RE: WrkSysSts -- MIDRANGE-L

There is something wrong here.

Drive 1 is full size.  Drives 2-5 have reduced sizes like they are in a RAID
set.  If you create a RAID set from 5 drives, the parity data is only
carried on four of the drives - drives 5, 6, and 7 will show as full sized.
If you create a RAID set with 8 drives, they all show size reduction but
drives above 8 show as full sized.  So, except for the fact that they are
all 100 full, they look properly balanced.

Drives 6-9 look like they form another RAID set except the first drive shows
100 full and the other 3 are 40 percent.  That is wrong.  You cannot divide
a RAID set between two ASP, can you?

Since the drives are different sizes, the two types cannot be in the same
RAID set.

Check this out!  Here are the read requests and write requests per second
for the 17 GB drives:

 R   W
4.1 2.1
1.2 3.4
1.9 2.3
1.5 2.6

The standard assumption is that the drives support 80 percent reads and 20
percent writes.  There are more like one third reads and two thirds writes.
The rates for the 17GB drives is more like 50:50 and that is weird too.  It
looks like you are restoring a large quantity of stuff to these drives - how
may database files?  Are the audit journals turned on?  I'm just fishing for
SOMETHING that is creating on those low end drives because those look like
journal receivers to me.

The fastest way to get the next clue is to run SMTRACE to figure out what is
going onto the  drives - in other words, what is being read and written.  I
want to know what is on drives 1 through 5 but that is a little tricky to
figure out.

There is something wrong here.  Did you call IBM?  What OS release are you
on?

Richard Jackson
mailto:richardjackson@richardjackson.net
http://www.richardjacksonltd.com
Voice: 1 (303) 808-8058
Fax:   1 (303) 663-4325

-----Original Message-----
From: owner-midrange-l@midrange.com
[mailto:owner-midrange-l@midrange.com]On Behalf Of
Contractor1@Parkdalemills.com
Sent: Thursday, September 07, 2000 2:03 PM
To: MIDRANGE-L@midrange.com
Subject: Re: WrkSysSts



Take a look at this:

                             Work with Disk Status
 Elapsed time:   00:06:02

                 Size    %    I/O  Request  Read  Write  Read   Write    %
 Unit  Type      (M)   Used   Rqs  Size (K) Rqs   Rqs    (K)     (K)   Busy
    1    6607    4194  99.9   7.3    8.9    5.4   1.8     7.3    13.6     5
    2    6607    3145  99.9   3.3   12.6    1.7   1.6    20.0     5.1     2
    3    6607    3145  99.9   2.8   14.6    1.6   1.2    19.3     8.4     2
    4    6607    3145  99.9   3.6   11.7    1.9   1.7    16.5     6.1     3
    5    6607    3145  99.9   4.2   11.0    1.8   2.3    18.7     4.8     1
    6    6714   13161  99.1   6.6   11.5    4.4   2.1    13.3     7.7     6
    7    6714   13161  38.8   4.7   19.1    1.2   3.4    27.3    16.2     2
    8    6714   13161  38.8   4.3   18.7    1.9   2.3    21.8    16.2     2
    9    6714   13161  38.8   4.2   20.6    1.5   2.6    26.7    17.1     2

Bottom
 Command
Patrick Conner
www.ConnecTown.com
(828) 244-0822




                    Pat Barber
                    <mboceanside@worldne        To:
MIDRANGE-L@midrange.com
                    t.att.net>                  cc:
                    Sent by:                    Subject:     Re: WrkSysSts
                    owner-midrange-l@mid
                    range.com


                    09/07/00 02:13 PM
                    Please respond to
                    MIDRANGE-L







Contractor1@Parkdalemills.com wrote:
>
> When I do a WrkSysSts and the F16 to look at the status of the DASD, How
> reliable is this information?
>
> The reason I as is that we've recently added some more disk to our
> development box. A few days after the install I looked at the disk and
the
> old disk were at 88.8%. The new ones at 10%.  We've been moving a lot
more
> information to the development box now and the old disks are at 99.7+%
and
> the old ones are at 37%. I expected the new disks to go up, but not the
old
> ones. I'm wondering if I'm misinterpreting the WrkSysSts F16 display.


That's the way it works... That information is correct. The only way
you can get a better balance is to do a complete reload which you are
proably not wanting to do. There is a newer command on disk balance
but that may only work on very recent releases. The system will attempt
to keep ALL of OS/400 on the first drive if at all possible.
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to
MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator:
david@midrange.com
+---




+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator:
david@midrange.com
+---

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---