Re: can't prevent root lockout under Tru64/C2 security
From: Nico Kadel-Garcia (nkadel_at_comcast.net)
Date: 01/31/05
- Previous message: Darren Tucker: "Re: can't prevent root lockout under Tru64/C2 security"
- In reply to: Darren Tucker: "Re: can't prevent root lockout under Tru64/C2 security"
- Next in thread: Darren Tucker: "Re: can't prevent root lockout under Tru64/C2 security"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Sun, 30 Jan 2005 22:58:23 -0500
"Darren Tucker" <dtucker@gate.dodgy.net.au> wrote in message
news:slrncvrace.ib9.dtucker@gate.dodgy.net.au...
> Warning: topic drift ahead. For a potential solution see my other
> post up-thread.
>
> On 2005-01-30, Nico Kadel-Garcia <nkadel@comcast.net> wrote:
>> Number 1: Tru64 support is basically dead as a doornail.
>
> HP may be fitting it for a coffin but it's not dead yet.
>
>> I strongly urge you
>> to take the money for electricity used powering those older systems and
>> invest it in a modest modern Opteron box running a decent Linux distro.
>
> That's not always easy, cost effective or even possible.
I've seen that analysis tried. I've also run some numbers several times in
the past year, comparing the costs of a Tru64 maintenance and the hardware
maintenance costs, cooling requirements, software issues with the
limitations of the older and no longer developed software vs. re-engineering
any core applications from scratch or based on newer core tools.
The new tools won hands down, but you did have to amortize over at least one
year to see the savings. By now, since ASU is dead, Matlab is no longer
supported on Tru64, and it's difficult if not impossible to bring gcc over
to Tru64 to compile other useful tools anymore rather than the amazingly
optimized-to-the-point-of-incompatible compiler, it's an even shorter
period. Call it six months in a typical multi-user environment, especially
if you need large storage arrays. (Ever tried using one of these with large
IDE or SATA arrays instead of the vastly more expensive SCSI? Don't.)
> In many environments, the new hardware cost is pocket change compared to
> the the cost of time, effort, application upgrades, approvals, testing,
> (re)certifications and contract renegotiations required to implement it.
That's quite true. But the cost of maintenance, ye ghods, man!
> An application may not even be available on a newer platform at all.
> Contractual obligations, regulations and/or certifications may prevent.
Yup. Contractual obligations and paperwork can trump any wise technical
idea.
>> There just aren't enough of these systems left alive to constitute an
>> open
>> source community to keep them going.
>
> Digression: as far as I'm concerned, OpenSSH support is alive on a given
> platform as long as *someone* is prepared to do the work, and it can be
> done without compromising the integrity of it.
>
> For example, I still test releases on AIX 4.2.1 even though that version
> has been EOLed for nearly half a decade. It's similar enough to modern
> versions that the effort involved is small (and I get a kick out of the
> occasional email I get telling me that it works on those systems or ones
> even older).
Cool. I've found OSF releases to be amazingly painful to deal with, in
particular getting popular open source system tools (such as gcc and ddd)
ported over to it.
> In the case of sshd's SIA support, it's pretty well written and the
> author usually pops up when a pre-release call-for-testing goes out.
> Even if he stops doing that, as long as someone steps up it'll still be
> "supported". And if no one does and you still need support, drop me a
> line, we might be able to work something out :-)
But getting gcc over to it to compile OpenSSH is non-trivial, and the
built-in compiler is just not up to the job, at least the last time I tried
last year. And that underlying support community is evaporating if they're
not already gone.
- Previous message: Darren Tucker: "Re: can't prevent root lockout under Tru64/C2 security"
- In reply to: Darren Tucker: "Re: can't prevent root lockout under Tru64/C2 security"
- Next in thread: Darren Tucker: "Re: can't prevent root lockout under Tru64/C2 security"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
|
