[Brahms-dev-l] Re: [Brahms-soft-l] life after redhat

Dear Hiro

Thanks for the update -
It is not so obvious what we can do, but certainly SUSE is not an option
since it was just purchased by Novell. The real issue is of course the
updates; e.g. the RCF internal farm machines can due to the isolation on the
firewalls care less for at least extened periods on what
security will be done.
I learne today the the req from Phenix in term of Obj-y is realy only due to
legacy Database stuff, the newer runs are base on Ingres, so why should the
rest of us care?

cheers
    Flemming




----- Original Message ----- 
From: "Hironori Ito" <hito@rcf.rhic.bnl.gov>
To: <brahms-soft-l@lists.bnl.gov>
Sent: Friday, November 07, 2003 6:24 PM
Subject: [Brahms-soft-l] life after redhat


> Hello.  As some of you might know, RedHat is changing their business
> course (to make money!!!)  (I suppose that there is nothing wrong with
> that.) But, it is (or will be) making a very BIG mess not only in BNL
> but also in entire DOE (and probably entire physics or scientific
> community).  The reason for this mess is that RedHat will stop
> supporting many of their products (up to RH9) at the end of April 2004.
> There will be no RH after 9, and they will force their customers to a
> subscription based service with quite high price tag.  (The cheapest
> price is  about $180 per year per computer!!!)  The problem we (or at
> least people in DOE labs) have is that we are obligated to update our
> security patches as they arrive.  For RedHat machine in BNL, this was
> accomplished by RedHat network, which is a paid service (small price)
> with very easy web interface.  (And, we have never paid for  use of
> RedHat linux OS.)  But, with the subscription based system, we have to
> pay to use it as well as its support for patches. As a result, there was
> a meeting in BNL physics department (not RCF) (there will be more
> meeting for physics as well as entire BNL).  So, here is the first
> report from that meeting.
>
>
> 1.  None of us is interested in paying $180 (at least) per year per
> machines to use linux.  (I thought that this is the reason physics
> community moved away from VMS.)
> 2.  Negotiation between RedHat and DOE (not just BNL) is under way.  The
> price might get lower for DOE lab.  (not necessary for other people.
> This might be a problem if people want to use exactly the same system at
> your own home/institution.)  (Rumor:  RedHat wants to have $1.3M from
> DOE for assumed 25k linux machines($52 per pc). )
> 3.  Not many people are eager to change from RedHat to other linux.
> 4.  What are other alternative?  There are many other linux out there
> without subscription fees:  Debian, Suse, Mandrake, Fedora, etc...   (I
> can hear very loud cheer for certain one from the other side of Atlantic
> ocean :) ).   Here is the brief description for some of them I know.
>     a.  Debian:  Pros---truly community linux (no business association).
> Linux hard core favorite.   Cons--- support for third party software
> (objectivity and oracle???  It does not matter to us, but it matters to
> other collaboration in RCF.)
>     b.  Suse or Mandrake ---  business association (like old RedHat).
> Suse has large support from corporation.
>     c.  Fedora --- community supported RedHat.  This is a sort of like
> Debian.  (Debian wanabee???)   Basically the same as the current and/or
> future RedHat.  Cons--- Since it has just started, its future is very
> unknown.
>     d.  Fermi or CERN linux ---  Here, it gets interesting.  Some big
> Labs are thinking (or already making) own version of linux. If you are
> interested in Fermi linux, go to
> http://www-oss.fnal.gov/projects/fermilinux/  .  If you are interested
> in CERN linux, go to http://linux.web.cern.ch/linux/  . (Note:  Both of
> them are based on RedHat.   It is unclear now what they will do in the
> future.)
>
>
> Anyway,  I will report as I hear more information (or rumors :) ). If
> you have a particular favorite, you should make sure your words are
> heard by the people in the higher-up of a food chain at every big/small
> lab. (well particularly at a big lab).  Your opinion might actually
> count.  (Ok, That is not entirely true.  I don't think RCF is switching
> to Mac OSX no matter what you say in the near future.  Sorry, Apple
> fans!!!)
>
>
> Hiro
>
>
> _______________________________________________
> Brahms-soft-l mailing list
> Brahms-soft-l@lists.bnl.gov
> http://lists.bnl.gov/mailman/listinfo/brahms-soft-l
>


_______________________________________________
Brahms-dev-l mailing list
Brahms-dev-l@lists.bnl.gov
http://lists.bnl.gov/mailman/listinfo/brahms-dev-l