Re: Lost contact with partition after applying latest CUME and groups

[Pete Helgren <pete@xxxxxxxxxx>]

Okee Dokee...

Further investigation reveals that I had a drive fault back in July but 
it is in the storage module.  If I remember correctly, and I can't say I 
do, the JS12 had a mirrored set of drives in the blade itself and then I 
think I mirrored a set of drives in the storage module for the OS so I 
am hoping that replacing the drive in the storage module will get me 
close to what I need.  I just can't remember how things are configured 
and NOW the IVM isn't reporting in (but I have seen that happen 
before).  The 7.1 partition and Linux are still running so I am hopeful 
a drive replacement will get things going.  I haven't replaced a 
mirrored drive before although I have replaced a single drive in a RAID 
5 set.  More adventures await.

I would have thought that a mirrored set would have started up but been 
in a degraded state. Right now I am trying to figure out what drives I 
have in the Storage Module so I can order a few...thought I had some 
spares but can't seem to locate them....

www.petesworkshop.com
GIAC Secure Software Programmer-Java
Twitter - Sys_i_Geek  IBM_i_Geek

On 1/6/2018 2:12 PM, DrFranken wrote:
> Loss of disks for sure. The A6xx was probably incrementing through the 
> number of disks you had.
>
> In any case that's also what the 4158 says only now it's complaining 
> that it cannot find the load source.
>
> Personally I think all your stuff is there, albeit a tad upset with 
> you for not doing a PWRDWNSYS neatly. It'll survive.
>
> You DO need to check carefully the connections from the i Partitioni 
> to the VIO and assure all the mappings and such are correct. Your 
> problem lies within.
>
>         - Larry "DrFranken" Bolhuis
>
> www.Frankeni.com
> www.iDevCloud.com - Personal Development IBM i timeshare service.
> www.iInTheCloud.com - Commercial IBM i Cloud Hosting.
>
> On 1/6/2018 12:49 PM, Pete Helgren wrote:
> > My usual seat of the pants update procedure probably coming back to 
> > haunt me.....
> >
> > Ordered the latest CUME and Groups for my 7.2 partition, built the 
> > image catalog and applied it with the option to re-IPL.  As usual I 
> > didn't bother to read any MTU's (I can hear Rob Berendt groaning...) 
> > and the apply and re-IPL went smoothly.  All was well yesterday 
> > afternoon, the only thing I did was to start SSHD and update my 
> > LetsEncrypt certificate and change supported ciphers due to the 
> > return of the Bleichenbacher's Oracle Threat (ROBOT).  No problem.  
> > All was well.
> >
> > Checked on a couple of things this morning and the 7.2 partition was 
> > no longer online.  This is a JS12 blade with a Power 6.  The IVM is 
> > up, the 7.1 partition is up, the Linux partition is up but the IBM i 
> > console wouldn't connect and there was an A6XX0255 error on the 
> > partition in the IVM.  So, I did what any intrepid developer would 
> > do:  I bounced the partition.
> >
> > The error I see now is:
> >
> >
> > Reference code:    B2004158 LP=00002
> > Description:    System log entry only, no service action required
> > Time stamp (UTC):    1/6/18 4:21:59 PM
> > Word 2:    03B00000
> > Word 3:    000000FF
> > Word 4:    00000000
> > Word 5:    30000002
> > Word 6:    000300FF
> > Word 7:    00000001
> > Word 8:    00007F00
> > Word 9:    00000000
> > FRU callout:
> >
> > A couple of posts and references point to a load source issue. It 
> > isn't hardware (I don't think) since all the other operational 
> > partitions use the same hardware (disks, etc) that have been 
> > virtualized for the partitions. So I am guessing that there was 
> > something in an MTU that I failed to read that is causing the issue. 
> > With no console or virtual control panel so getting this guy back on 
> > his legs will be a  little tricky.  The good news is that the IVM 
> > allows me to change IPL modes and load source adapters so hopefully 
> > there will be a way to boot this guy....
> >
> > Ideas?  The "good" news is that this is a developer machine so if I 
> > have to restore, I will (painful as that may be...)