Okay, I have tried this.

I booted of hdisk1, and while updating the ML, I get this problem now


installp: APPLYING software for:
devices.chrp.base.rte 5.2.0.85

Replaying log for /dev/fwdump.
Failure replaying log: -1
mount: /dev/fwdump on /var/adm/ras/platform: Unformatted or
incompatible media
The superblock on /dev/fwdump is dirty. Run a full fsck to fix.
update: Failed while executing the devices.chrp.base.rte.post_u
script.


installp: The installation has FAILED for the "root" part
of the following filesets:
devices.chrp.base.rte 5.2.0.85

installp: Cleaning up software for:
devices.chrp.base.rte 5.2.0.85

Was I supposed to do anything special with my dump devices?

Any ideas?



Lv wrote:
> ML Update with other disk
>
> --------------------------------------------------------------------------------
>
> Hi All,
>
> I am planning to do ML update (5.2 ML 05 to TL 08).
>
> I was planning something like this:
> 1) unmirror rootvg (hdisk0, hdisk1)
> 2) perform ML update on hdisk1
> 3) bootoff hdisk1
>
> Does that sound right? If so, I am not sure how to do step #2. For step
> #3, I can just change the bootlist order, correct?
>
> Any thoughts about this?
>
>
> TIA
>
>
>
>
> Mag
>
> #2 (permalink) 10-15-2006, 10:16 PM
> Friedhelm Neyer Posts: n/a
>
> Re: ML Update with other disk
>
> --------------------------------------------------------------------------------
>
> ....
> 1. unmirrorvg rootvg
> 2. reducevg rootvg hdisk1
> 3. alt_disk_install -C hdisk1
>
> With the lspv command you see at this point a rootvg and an alt_rootvg
>
> Now you can boot from hdisk1 and perform your update to TL08
> After updating you should run the systemwith the new TL and see if all
> apps run properly - if so do the following
>
> At this point you have a rootvg and an old_rootvg (yes OLD...), so you
> do the following:
>
> 1. alt_disk_install -X old_rootvg
> 2. extendvg -d rootvg hdisk1
> 3. mirrorvg -S rootvg
>
> Now you have a mirrored rootvg again with TL08