Shadowbase Support Tips

William Holenstein
Sr. Manager of Product Delivery

Shadowbase Secondary Usage: How to Measure Audit Generation from Audit Trail History Using Shadowbase Software

Here is a helpful method for determining the audit generation for a system to aid in the sizing and scaling of a Shadowbase environment without running a measure. This method is performed by generating a simple uni-directional Shadowbase configuration with a single DBS (an EXCLUDE SPEC DBS), and then running Shadowbase replication against the MAT and AUX audit trails for summarizing the previous days’ activity. During the run, the Shadowbase collector will compile statistics about the data read from the audit trail. The configuration will not replicate nor send any data to the replicate/target node.

There are a few limitations to using this method. First, the master and auxiliary audit-trail(s) must still be available on the system. The Shadowbase collector, by rule, cleans up tracking activity for a file or table unless there has been at least one audit event generated for the file/table in the previous seven days. Therefore, this method will work for statistical queries of seven days or less.

The tracking statistic will show all system-generated audit events (i.e., catalog and SMF updates). By default, reload events will not be tracked. The collector’s ARRETURNNDCRECS attribute can be enabled to view reload events.

Note, this example assumes the reader is familiar with a Shadowbase uni-directional EXPAND configuration, along with the product usage and the SBUNIDIR uni-directional configuration generator. For details on Shadowbase operation, please reference the ‘HPE NonStop Shadowbase Command Reference Manual,’ the ‘HPE NonStop Shadowbase Operations Manual,’ the ‘HPE NonStop Shadowbase Installation and Planning Manual,’ and the ‘HPE Shadowbase Tools and Utilities Manual’ for details on how to use SBUNIDIR.

Using SBUNIDIR, configure source and target replication on a same node loopback using EXPAND:

LOAD /KEEP 1/ $H2.$QA.QA6230.SBUNIDIR

SB_SET SB_CONFIGURATION_TYPE DIRECT

SB_ADD SRCNODE \H2 \H2.$QA.QA6230 \H2.$QA.SCRATCH &

~[#MYTERM~] \H2.$QA.QA6230.SHADPASS &

$SBCL 0 1 145 $SBMON 0 1 145

SB_ADD TGTNODE \H2 \H2.$QA.QA6230 \H2.$QA.SCRATCH

SB_ADD CONS $SBCN1 1 0 145 \H2.$QA.SCRATCH.TIDFIL1

Now add an override to configure an ADTSTARTTIME collector parameter for up to a week in the past.

SB_ADD COLL_PARAM ADTSTARTTIME (SEP 23 2016,16:40:00.0.0)

SB_ADD COLL_PARAM ADTSTARTEOF OFF

Configure a DBS exclude for all volumes, subvolumes and files

SB_ADD DBS X $*.*.* $SBCN1 – – –

==and set it as a type EXCLUDE with a DBS override:

SB_ADD DBS_PARAM X * SPECTYPE EXCLUDE

Change the internal Shadowbase source cleanup period to the maximum seven days (10,080 minutes) with a TACL override. The default is normally 12 hours.

SB_ADD TACL_PARAM * SBSRCCLEANUPAGE 10080

SB_ADD TACL_PARAM * SBSRCCLEANUPRECUR 10080

Generate the sbunidir scripts with sb_gen.
Make certain the Shadowbase restart file does not exit.
Run sbstrtsb to cold-start the Shadowbase environment, using the ADTSTARTxxx parameters specified, and with the single DBS EXCLUDE spec.
Use stats and estats in audcom to view the audit details once all the audit trail has been processed by the collector. Here is a way from AUDCOM to send the estats output to the spooler or edit file.

+ STATS /out $s.#consstatsout/ CONS *, DETAIL

+ ESTATS /out $s.#estatsout/ AUD, COLLSTATS ALL

Sample ESTATS output…

+estats aud,collstats all

SHADOWBASE – T1122 – V6230?06 – (15MAY16)

ESTATS AUD AT 10-03 12:33:48

===============================================================================

AUDMON: \H2.$SBMON CPU: 00 STATS RESET AT: 10-03 12:20:11 (COMMAND)

——————————————————————————-

—————— COLL/QMGR/CONS AS OF 10-03 12:33:47.931033 —————–

NAME CPU PNAME ADT POSITION LE TIME DIFF (CUR)

—————— — —— ——————– ————– ————-

SYSTEM: \H2

COLL-SBCL 0 $SBCL AA000060/01655025997 10-03 12:33:42 00:00:05.856

SYSTEM: \H2

CONS-DIR-SBCN1 1 $SBCN1 000000/00000000000 ??-?? ??:??:?? 00:00:00.000

—————————-COLL PARTITION STATS—————————–

NAME READ USED DO UNDO

—————– ————– ————– ————– ————–

SYSTEM: \H2 COLL: COLL-SBCL

\H2.$DATA.QAMETRO.METRO ADT: BB LE: 09-27 12:28:23.658 EXCL

INSERT 0 0 0 0

UPDATE 5507567 0 5507567 0

DELETE 0 0 0 0

CREATE 0 0 0 0

ALTER 0 0 0 0

PURGE 0 0 0 0

PURGEDATA 0 0 0 0

TOTALS 5507567 0 5507567 0

[…]

\H2.$VDV128.ZSMSCAT.ANT ADT: AA LE: 10-03 11:57:53.421 EXCL

INSERT 33 0 33 0

UPDATE 0 0 0 0

DELETE 27 0 27 0

CREATE 1 0 1 0

ALTER 0 0 0 0

PURGE 0 0 0 0

PURGEDATA 0 0 0 0

TOTALS 61 0 61 0

COLL COLL-SBCL TOTALS

INSERT 133205 0 133070 135

UPDATE 28393815 0 28393503 312

DELETE 25307 0 24299 1008

CREATE 703 0 703 0

ALTER 3260 0 3260 0

PURGE 0 0 0 0

PURGEDATA 65 0 65 0

TOTALS 28556355 0 28554900 1455

===============================================================================

Please reference our Newsletter Disclaimer.