Hi everybody. I'm ideally using the closed thread thread52-1099809 in this area to put my question.
I successfully installed a dual Linux-AIX system on a p650, with everything upgraded - thanks to the quoted thread - to latest microcode, so I shouldn't have any kernel-limit problem.
This is the snapshot of the WORKING system:
- hdisk0 (U0.1-P2/Z1-A8) Debian 'etch' v4.0 ppc64 Linux
- hdisk1 (U0.1-P2/Z1-A9) rootvg AIX v5.3(ml04)
(AIX is only used for maintenance purposes...)
Until now, everything's fine. If I would start AIX, my SMS boot sequence: ( or, bootlist -m normal -o )
1 hdisk1 blv=hd5
2 hdisk0
... everything's OK and Aix runs fine.
If I would start Linux, changing the normal boot sequence by bootlist, diag or SMS too, in:
1 hdisk0
2 hdisk1 blv=hd5
... AIX is started, not linux.
If hdisk1 is not available (for example, removed, or not in list), I got the well known !20EE000B! error.
On the other hand, if I enter SMS and manually choose hdisk0 (A8), linux runs like a charm. But manually, because even it's the first disk in the boot-chain, it isn't booted.
I found this strange workaround.
With this bootlist:
1 hdisk0
2 cd0
3 rmt0
4 hdisk1 blv=hd5
without entering SMS and manually booting from hdisk0, if there's a bootable cdrom in the device, hdisk0 is SOMETIMES correctly started...
Just like the boot sequence would double check itself when 'broken' by a bootable device between the disks.
Any help or hint how to solve this weird behavior ?
Thanks anybody.
Valerio
I successfully installed a dual Linux-AIX system on a p650, with everything upgraded - thanks to the quoted thread - to latest microcode, so I shouldn't have any kernel-limit problem.
This is the snapshot of the WORKING system:
- hdisk0 (U0.1-P2/Z1-A8) Debian 'etch' v4.0 ppc64 Linux
- hdisk1 (U0.1-P2/Z1-A9) rootvg AIX v5.3(ml04)
(AIX is only used for maintenance purposes...)
Until now, everything's fine. If I would start AIX, my SMS boot sequence: ( or, bootlist -m normal -o )
1 hdisk1 blv=hd5
2 hdisk0
... everything's OK and Aix runs fine.
If I would start Linux, changing the normal boot sequence by bootlist, diag or SMS too, in:
1 hdisk0
2 hdisk1 blv=hd5
... AIX is started, not linux.
If hdisk1 is not available (for example, removed, or not in list), I got the well known !20EE000B! error.
On the other hand, if I enter SMS and manually choose hdisk0 (A8), linux runs like a charm. But manually, because even it's the first disk in the boot-chain, it isn't booted.
I found this strange workaround.
With this bootlist:
1 hdisk0
2 cd0
3 rmt0
4 hdisk1 blv=hd5
without entering SMS and manually booting from hdisk0, if there's a bootable cdrom in the device, hdisk0 is SOMETIMES correctly started...
Just like the boot sequence would double check itself when 'broken' by a bootable device between the disks.
Any help or hint how to solve this weird behavior ?
Thanks anybody.
Valerio