Re: Unload/Reload -- MIDRANGE-L

Well the 'drives selected for end allocations' is a consequence of the Add and Balance option. What it does is prevent all writes of new data from ending up on those drives potentially causing them to 'go hot' even though it is still moving old data to those drives. Once balancing ends normally with all disk units evenly balanced that flag is reset and life proceeds as normal.

Looking at the system status the machine pool is acceptable but not swimming in memory on a system with 16GB. Most of the work once it's started is done in *BASE and that one is showing a lot of paging but faulting isn't too bad. The disks are very busy so I suspect that the balance function is doing it's thing. You can see the "Write K" is large on the new drives and the "Read K" is larger on the old ones. What's a bit odd to me is that even the new drives have more Read requests than Write requests.

I would let it run and certainly would not do ENDASPBAL because there is a significant amount of work done before any movement begins when you run STRASPBAL.

- DrF

On 9/17/2012 8:47 AM, rob@xxxxxxxxx wrote:

Work with Disk Status
Elapsed time: 00:50:23
Size % I/O Request Read Write Read Write %
Unit Type (M) Used Rqs Size (K) Rqs Rqs (K) (K) Busy
1 4328 141129 71.3 38.3 17.9 27.3 11.0 21.3 9.5 25
2 4328 105847 71.2 25.5 22.3 15.0 10.5 32.5 7.6 26
3 4328 105847 70.9 43.4 13.3 17.0 26.4 23.3 6.9 50
4 4328 105847 71.2 28.0 18.1 15.6 12.3 26.4 7.5 32
5 4328 105847 71.0 33.0 15.2 17.4 15.5 22.5 6.9 34
6 4328 141129 71.4 54.0 13.4 28.6 25.3 19.4 6.7 45
7 4327 52923 71.0 22.3 18.8 11.1 11.2 25.6 12.2 21
8 4327 52923 70.9 30.6 20.3 15.0 15.5 23.7 17.0 25
9 4327 52923 70.9 34.2 21.1 17.5 16.6 22.6 19.5 28
10 4327 52923 70.9 39.7 19.0 19.2 20.5 21.5 16.6 34
11 4328 141129 72.3 47.9 13.4 32.0 15.8 16.3 7.6 33
12 4327 70564 71.0 25.1 16.4 15.6 9.5 19.7 10.9 18
13 4327 70564 70.9 27.5 16.9 16.4 11.0 20.3 12.0 20
14 4327 70564 71.1 28.4 14.7 16.8 11.6 18.9 8.5 22
15 4328 141129 8.8 10.9 20.0 8.1 2.8 8.1 54.3 8
16 4328 141129 8.5 9.3 20.6 6.7 2.6 7.4 54.6 7
17 4328 141129 8.7 10.3 19.8 7.4 2.9 7.5 51.3 8
18 4328 141129 8.7 12.5 18.4 8.0 4.5 7.4 38.1 11

Work with System Status
Elapsed time . . . . . . : 00:39:29
System Pool Reserved Max -----DB----- ---Non-DB---
Pool Size (M) Size (M) Active Fault Pages Fault Pages
1 724.43 308.16 +++++ 2.5 2.5 12.9 26.3
2 2299.31 69.42 376 3.2 173.5 91.4 706.3
3 35.34 .00 10 .0 .0 .0 .0
4 10393.80 .00 98 1.1 38.1 168.4 372.6
5 3021.10 .03 478 2.0 492.0 8.1 33.8

DSPLOG PERIOD((*AVAIL 091512)) MSGID(CPI1474 CPI1475)
ASP balancing successfully started for ASP 1. 09/15/12 - 19:55:17
ASP balancing for ASP 1 ended. 09/16/12 - 11:53:39
ASP balancing successfully started for ASP 1. 09/16/12 - 14:25:46
09/15/12 - 19:55:17: Add drives and balance data.
09/16/12 - 11:53:39: Let's try ending the asp balancing to get rid of scary messages that drives are selected for end allocation.
09/16/12 - 14:25:46: STRASPBAL *CAPACITY

Development lpar. Six people currently in QINTER. One job in QBATCH. Two Domino servers, very lightly used (one email user, me; some minor development work).

Little graphic from iNav attached that the list will probably strip off.

Rob Berendt
--
IBM Certified System Administrator - IBM i 6.1
Group Dekko
Dept 1600
Mail to: 2505 Dekko Drive
Garrett, IN 46738
Ship to: Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com <http://www.dekko.com/>