Re: Quick and dirty system recovery: Is this method OKAY ?
- From: Robert Heller <heller@xxxxxxxxxxxx>
- Date: Fri, 10 Feb 2006 21:08:54 +0100
"G Dahler" <gd-nntp3@xxxxxxxxxx>,
In a message on Fri, 10 Feb 2006 12:08:15 -0500, wrote :
"D> Hi,
"D>
"D> I have 2 sun V20z running RHEL3. One of them is a production box. The other
"D> is a development box and could be shut down for a couple of hours.
"D> Unfortunately, those server do not have dual redundant power supplies. They
"D> do have hardware mirrored disks. I have a same day contract support on both
"D> servers from sun.
"D>
"D> I would like to do the following in case my production server motherboard or
"D> power supply crashes:
"D>
"D> Shut down the dev server, take both the production disks and put them in the
"D> dev box (in the same spot), boot the dev box (which is now supposedly the
"D> production box) and wait for the sun technician to replace the PS or
"D> motherboard before reversing the procedure.
"D>
"D> Do you think there might be a problem with it ? Apart from the MAC adress
"D> change for the NIC card, I don't see a problem (and I will remove the HWADDR
"D> line from my eth0-conf file)
So far everything is golden. Not really much different from how I
installed WBL 3.0 on my Laptop (which lacks a CD-ROM and whose PCMCIA
NIC is in the 'unsupported' list) -- I put the laptop's HD in my desktop
and installed using the desktop's mobo, etc. and then swapped the hd
back in the laptop.
"D>
"D> The only other problem I see would be with the hardware RAID.
Depends on how the hardware RAID controller stores its configuration.
If it stores it on the *disks*, things should be transparent. If it is
storing things in its NVRAM, then it depends on exactly what it is
storing in its NVRAM. If it is storing disk serial numbers, the dev
box will be unhappy, claiming that both disks died and/or were replaced
and it might not trust the config (if any) on the disks themselves. If
the RAID controller has the option of backing its config up to a
floppy, then you should make this floppy backup (you should anyway, in
case the RAID controller's battery goes bad or something). Then when
the dev box boots with the production system's disks, 'restore' the
config from the floppy. You should also have a backup of the dev
machine's RAID config as well (so you can restore that later).
"D>
"D>
"D>
"D>
\/
Robert Heller ||InterNet: heller@xxxxxxxxxxxx
http://www.deepsoft.com/ ||FidoNet: 1:321/153
http://www.deepsoft.com/~heller /\
.
- References:
- Quick and dirty system recovery: Is this method OKAY ?
- From: G Dahler
- Quick and dirty system recovery: Is this method OKAY ?
- Prev by Date: Re: Question about nvidia video card
- Next by Date: Newb Bash script question...
- Previous by thread: Quick and dirty system recovery: Is this method OKAY ?
- Next by thread: Question about nvidia video card
- Index(es):
Relevant Pages
|