Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: KVM question - skip this technology entirely



 I worked with Sun systems years ago.  Yes, you could shut them off at 
the power supply, but the 'inteligent' power supply kept a small 
processor (think 'bios') running.  From the serial port or network it 
was possible to login and 'power up' the system and see the entire boot 
process on the serial port.  It was possible to power down similarly. 
The ability to run FORTH in the Bios was a hoot and allowed for all 
kinds of things, but few used this power. 

On Mon, 2008-05-12 at 09:15 -0400, Matt Shields wrote: 
> On Mon, May 12, 2008 at 7:44 AM, Jerry Feldman <[hidden email]> wrote: 
> > On Sun, 11 May 2008 22:42:52 -0400 (EDT) 
> > 
> > "Rich Braun" <[hidden email]> wrote: 
> > 
> > 
> > > Hmm I don't understand this argument.  If main power to the computer is off 
> >  > then you can't turn it on no matter what kind of options you have.  In my case 
> >  > I'm working with systems that each have two power supplies so hopefully the 
> >  > above scenarios would be virtually impossible (at least, no more likely than a 
> >  > scenario where a DRAC could somehow put power into the system by some 
> >  > alternate method).  The KVMs don't have any mechanism for power cycling or 
> >  > resetting so that's what interests me about the DRAC/ILO solution. 
> >  > 
> >  > Didn't know you could connect a separate power supply to the DRAC or ILO, I 
> >  > really am just starting out with this stuff, just wanted to post here that 
> >  > this is an alternative to the KVMs (which have *always* frustrated me).  Maybe 
> >  > the console front-ends will ultimately prove just as frustrating but so far I 
> >  > haven't had any troubles.  And the "user experience" is as good as or better 
> >  > than the Raritan or Avocent connection, at least through a 100-Mbit or faster 
> >  > network link. 
> > 
> >  As mentioned, on an HP ILO, when you power off the computer (eg. 
> >  shutdown -h or via the power button) the ILO board remains powered on. 
> >  Certainly, if you pull the plug (or plugs) that will certainly power 
> >  down the ILO. 
> 
> The ILO's have a DC input for an extra power brick.  So even if you 
> the main power cables were pulled, it's still possible to get into the 
> ILO, which would have an alert that says it's main power source has 
> been lost.  The ILO's are nice, but I found they cost more to use 
> them.  I won't run lower end switches just to maintain a KVM network, 
> when I used them in the past I put them on my regular enterprise class 
> switches on a different VLAN.  Once you figured in the cost of the ILO 
> license, the extra network ports, the extra power ports for the brick, 
> it did cost more than a KVM solution.  But whatever, to each his own. 
> 
> >  Some KVM's do have power options. In our case, we do have physical 
> >  access 9-5 and usually do not have any mission critical applications 
> >  requiring access outside of normal business hours, but we do have 
> >  employees who do log in nights and weekends. 
> > 
> >  So, my objective is to be able to manage the systems remotely more as a 
> >  convenience. I'll take a look at the DRAC solution since the Intel 
> >  whiteboxen are Supermicro. 
> > 
> >  Note that while at HP, we had a very large lab in the Linux Expertise 
> >  Center (about 500 square feet) with equipment ranging from smaller 
> >  ProLiant systems up to large SuperDome with mostly Linux, but also 
> >  Windows Server, HP-UX, and different virtualization strategies.  We 
> >  did use KVMs so we could have console access from our desks although 
> >  the lab was accessible to us 24x7. I think we might have used Brocade 
> >  at the time. 
> > 
> > 
> > 
> >  -- 
> >  -- 
> >  Jerry Feldman <[hidden email]> 
> >  Boston Linux and Unix 
> >  PGP key id: 537C5846 
> >  PGP Key fingerprint: 3D1B 8377 A3C0 A5F2 ECBB  CA3B 4607 4319 537C 5846 
> > 
> > _______________________________________________ 
> >  Discuss mailing list 
> >  [hidden email] 
> >  http://lists.blu.org/mailman/listinfo/discuss
> > 
> > 
> 
> 
> 
> -- 
> -matt 
> 


BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org