× The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.



Hi Rob

It would be even easier to just name the library in such a way that the restore doesn't have an issue...

I have seen performance issues using *NEW with a lot of objects, and particularly noticeable on the old tape drives. With an LTO3 it is not such a big deal but of course it depends on just how many libraries have missing objects. It seems that checking whether the object on tape is on the system or not can add quite a bit of overhead depending on how many objects there are and how many _won't_ turn out to be new.

The performance overhead of *NEW over a 400gb system using LTO1 was bad enough that rather than run through the whole tape again I scanned the job log for message CPF3779 to determine which libraries were not completely restored and for message CPF3773 to determine the number of objects in the library, the number not restored and the tape sequence number. (It's always worth specifying OUTPUT(*PRINT) on restores to ensure a record of the restore operation hangs around)

Running the restore specifying the sequence number and just using a vanilla RSTLIB was still much faster then specifying OPTION(*NEW) and letting the whole tape get read again. The downside was that each restore must be checked to ensure the right number of objects ended up in the library - the RSTLIB output will detail which objects were legitimately not restored on the second restore, for instance *DTAQs don't get restored if they already exist, so this must be allowed for.

Running individual RSTLIBs, specifying the SEQNBR, ENDOPT(*LEAVE) and also specifying OPTION(*NEW) was easily the fastest way to get this done, but obviously requires a bit more work than just specifying OPTION(*NEW) and letting the whole thing run.

My personal opinion is that once there are any objects missing from the initial restore, then the subsequent restores need to be checked whatever the method used to perform them to ensure all objects ended up back on the system. Any assumptions at all are likely to bite you at some stage.

In the same vein, libraries containing journal objects that get restored after the files in the journal are already restored can cause you more than just a few minor problems, particularly because you don't get any error messages until you try and use the journal entries (like in a commit control operation !) Everything restores just handy dandy, but journaling of the files doesn't get started. I got bit bad on this one once on a 520 migration at 2:00am in the morning....

Regards
Evan Harris


At 12:44 a.m. 1/05/2007, you wrote:
Then again, what problem are you trying to solve? What you need to
restore after a catastrophe? Frankly running the second restore with the
*NEW is pretty fast. And, unless you have that file dumped to something
else JUST prior to the catastrophe you won't have the necessary
information, right? Because after the restore, until you get the cross
library logicals restored, they will not be in that file, right? And, if
the last time you dumped that file was prior to any new cross library
logicals then it's out of date also.

Rob Berendt
--
Group Dekko Services, LLC
Dept 01.073
PO Box 2000
Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





PaulMmn <PaulMmn@xxxxxxxxxxxxx>
Sent by: midrange-l-bounces@xxxxxxxxxxxx
04/27/2007 09:29 PM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
midrange-l@xxxxxxxxxxxx
cc

Fax to

Subject
Re: Cross library logicals.







You don't need a fancy audit; the system-maintained file,
QSYS/QADBFDEP, maintains a file of cross-library references.

A query over this file will take just a few minutes to run.

--Paul E Musselman
PaulMmn@xxxxxxxxxxxxxxxxxxxx



Query . . . . . . . . . . . . . . . . . PFLFXLIB
Library . . . . . . . . . . . . . . .your.lib.name.here
Query text . . . . . . . . . . . . . . Cross-Library File Dependencies
Query CCSID . . . . . . . . . . . . . . 65535
Query language id . . . . . . . . . . . ENU
Query country or region id . . . . . . US
*** . is the decimal separator character for this query ***
Collating sequence . . . . . . . . . . Hexadecimal
Processing options
Use rounding . . . . . . . . . . . . Yes (default)
Ignore decimal data errors . . . . . No (default)
Ignore substitution warnings . . . . Yes
Use collating for all compares . . . Yes

Selected files
ID File Library Member Record Format
T01 QADBFDEP QSYS *FIRST QDBFDEP

Result fields
Name Expression Column Heading Len Dec
DBFFIL10 substr(DBFFIL,1,10)
DBFFDP10 substr(DBFFDP,1,10)
DBFLB210 substr(DBFLB2,1,10)

Select record tests
AND/OR Field Test Value (Field, Numbers, or 'Characters')

DBFLIB NE DBFLDP
AND DBFLIB NE 'QSYS'

OR DBFLIB NE DBFLDP
AND DBFLDP NLIST 'QSYS2' 'QIWS'

Ordering of selected fields

Field Sort Ascending/ Break Field
Name Priority Descending Level Text

DBFLIB 10 A Dependency Library name
DBFFIL10
DBFLDP 30 A Library name of dependent file
DBFFDP10
DBFTDP Dependency: D-data,V-view,I-indirect
DBFRDP Format name of dependent file
DBFLB210

Report column formatting and summary functions
Summary functions: 1-Total, 2-Average, 3-Minimum, 4-Maximum, 5-Count
Overrides
Field Summary Column Dec Null Dec Numeric
Name Functions Spacing Column Headings Len Pos Cap Len Pos Editing

DBFLIB 0 DEPENDENCY 10
LIBRARY
DBFFIL10 2 DBFFIL10 10
DBFLDP 2 DEPENDENT 10
FILE
LIB
DBFFDP10 2 DBFFDP10 10
DBFTDP 2 DEP 1
TYPE
DBFRDP 2 DEPENDENT 10
FORMAT
NAME
DBFLB210 2 DBFLB210 10

Selected output attributes
Output type . . . . . . . . . . . . . . Printer
Form of output . . . . . . . . . . . . Detail
Line wrapping . . . . . . . . . . . . . No
Printer Output
Printer device . . . . . . . . . . . . *PRINT

Report size
Length . . . . . . . . . . . . . . . 66 (default)
Width . . . . . . . . . . . . . . . . 132
Report start line . . . . . . . . . . . 6 (default)
Report end line . . . . . . . . . . . . 60 (default)
Report line spacing . . . . . . . . . . Single space
Print definition . . . . . . . . . . . No

Printer Spooled Output
Spool the output . . . . . . . . . . . (Defaults to value in print
file, QPQUPRFIL)
Form type . . . . . . . . . . . . . . . (Defaults to value in print
file, QPQUPRFIL)
Copies . . . . . . . . . . . . . . . . 1
Hold . . . . . . . . . . . . . . . . . (Defaults to value in print
file, QPQUPRFIL)
Cover Page
Print cover page . . . . . . . . . . . Yes
Cover page title
Cross-Library File Dependencies
Page headings and footings
Print standard page heading . . . . . . Yes
Page heading
Cross-Library File Dependencies
Page footing
* * * * * E N D O F Q U E R Y P R I N T * * * * *
--

As an Amazon Associate we earn from qualifying purchases.

This thread ...

Follow-Ups:
Replies:

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].

Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.