Re: Fw: IBM i Disk system statistics

I'm not certain you'll find SSD replacements for those drives, I'm not sure
they existed back then.

Larry's comment about the age of the drives is the real point. IBM i
Storage Management is constantly looking at the performance of the drive,
spin rate, head movements, etc. to see if the drive meets performance
specifications. When it does not, the drive is marked bad.

I've got a couple of customers in your same situation, and I'm replacing
drives monthly. Soon they will run out of stock and we will dive into
Larry's stash. The trick is to keep up with it to avoid losing a RAID
set. Sounds like your doing that now but stopping/starting RAID sounds a
bit too much like work.

To avoid unpleasant results, I suggest you remove the "replacement drives"
from the system, and stop RAID on them. Now all you have to do is pull one
of them and replace the dead one. Far less work, same result.
--
Jim Oberholtzer
Chief Technical Architect
Agile Technology Architects


On Wed, Nov 28, 2018 at 9:39 AM Armand Borick <armand@xxxxxxxxxxxxxxxxxx>
wrote:

I never received an "impending failure" message fron any of the dead
drives.
They were all parity errors , displayed by a "critical system error"
message
and an entry in the
"work with problems" screen.

I have 10 parity sets running on the machine. When one parity set faults,
I
have been copying the data,
ending parity on the affected set, removing the drive , and restarting the
parity set.
It's been working out OK so far.

I plan to do a survey of manufacture dates on the installed drives. Maybe
that will show something.

I am well aware of the vintage of the hardware I have. The 810 is still
on
V5R4!

Were I to replace drives, I would be more inclined to go for compatable
SSDs.
That is a fantasy though. These guys don't want to spend money right now.

I have to put a business case together to justify some consultant time to
help me with the HMC issue I mentioned earlier.
That is just how it is here right now.

Armand

----- Original Message -----
From: "DrFranken" <midrange@xxxxxxxxxxxx>
To: "Midrange Systems Technical Discussion" <midrange-l@xxxxxxxxxxxx>
Sent: Tuesday, November 27, 2018 23:59
Subject: Re: IBM i Disk system statistics

'Not Replacing' them is, um, 'fraught with peril' because when the

second

drive does in a Parity set, you're toast.

You do realize that a 525 is OLD and I did some math a while back that

the

drives in that thing have likely rotated over 1 BILLION times so things

in

there are getting tired. We ARE seeing those drives beginning to fail at
higher rates simply due to age.

IBM i constantly is looking at disks and deciding if the number of

errors

is such that a drive will fail. If the numbers get to that point you

will

get 'impending DASD failure' messages in QSYSOPR meaning replacing the
drive is a good idea.

I would suspect that those folks claiming that other servers are

'better'

are not comparing to decade old equipment.

Also in case you're confused on dates, the i810 is even older than a

525!

Those came out in 2003!

<vendor response>

If you are looking for replacement drives for the 525 I have MANY of

them.

35, 70, and 140G and they are CHEAP.

</Vendor response>

- Larry "DrFranken" Bolhuis

www.Frankeni.com
www.iDevCloud.com - Personal Development IBM i timeshare service.
www.iInTheCloud.com - Commercial IBM i Cloud Hosting.

On 11/27/2018 4:59 PM, Armand Borick wrote:

Hi Everyone!

I am the inheritor of a 9406-525 that we would like to use as a backup

to

our production 9406-810.

I have the system set up and running, but the disk drives keep throwing
parity errors.

The new system has about 3TB of disk in the main LPAR, so I have been
just pulling the bad drives out of the
array without replacing them.

My question is: Is there a way to display the error statistics for the
drives in the array, so I can identify upcoming failures?

I ran the Surface Analysis about 2 weeks ago and it reported as OK.
This morning another unit reported a parity error.
That is the 4th to go bad since I brought up the system about 2 months
ago.
(the array started out with 75 disks)

The system is just waiting right now, while I finish off some other
projects.
The plan is to clone the applications and data, and use Journaling and
remote data queues to
keep the data files up to date.

I am not inclined to load up all the files just to have the system

start

dropping drives.
I would like to identify the worst of the remaining drives, and remove
them.
My production box only has 180GB on it, so there is plenty of room.

I am trying to pitch the i as a good platform for high availabliity,

but

its a tough sell with drives dropping dead left and right.

I have a bunch of Milleneals in management here who think Windows

servers

are the best thing, and IBM i is "legacy".

Any suggestions?

Thanks!
Armand Borick

--
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: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com

--
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: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com