RE: I/O Battery cache replacement problems

["Thomas Garvey" <tgarvey@xxxxxxxxxx>]

I suppose it is possible that the FSP is OS managed because I did not get
this while thing running back at install, and the only time we've ever had
to boot both the system and the HMC is when we had a power blackout a year
ago.

So, I think I should do the following, based on suggestions here:
Find out what level we are at with the FSP.
If really old (like 225) which I'm guessing we are, look into going to an
intermediate level
Check into whether FSP is OS managed and, if so, change to HMC managed.
Upgrade the FSP to some level before 240 (which is the latest?)
Try to boot the iSeries again. 


Best Regards,

Thomas Garvey
Unbeaten Path International
Office: 630-462-3991
web:   www.unpath.com



-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Pete Massiello
Sent: Thursday, February 12, 2009 12:22 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: I/O Battery cache replacement problems

Since you are SO FAR behind in your FSP code (from your first email about
not updating anything like i5/OS, PTFs, FSP, and HMC) you will have a
problem going from a lower level directly up to the 240 code stream.  You
need to find out what level you are at, and if you are at something really
old like 225, then you should upgrade to an intermediate level before going
to 240.  Be careful, because if your FSP is OS Managed (which is the wrong
setting with an HMC), and you are back on the 225 code level, you should NOT
put the latest cumulative PTFs on until you do the intermediate FSP update. 

Pete

Pete Massiello
iTech Solutions
http://www.itechsol.com

 
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx
Sent: Thursday, February 12, 2009 12:51 PM
To: Midrange Systems Technical Discussion
Subject: RE: I/O Battery cache replacement problems

Tangent 1:  If your HMC is defined as your system console then there's a way
to get the system console screen on that and you wouldn't have needed to go
to some PC with iSeries access loaded on it.  I could go on about the beauty
of this but you have other priorities.

Sounds like you were using your HMC recently so, unless your firmware got
upgraded because this is your first full system shutdown in years then that
shouldn't be the issue.  We get this up and running remind me how to show
you how to check your firmware levels.  Also remind me how to show you how
to check who does firmware upgrades:  hmc or cumes.  If you're using a hmc
ALWAYS set it to hmc.

Here's a V5R4 thing they buried in a V5R4 cume that mentions D200B077:
MH01117  NONE  09/30/08 HOT:UPT:HE:LPM: DSI in pagerEndCopy       MH01136  
 
                         during J2 stress te
                         Migration fails to complete if processors  
                          were removed prior
                         D200B077 - Partition Shutdown Hangs  
                         P6 L2 Cache UE's not writing an entry  
                         into the line del
                         iFST:  B60069D9 error logs surfaced by  
                         HEA on RCHASTMN
                         Tower resources either Unknown or  
                         Occupied status after
                         GA5:IH:FST:manhunter: SRC:A181D000 is  
                         recorded when I i
                         HEA:  Missing Interrupt Handler  
                         Incorrectly Handling EQ
                         GA5: FST:  HMC went to incomplete while  
                         FW update.
                         P6_Field:  /tmp contents not gathered  
                         with FSPDUMP
                         MPV: IH 154: sunfire: Processor APR did  
                         not recover cor
                         MPV: Harbug (TOD_OSC0_OUT_R to GND) on  
                         HV8 machine syst
                         GA5:FST:CLD:bentleyfp1: PE B181EF88 in  
                         Slow IPL
                         HV8-ODC:Shimlafp1:System in "Incomplete"  
                         state when add
                         GA3SP2: B181E630 reported while GA1SP1->  
                         GA3SP2
                         IH: Balm:  Rings missing for p2 c0 in SH  
                         dump
                         FST:GA5:IH:Joker:Lost IO on IPL and  
                         B7006981 and B70069
                         Phyp usable lmb count greater than target  
                          after node ad
                         System Level Activate Does Not Complete  
                         PRD simulator changes  
                         P6_Field:  eMLx system crashed during  
                         runtime & posted SRC B
                         Fix Pack 01EL320_083_045  
                         HDW CIRC - Apply "cortina" circumvention  
                         to all P5IOC2/
                         INCORROUT IB B170E560 during Memory  
                         Region allocation
                         INCORROUT IB B700F103 During HCA Free  
                         Resource
                         CCM-FFY: system checkstops post repair  
                         due to processor
                         IPL: Slave FSP didn't get psi link  
                         assigned
                         latest js22 scripts no longer work  
                         through UX/MCP insta
                         P6FIELD_FHA:  Bloomberg system crashed  
                         and should have 
                         Implement changes to allow second non-mem  
                          presv re-ipl
                         Hardware Circumvention:  Galaxy1 RC QP  
                         Checkstop Workaround
                         Unsupported svpdMFGtool option used by  
                         mnfgclean -acvpd
                         pPET: HMC lost access to LPARs - Vio  
                         locking issue
                         HEA Lport Config Result in SRC B700F103  
                         P6FIELD: IPL-SRCB6000103-UNPRED  
                         HYPERVISOR HEAP CORRUPTION D

You may have gotten burned by the "if it ain't broke don't fix it
mentality".  Apparently IBM issued a fix for this issue but you don't have
it.  And now you've activated the situation that duplicates the problem. I
believe that MH* ptfs are firmware markers.

Here's some more hits:
http://www14.software.ibm.com/webapp/set2/firmware/gjsn?mode=30&page=01SF240
_299_201.html
http://download.boulder.ibm.com/ibmdl/pub/software/server/firmware/01SF240_3
58_201.html
http://download.boulder.ibm.com/ibmdl/pub/software/server/firmware/01EL320_0
83_045.html

You really need to find out why the HMC will not talk to the i.


Rob Berendt
--
Group Dekko Services, LLC
Dept 01.073
Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





From:
"Thomas Garvey" <tgarvey@xxxxxxxxxx>
To:
"'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx>
Date:
02/12/2009 12:14 PM
Subject:
RE: I/O Battery cache replacement problems
Sent by:
midrange-l-bounces@xxxxxxxxxxxx



All was well before I shut my server down using the HMC.  We have a single
server (model 52) configured with two partitions.   I was going to use the
HMC to do a PWRDWNSYS on each partition, but I got a warning dialog box 
(on
the HMC) saying that I should do the PWRDWNSYS from the partition itself.
So I went to a PC with an iSeries Access session on each partition and
executed the PWRDWNSYS command with the parameter to NOT restart, for each
partition.  When the HMC showed that both partitions were shut down, I 
used
the HMC to actually turn the iSeries power off completely.  When the 
iSeries
power was off completely, I unplugged the AC power cord from the back of 
the
server.  Then I pulled the card with the battery cache and replaced the
battery.  The HMC was still running.  I then powered up the iSeries with 
the
white power button and my problems began.

As for upgrading firmware...  I am only now becoming aware of this 
firmware
distinction.  My experience has always been only with the SLIC and OS
upgrades, PTF's etc.  I know we never did anything specifically to upgrade
either firmware or the HMC software.  As for the SLIC and OS updates and
PTFs, we are one of those who have NOT been scrupulous about keeping up.
The only PTF that was installed in the last 2 to 3 years was the one
required to use the ANZOBJCVN command to help decide what we might need to
upgrade to v6 of the OS.  So I don't know how our system could be at a
higher level than our HMC. 

We are currently at v5r3 of the OS and decided to upgrade to v5r4.  Our 
plan
was to later upgrade our 2nd partition to v5r4, then later on upgrade our
primary partition to v6r1 (keeping our secondary at v5r4 for a while).

At least we are still talking software here.  My fear is that we have a
hardware issue for which we don't have support.

I greatly appreciate your responses and look forward to any advice you may
have.

Best Regards,

Thomas Garvey
Unbeaten Path International
Office: 630-462-3991
web:   www.unpath.com



-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of rob@xxxxxxxxx
Sent: Thursday, February 12, 2009 10:27 AM
To: Midrange Systems Technical Discussion
Subject: RE: I/O Battery cache replacement problems

When was the last time you were able to see this server on the HMC?  The
concern I am trying to address is:
Have you upgraded your firmware beyond the support capability of your HMC? 

 Classic situation is that company has been running fat and happy for 
years
just doing 'traditional' system maintenance, cumes and what not and 
haven't
laid a hand on their HMC in years.  They set their system i to get 
firmware
updates from the cume.  Now that they are in a situation where they have 
to
use the HMC but they cannot because their system i is at a firmware level
beyond the capability of that HMC.
https://www14.software.ibm.com/webapp/set2/sas/f/power5cm/power5.html

Did you record the IP address of your FSP somewhere so that you can step
through the HMC configuration and check it?
Is the connection between your HMC and your FSP on your lan or on some
dedicated lan?

Rob Berendt
--
Group Dekko Services, LLC
Dept 01.073
Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





From:
"Thomas Garvey" <tgarvey@xxxxxxxxxx>
To:
"'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx>
Date:
02/12/2009 09:42 AM
Subject:
RE: I/O Battery cache replacement problems
Sent by:
midrange-l-bounces@xxxxxxxxxxxx



Yes, we have an HMC but it's apparently not communicating with the 
iSeries.
When I click on Servers and Partitions it shows an IP address we don't use
(10.0.0.254) and says 'No Connection'.  Normally it would show '9406-520
xxxxxxxx' where xxxxxxx is the Serial Number.  Under that it would show 
the
partitions.  None of that shows up.

We don't have a maintenance contract with IBM.  Apparently the system 
called
IBM and they called us to see if we wanted their help.  We said sure, but
then they wanted to charge $4250 to our credit card before we could even 
ask
a question.


Best Regards,

Thomas Garvey
UPI



-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Paul Nelson
Sent: Thursday, February 12, 2009 6:29 AM
To: 'Midrange Systems Technical Discussion'
Subject: RE: I/O Battery cache replacement problems

Google shows the machine is waiting for reset slots, but I have no idea 
what
that means. Do you have an HMC?

Can you call IBM?

Paul Nelson
Office 512-392-2577
Cell 708-670-6978
nelsonp@xxxxxxxxxxxxx


-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Thomas Garvey
Sent: Wednesday, February 11, 2009 7:00 PM
To: 'Midrange Systems Technical Discussion'
Subject: I/O Battery cache replacement problems

After powering my system down completely and replacing the cache battery I
am unable to get the system to IPL.  I keep getting stuck at D200B077. 
I've
tried reseating all the cards in their slots but nothing seems to help.
Any ideas?  Anyone else every try this? 
 

Best Regards,

Thomas Garvey

UPI

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing 
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe, 
unsubscribe,
or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take a
moment to review the archives at http://archive.midrange.com/midrange-l.