle, but the goal is to migrate most of the SMTP 
stuff off of some BSD relay servers to something like Sendgrid.  I'm 
trying to see if there's an "as a service" solution for hosting these 
lists, considering there's only a few of them and I don't think they need 
to maintain hardware (or even cloud infrastructure) for them.

Alternatively, majordomo or mailman may still be viable alternatives, 
especially with good BSD support for both.  Does anyone have any 
suggestions?  Definitely want to move away from ezmlm as it hasn't been 
supported in over a decade.

Thanks,

Matt


From bcallah at devio.us  Sat Mar 23 20:26:09 2013
From: bcallah at devio.us (Brian Callahan)
Date: Sat, 23 Mar 2013 20:26:09 -0400
Subject: [nycbug-talk] Modern Mailing List as a Service?  Or software?
In-Reply-To: <alpine.BSF.2.00.1303232302100.16643@endeavour.bitlancer.net>
References: <alpine.BSF.2.00.1303232302100.16643@endeavour.bitlancer.net>
Message-ID: <514E4821.3060603@devio.us>

On 3/23/2013 7:07 PM, Matt Juszczak wrote:
> Hi folks,
>
> We have a client using Monarch for mailing list archiving with ezmlm for
> the actual mailing lists, hosted internally in their own data center.
> It's part of a bigger puzzle, but the goal is to migrate most of the
> SMTP stuff off of some BSD relay servers to something like Sendgrid.
> I'm trying to see if there's an "as a service" solution for hosting
> these lists, considering there's only a few of them and I don't think
> they need to maintain hardware (or even cloud infrastructure) for them.
>
> Alternatively, majordomo or mailman may still be viable alternatives,
> especially with good BSD support for both.  Does anyone have any
> suggestions?  Definitely want to move away from ezmlm as it hasn't been
> supported in over a decade.
>
> Thanks,
>
> Matt

Off the top of my head, you could try mlmmj. It's a modern codebase (the 
current release was made mid-2012) and IIRC, is supposed to emulate (or 
at least is inspired by) ezmlm.

Ports might lag behind one version (but should be trivial to bring 
up-to-date).

NOTE: I haven't tried it myself, just going off memory.

~Brian


From george at ceetonetechnology.com  Sat Mar 23 21:22:57 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sat, 23 Mar 2013 21:22:57 -0400
Subject: [nycbug-talk] Modern Mailing List as a Service?  Or software?
In-Reply-To: <514E4821.3060603@devio.us>
References: <alpine.BSF.2.00.1303232302100.16643@endeavour.bitlancer.net>
 <514E4821.3060603@devio.us>
Message-ID: <514E5571.90609@ceetonetechnology.com>

On 03/23/13 20:26, Brian Callahan wrote:
> On 3/23/2013 7:07 PM, Matt Juszczak wrote:
>> Hi folks,
>>
>> We have a client using Monarch for mailing list archiving with ezmlm for
>> the actual mailing lists, hosted internally in their own data center.
>> It's part of a bigger puzzle, but the goal is to migrate most of the
>> SMTP stuff off of some BSD relay servers to something like Sendgrid.
>> I'm trying to see if there's an "as a service" solution for hosting
>> these lists, considering there's only a few of them and I don't think
>> they need to maintain hardware (or even cloud infrastructure) for them.
>>
>> Alternatively, majordomo or mailman may still be viable alternatives,
>> especially with good BSD support for both.  Does anyone have any
>> suggestions?  Definitely want to move away from ezmlm as it hasn't been
>> supported in over a decade.
>>
>> Thanks,
>>
>> Matt
> 
> Off the top of my head, you could try mlmmj. It's a modern codebase (the
> current release was made mid-2012) and IIRC, is supposed to emulate (or
> at least is inspired by) ezmlm.
> 
> Ports might lag behind one version (but should be trivial to bring
> up-to-date).
> 
> NOTE: I haven't tried it myself, just going off memory.
> 

+1 on mlmmj.  Well done mail manager.  NYC*BUG would be using it if the
migration from mailman wasn't such a nightmare.

I've managed some lists with it before, and I enjoy the Unixey approach.
 Haven't used any of the add-ons, but there's a lot of stuff out there
AFAIK.

g



From george at ceetonetechnology.com  Sun Mar 24 20:12:32 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sun, 24 Mar 2013 20:12:32 -0400
Subject: [nycbug-talk] SparkleShare
Message-ID: <514F9670.2010902@ceetonetechnology.com>

Has anyone run it on FreeBSD?

There's instructions for a variety of Linux distros, and the one missing
eyesore in configuring on FreeBSD is the Mono dependencies.

I could wade through that, but wondering if anyone has already.

Also, I assume one could just run git instead of the full-blown
application, to access a SparkleShare host... anyone?

g


From george at ceetonetechnology.com  Sun Mar 24 21:14:33 2013
From: george at ceetonetechnology.com (George Rosamond)
Date: Sun, 24 Mar 2013 21:14:33 -0400
Subject: [nycbug-talk] armed.nycbug.org update
In-Reply-To: <514F9663.8050203@gmx.com>
References: <mailman.3.1364140803.5131.talk@lists.nycbug.org>
 <514F9663.8050203@gmx.com>
Message-ID: <514FA4F9.7050202@ceetonetechnology.com>

On 03/24/13 20:12, Brett Mahar wrote:
>> Subject: [nycbug-talk] armed.nycbug.org update
>> Message-ID: <514DE2DE.6030209 at ceetonetechnology.com>
>> Content-Type: text/plain; charset=ISO-8859-1
>>
>> This box was setup to create FreeBSD images for BeagleBones and
>> Raspberry Pis to make life easy for those without build machines.
>>
>> The box has been updated to 9.1.  It ends up the issues we were having
>> with Gonzo's script was due to FBSD 8.x's make(1) not supporting
>> MAKESYSPATH.
>>
>> So we should have workable images up and regular builds in the near
> future.
>>
>> g
> 
> Hi,
> 
> I wonder if it would be possible for this machine to make images for the
> beagleboard-xm as well?

Well, there is a Summer of Code port if you check out wiki.freebsd.org,
but neither of us have that board.  There is code in svn available.

Look at Tim's scripts... it wouldn't be too much hassle to add the other
boards his scripts already support (VersatilePB, PandaBoard), but it's
not on our immediate agenda to do the Board-xm if we can't test it.

If you figure out something to easily build and test with the svn code,
hit me offline and we can set something up.

g



From brett.mahar at gmx.com  Sun Mar 24 20:12:19 2013
From: brett.mahar at gmx.com (Brett Mahar)
Date: Mon, 25 Mar 2013 11:12:19 +1100
Subject: [nycbug-talk] armed.nycbug.org update
In-Reply-To: <mailman.3.1364140803.5131.talk@lists.nycbug.org>
References: <mailman.3.1364140803.5131.talk@lists.nycbug.org>
Message-ID: <514F9663.8050203@gmx.com>

 > Subject: [nycbug-talk] armed.nycbug.org update
 > Message-ID: <514DE2DE.6030209 at ceetonetechnology.com>
 > Content-Type: text/plain; charset=ISO-8859-1
 >
 > This box was setup to create FreeBSD images for BeagleBones and
 > Raspberry Pis to make life easy for those without build machines.
 >
 > The box has been updated to 9.1.  It ends up the issues we were having
 > with Gonzo's script was due to FBSD 8.x's make(1) not supporting
 > MAKESYSPATH.
 >
 > So we should have workable images up and regular builds in the near 
future.
 >
 > g

Hi,

I wonder if it would be possible for this machine to make images for the 
beagleboard-xm as well?

Brett.


From okan at demirmen.com  Mon Mar 25 11:17:47 2013
From: okan at demirmen.com (Okan Demirmen)
Date: Mon, 25 Mar 2013 11:17:47 -0400
Subject: [nycbug-talk] Modern Mailing List as a Service?  Or software?
In-Reply-To: <514E5571.90609@ceetonetechnology.com>
References: <alpine.BSF.2.00.1303232302100.16643@endeavour.bitlancer.net>
 <514E4821.3060603@devio.us> <514E5571.90609@ceetonetechnology.com>
Message-ID: <20130325151747.GB20856@clam.khaoz.org>

On Sat 2013.03.23 at 21:22 -0400, George Rosamond wrote:
> On 03/23/13 20:26, Brian Callahan wrote:
> > On 3/23/2013 7:07 PM, Matt Juszczak wrote:
> >> Hi folks,
> >>
> >> We have a client using Monarch for mailing list archiving with ezmlm for
> >> the actual mailing lists, hosted internally in their own data center.
> >> It's part of a bigger puzzle, but the goal is to migrate most of the
> >> SMTP stuff off of some BSD relay servers to something like Sendgrid.
> >> I'm trying to see if there's an "as a service" solution for hosting
> >> these lists, considering there's only a few of them and I don't think
> >> they need to maintain hardware (or even cloud infrastructure) for them.
> >>
> >> Alternatively, majordomo or mailman may still be viable alternatives,
> >> especially with good BSD support for both.  Does anyone have any
> >> suggestions?  Definitely want to move away from ezmlm as it hasn't been
> >> supported in over a decade.
> >>
> >> Thanks,
> >>
> >> Matt
> > 
> > Off the top of my head, you could try mlmmj. It's a modern codebase (the
> > current release was made mid-2012) and IIRC, is supposed to emulate (or
> > at least is inspired by) ezmlm.
> > 
> > Ports might lag behind one version (but should be trivial to bring
> > up-to-date).
> > 
> > NOTE: I haven't tried it myself, just going off memory.
> > 
> 
> +1 on mlmmj.  Well done mail manager.  NYC*BUG would be using it if the
> migration from mailman wasn't such a nightmare.

++

> I've managed some lists with it before, and I enjoy the Unixey approach.
>  Haven't used any of the add-ons, but there's a lot of stuff out there
> AFAIK.
> 
> g
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk


From mark.saad at ymail.com  Mon Mar 25 12:29:42 2013
From: mark.saad at ymail.com (Mark Saad)
Date: Mon, 25 Mar 2013 12:29:42 -0400
Subject: [nycbug-talk] WPS ( Wireless Protected Setup) and FreeBSD
Message-ID: <CAMXt9Nb9tMJTY1jBHFZoSoCjCsv9rVRc9Tt9ZP3d66azdyG3aQ@mail.gmail.com>

Hello Talk
 I am thinking of adding some WiFi Range extenders / repeaters to my
home network.  Most of the the cost effective models you can find on
amazon , tigerdirect etc use WPS to get their configuration. Has
anyone tried to setup a WPS device using FreeBSD either via PFSense,
M0n0Wall
or a stock setup ?  Are there any pitfalls or pointers ?



-- 

Mark Saad | mark.saad at ymail.com