[blag-whereto] BLAG 90K, 100K and the resources that are available

[john maclean]

2009/1/6 Abdur-Rahman Morgan <21stcenturytechnosapien at gmail.com>:
> HAPPY 2009!
>
> I hope everyone is well and in good spirits. I am writing you all today
>  in hopes that we all have been progressing our knowledge in how we can
> contribute to the distribution in order to maintain BLAG and GNU Linux.
> This e-mail contains mostly questions, ideas and thoughts that I have
> had over the past month as I have not been as active as I had hoped, but
> after reading a good majority of the forums, all of the blag-whereto
> list archive(s) and documents posted in crabgrass and others on the
> site, I have the time, energy and resources where I can.
>

Watch this space. So far I'm grinding my teeth, wondering wether to
change anaconda back to the "pre fedora 9" way. Also a diskboot.img in
the final iso image would be nice. The Fedora team don't do that any
more :-(

Been praticing with CentOS, that stable low-hanging fruit. FC10-jx is
a goer but does not run on most machines. Loads of notes, coffee and
swearing!


> Yesterday, I posted a conversation to Crabgrass pertaining to a
> conversation between renilgh,Jaromil and lxo on indymedia. At the end of
> the conversation, what I concluded was that our resources need to be
> defined in order to understand not only how we will be developing, but
> distributing BLAG both online and offline.


At the risk of being an *hole, I think that we agree on a collection
of tools that we can all use.Example;

- Editor 1,2,3
- Development tools, 1,2,3
- revision control, 1,2,3 # fuck that, just stick with git!
- scripting language(s) 1,2,3

This way we can have "teams" that can stick to the tool(s) that they
like or are good with, rather than an individual having to learn
-everything-. I'm sure that this will work itself out in time.

 In sum, if we were at the
> point of rolling out updates for 90k or rolling out 100k and actively
> maintaing it:
>
> 0.How much bandwidth would we need in order to do this successfully?

I remember Jebba saying that there was about 50,000 downloads of 70k.
I've told as many people to grab as much as possible from "kimsufi" as
possible. It's not mirrored as far as I know. What would be /really/
nice is if kimsufi went doen at 02:37 and another server came up soon
after. Even if it was a day later.

> 1.Where and who would be hosting the space?

I can get servers with reasonable specs and jolly good throughput.
Even have a choice of UK data centers to put them in. Me and my big
mouth told one bunch about BLAG and they got nervous. So I'm looking
around for bandwith/throughput deals.

 > 2.What equipment would be serving those resources primarily and what
> resources what would be our back up?


Dell 750 and Dell 850 servers for now. That's what I can do any way.

> 3.Where would major development be taking place in orer to work effectively

We make a "blueprint" or methodology. I.e, "we aim to use this box
like this...." I think that every box  that we put up should be RAIDED
and mirrored elsewhere. The physical locations of the boxes is
irrelevant if the connection is good.

[We have a gigabit link in the office but I'd be washing dishes for
years if i made that public]

>
>
> These are a few of the first questions that I have at the present
> because I do not want to see the resources we have disappear slowly.
>
> As, I had been writing this e-mail in my mind. I thought we might leave
> kimsufi up and contribute towards the cost of to ensure that it has the
> resources we will be working on.

Leave that to me for now. There /will/ be some maintenance -soon- and
it may be down for a week or so. Watch this space.

 Did jebba pay for all this himself
> while he was developing BLAG?

Man, makes my heart weep how much he put towards it. That's another story.
>
> Jayeola, how much would it cost and what type of resources would be
> adequate so that we can use it in the next few months to roll any
> updates for 90k and use for our next version. C

an we just contribute to
> you directly for the cost?

Uh, gimme your knowledge! That is what I want ;-) And feed back when i
get my act together.

 What is our alternative? What is our back-up
> source? I would prefer that we continuing hammering the bandwidth in
> order to make the progress that we need, rather than limit/stunt/delay
> our accomplishments as we've come far and still have a great deal do.
>
> ...as I had been writing this e-mail, noldrin, provided me the the link
> to http://wiki.blagblagblag.org/Servers, a nice blueprint that clarifies
>  questions 1 and 2...so do we wish to follow the same model and setup
> severs similarly or how can we prove upon it, to take into account our
> present needs.
>
> What alternatives are possible?

Loads of choices. None of them are silly. The only truley silly option
is to do nothing!

I will add some more commentary in the
> next day or so, but I await your replies.
>
> Abdur-Rahman Morgan
>
> Resilence | Longevity | Ingenuity
>
> _______________________________________________
> blag-whereto mailing list
> blag-whereto at lists.aktivix.org
> https://lists.aktivix.org/mailman/listinfo/blag-whereto
>

Hope this email made sense. Throw some shoes at me if it didn't.

-- 
John Maclean
07739 171 531
MSc (DIC)

Timezone: GMT