Checking Cache Battery Status Without STRSST

September 28, 2011 Hey, Joe

Concerning your article about debunking cache battery rumors, did you know that there’s another way to see cache battery information without going into System Service Tools (STRSST)? IBM is offering a new program called QSMBTTCC that allows you to display cache battery status from the command line. It’s enabled by PTFs.

–Steve

As a refresher, IBM uses batteries in its disk controllers to provide caching for its disk drives. The batteries have a useful life of about 2.75 to 3 years and the system will start sending warning messages to change your batteries when your controller cache batteries are older than about 2.33 years. When one of these batteries dies, it may not cause a disk drive failure, but it will disable disk caching for the disk array the controller manages, and that will have a significant impact on hard drive performance.

The short version of all this is that it’s wise to replace your disk cache batteries before they run down.

Generally, disk cache batteries are hot swappable, meaning that they can be replaced while your system is up and running. However, depending on your machine type, one of the batteries may be controlling a disk array inside the system CPU and the system must be taken down to replace that battery.

Historically, the operating system sends battery status warning messages to the system history logs and the QSYSOPR message queue. The problem with this approach is that it relies on you reactively catching the message and then arranging for cache battery replacement before the batteries run down. You can check battery status on your own, but that involves running through a number of System Service Tools (STRSST) options, which are somewhat kludgy to remember and run.

As Steve mentioned, IBM delivered a green screen method last year for cache battery checking by adding a PTF to i OS operating systems V5R4Mx, V6R1Mx, and V7R1Mx. This PTF delivers a new operating system program called QSMBTTCC, and can be run with the following Call Program (CALL) command.

CALL PGM(QSYS/QSMBTTCC)

Once the program is called, it displays and prints a report containing information on partition resources that contain cache batteries and the expiration date of each set of cache batteries. This information includes the following items as shown on the sample report at the end of this tip.

The name of each system resource containing a cache battery.
The serial number of each cache resource.
The type and model number of each resource.
The frame ID and card position where each of the resources is installed in your rack.
Battery pack information, such as whether the resource contains a concurrently maintainable battery pack, what the estimated time is (in days) until the system starts issuing battery change warnings, and what the estimated time is (in days) until the battery pack might fail.

The trick with reading battery cache information is to change the batteries within a reasonable amount of time after you receive your first battery warning message, but well before you exceed the estimated time of error (when the batteries are scheduled to exceed their battery life). Remember that these times are only estimates. Your actual battery life may be longer or shorter than the number of days listed. The actual number of days until battery failure will depend on machine usage and a number of other variables affecting battery life. To avoid unexpected failure, it’s wise to replace your cache batteries not too long after you receive the first battery warning message.

Your partition needs a special PTF to enable QSMBTTCC. Here are the PTF numbers that need to be loaded on your system to use QSMBTTCC.

For i5/OS V5R4Mx, load and apply PTF SI40403.

For i OS V6R1Mx, load and apply PTF SI40404, which is also available in cumulative PTF level C1102610 and above.

For i OS V7.1, load and apply PTF SI40406, which is also available in cumulative level PTF C1116710 and above.

Each PTF can be applied immediately or on a delayed basis.

Special thanks to Steven Segars for providing me information for this tip.

HTH

–Joe

Sample QSMBTTC output

RUNNING MACRO: BATTERYINFO                   -LIST -ALL
***LIST OF ALL RESOURCES THAT HAVE CACHE***
                                                       CONCURRENTLY   CAN BE
RESOURCE  SERIAL         TYPE      FRAME     CARD      MAINTAINABLE   SAFELY
NAME      NUMBER         MODEL     ID        POSITION  BATTERY PACK   REPLACED
DC01      YL10P1057198   2BE1-001  3C00      C13       NO             NO
DC04      YL10D1112023   575C-001  3C03      C2        YES            NO
DC05      YL10D1096007   575C-001  3C02      C2        YES            NO
DC06      YL10D1096015   575C-001  3C03      C5        YES            NO
DC07      YL10P1080273   2BD9-001  3C00      C19       NO             NO
DC08      YL10D1096007   572F-001  3C02      C1        YES            NO
DC09      YL10D1112023   572F-001  3C03      C1        YES            NO
DC10      YL10D1096015   572F-001  3C03      C4        YES            NO
RUNNING MACRO: BATTERYINFO                   -LIST -WARN
***LIST OF ALL RESOURCES THAT HAVE CACHE
   WITH THE ESTIMATED TIME TO WARNING IN DAYS***
                                                       EST. TIME     EST. TIME
RESOURCE  SERIAL         TYPE      FRAME     CARD      TO WARNING    TO ERROR
NAME      NUMBER         MODEL     ID        POSITION  (IN DAYS)     (IN DAYS)
DC01      YL10P1057198   2BE1-001  3C00      C13          844           934
DC04      YL10D1112023   575C-001  3C03      C2           848           939
DC05      YL10D1096007   575C-001  3C02      C2           848           939
DC06      YL10D1096015   575C-001  3C03      C5           844           934
DC07      YL10P1080273   2BD9-001  3C00      C19          844           934
DC08      YL10D1096007   572F-001  3C02      C1           844           934
DC09      YL10D1112023   572F-001  3C03      C1           844           934
DC10      YL10D1096015   572F-001  3C03      C4           844           934

                     Post this story to del.icio.us
               Post this story to Digg
    Post this story to Slashdot