[Linux-cluster] Solution: Install over FC SAN works, boot doesn't

Rick Stevens rstevens at internap.com
Tue Mar 20 18:21:30 UTC 2007


On Tue, 2007-03-20 at 12:59 -0600, isplist at logicore.net wrote:
> After endless trial and errors, the solution was so much simpler than the 
> problem. 
> The problem was that I was able to install over the SAN to a volume but I 
> could not boot from that installation.
> It was as simple as going into the 'Advanced setting' of the boot installer 
> while installing the server and picking the proper drive to boot.
> 
> Wish RH would make it a pick list rather than having to highlight the drive, 
> then click it up one at a time but hey, it works. 
> 
> The answer was from a way cool guy at Qlogic who spend two days looking into 
> this and figured out the problem. Now I have my central storage AND SAN based 
> roots for each blade, no need for drives, flash drives, nothing, just the 
> storage.

Glad you got it sorted out.  So long as the drives don't move around,
you should be good to go.

> On Tue, 20 Mar 2007 17:44:35 +0000, William Rizzo wrote:
> > Hi you're trying to do that?
> > 
> > When you say "Install Linux over the FC Network", do you mean Boot From 
> SAN?
> > 
> > I guess that if can see the LUN at any stage, the LUN Masking and the
> > Zoning aren't the problems.
> > 
> > 
> > William
> > 
> > 
> > 2007/3/19, isplist at logicore.net <isplist at logicore.net>: > I am able to
> > install Linux over the FC network and onto a storage volume.
> >> However, when I try to boot these installs, I only get a blank screen and
> >> cannot boot.
> >> 
> >> I have found that I can install the /boot directory to a flash card on
> >> each
> >> blade and then am able to boot fine with swap and root on the remote
> >> volume.
> >> 
> >> Why is this and how can I make my install 100% remote?
> >> 
> >> Also strange (to me at least), I have found that if I connect the multiple
> >> volume storage device to the rest of the FC network, already installed
> >> machines with Qlogic cards in them no longer boot. They come up with many
> >> errors and all finally stall at the syslog starting point. Disconnecting
> >> the
> >> new storage allows them to boot again.
> >> 
> >> I believe this has something to do with LUN masking/zoning (which I've
> >> yet to
> >> learn, have to start reading Brocade 2800 docs next) but what I don't get
> >> is
> >> why the new blades can see their own LUNS if installed as noted above with
> >> flash /boot, but old machines get stuck?
> >> 
> >> Rather confusing.
> >> 
> >> Mike
> >> 
> >> 
> >> _______________________________________________
> >> Anaconda-devel-list mailing list
> >> Anaconda-devel-list at redhat.com
> >> https://www.redhat.com/mailman/listinfo/anaconda-devel-list
> 
> 
> 
> 
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
> 
----------------------------------------------------------------------
- Rick Stevens, Principal Engineer             rstevens at internap.com -
- VitalStream, Inc.                       http://www.vitalstream.com -
-                                                                    -
-         It is better to have loved and lost.  Cheaper, too!        -
----------------------------------------------------------------------




More information about the Linux-cluster mailing list