IBM Support

Reducing System ASP Disk Space (DASD) / Storage Used

Troubleshooting


Problem

This document describes common techniques for reducing the amount of disk space used.

Resolving The Problem

IBM i Support Documentation in this document:

644435, Go Disktasks Overview
6323041, Overview of RCLSTG
706651, Temporary Storage PTFs

An example of a system running at a high percentage of system ASP used is shown below. The percentage of system ASP used is high.
 
                       Display System Status           SYSTEM
                                            01/01/21  14:47:14
% CPU used. . . . :     26.7   Auxiliary storage:      
Elapsed time. . . : 00:00:27     System ASP . . . . . : 27.86 G
Jobs in system:         2767     % system ASP used  . : 99.8263
% addresses used:                Total  . . . . . . . : 27.86 G

  Permanent:            .009     Current unprotect used: 359 M  
 Temporary:            .005     Maximum unprotect . . : 369 M

If permanent storage growth, a DISKTASKS report is recommended. It shows the allocation of disk space across the system. At least 3% free space is suggested to generate the output. However, more space may be required. Note: The collection may take up to several hours to complete. For more information on DISKTASKS, refer to the following document, Go Disktasks Overview .

This document is divided into two categories. The first portion includes tips to reduce Permanent Space allocated; the second part shows areas to reduce Temporary Space allocated.
 
Caution: All of the following actions should be taken with caution and at the customer's discretion. Some of these functions may be long-running. They are common ways to reduce storage used. Some of these functions will delete, clear, or destroy data that may be unrecoverable. Also note that any of the following can consume the remaining amount of space without warning. In addition, some commands may not be available at all releases.

To reduce permanent space:
O To delete old history files (except the most recent one), on the operating system command-line type the following:

WRKOBJ OBJ(QSYS/QHST*) OBJTYPE(*FILE)

Press the Enter key. Select Option 4 to Delete.
O To verify CLEANUP is running, type the following:

CHGCLNUP

Press F4 to prompt. Verify that Allow cleanup is set to *YES. Then, verify the Number of days to keep objects is within reason.
O To clear QEZJOBLOG and QEZDEBUG output queues (if not needed to debug storage growth or other open problems), on the operating system command-line type the following:

CLROUTQ OUTQ(QEZJOBLOG)

Press the Enter key.

CLROUTQ OUTQ(QEZDEBUG)

Press the Enter key.
O To clear any job queues that are not needed, on the operating system command-line type the following:

CLRJOBQ JOBQ(XYZ)

Press the Enter key.
O To clear any user libraries that are not needed, on the operating system command-line type the following:

CLRLIB LIB(XYZ)

Press the Enter key.
O To delete journal receivers, on the operating system command-line type the following:

DSPOBJD OBJ(*ALL/*ALL) OBJTYPE(*JRNRCV)

If they are large, on the operating system command-line type the following:

WRKJRNRCV JRNRCV(*ALL/*ALL)

Press the Enter key. Select Option 5 to display and check the status. If SAVED or ON-LINE, they may be possibilities to delete.

Follow the GO DISKTASKS report from this link: https://www.ibm.com/support/pages/go-disktasks-overview

To get the information on journal receivers:
A) GO DISKTASKS
B) Option 2. Print disk space information
C) Check the top to verify date/time the collection was taken, if not current, collect one: https://www.ibm.com/support/pages/go-disktasks-overview
D) Specify 1 for Specific object
E) Objects . . . . . . . . . . . . *ALL
Object types . . . . . . . . *JRNRCV
Size of smallest object . . . . 0
F) That generates a report, enter WRKSBMJOB, the spooled file is named QPRTOBJRPT

At the top of the report, it shows the journal receiver consuming the most storage.

If Data Replication over remote journaling is being used, do not delete the journal receiver in the HA solution.

Via SQL: With current PTF levels at least V5R4x and up, to get a list of storage for all journal receivers for a journal on the system. (The most storage is at the top.)

select sys_dname, jrnname, JOURNAL_ASPGRP, SIZJRNRCV
from qsys2/jrninfo
order by SIZJRNRCV desc
O After deleting documents and folders, do the following:

RCLDLO DLO(*ALL)

Press the Enter key.
O Delete any License Program Products (LPPs) that are not needed (if unsure, save the LPP to tape, and then delete). On the operating system command-line, type the following:

GO LICPGM

Press the Enter key. Select Option 12, Delete licensed programs, Option 4 next to LPP(s) to delete.
O Delete any copies of Main Storage Dumps (MSD) that are not needed:

STRSST

Press the Enter key.

Select Option 1, Start a service tool, and press the Enter key.
Select Option 6, Main storage dump manager, and press the Enter key.
Select Option 2, Work with copies of main storage dumps, and press the Enter key.
Select Option 4, Delete, and press the Enter key.
O Clear performance libraries QMPGDATA and QPRFDATA (be aware that there may be other PM/400 issues involved).

On the operating system command-line, type the following:

If the system is pre-R450, ENDPFRMON
If the system is R450 or later, GO PERFORM, option 2, option 3 to end

Press the Enter key. On the operating system command-line, type the following:

CLRLIB LIB(QPFRDATA)

Press the Enter key. On the operating system command-line, type the following:

CLRLIB LIB(QMPGDATA)

Press the Enter key. To restart after the library has cleared, on the operating system command-line type the following:

GO PERFORM, option 2, option 1

Press the Enter key.
O Delete spooled files that are no longer needed, then run RCLSPLSTG. This removes the storage that the spooled files allocated however does not remove spooled files that are not printed. On the operating system command-line, type the following:

WRKSPLF

Press the Enter key. Select Option 4 to Delete next to spooled files to delete. Then, on the operating system command-line type the following:

RCLSPLSTG DAYS(*NONE)

Press the Enter key.
O Reduce the number of days to keep unused spooled storage, if possible.

WRKSYSVAL SYSVAL(QRCLSPLSTG)

Press the Enter key.
O CLRPFM on QRECOVERY/QSQ901S at R540 and beyond. This member logs information on all msgSQL0901's on the system up and can be cleared if you are not pursuing SQL issues.
O
Clear the directories where SQE logs errors: TMP/SQE (before May 2020) and /QIBM/UserData/OS400/SQE (after May 2020 PTFs)
Assuming that you are not pursuing SQE related issues.
O To archive old data to tape, on the operating system command-line type the following:

SAVLIB
or
SAVOBJ STG(*FREE)

Note: There is a parameter that will save data to tape, then delete the data on the disk while leaving the object headers in place so private authorities do not get lost.
O Clear old problem entries:

DLTPRB

Also check the system value QPRBHLDITV (it allows you to specify the number of days to keep the problem in the problem log).
O RCLSTG can be a long-running function and requires the system to be in a restricted state. Consider the Reclaim Storage command only if there is a warranted reason to run it. This could include results from Go Disktasks Reports in the Storage Affected by RCLSTG section.

To end the subsystem, on the operating system command-line type the following:

ENDSBS SBS(*ALL) OPTION(*IMMED)

Press the Enter key. Then, type the following:

RCLSTG

Press the Enter key.
O Delete PTF save files on the machine that you are not going to apply. On the operating system command-line, type the following:

DLTPTF PTF(XXXXXXX) LICPGM(XXXXXXX)

Press the Enter key (for the given product and PTFs).
O Delete program temporary fix (PTF) save files, their associated cover letters, and the records of PTFs that have been ordered as applicable:

DLTPTF PTF(*ALL) LICPGM(*ALL)
O Permanently apply temporary PTFs.

Note: Do not permanently apply test PTFs.
O Review and then, if possible, clear out the system recommended index file:
CLRPFM QSYS2/SYSIXADV
O
Review:
QRECOVERY/QSQ901S

At R540 (and later) this file will now record information on msgSQL0901 errors. It can have a CLRPFM done on it in low storage conditions.
O If SMTP TRACE DATA is being saved, QRPLOBJ and QUSRSYS will show high amounts of space in a DISKTASKS report. To end the traces and capture the space, on the operating system command-line, type each of the following commands and press the Enter key after each:

TRCTCPAPP APP(*SMTPSVR) SET(*END)
TRCTCPAPP APP(*SMTPCLT) SET(*END)
ENDTCPSVR SERVER(*SMTP)
STRTCPSVR SERVER(*SMTP)
For an alternate review, by user profile, you can use the SQL statement:
select                        
      AUTHORIZATION_NAME,     
      ASPGRP,                 
      STORAGE_USED,           
      MAXIMUM_STORAGE_ALLOWED 
from USER_STORAGE             
order by storage_used desc    
to see what profiles own what storage, sorted with the highest storage owning profiles at the top.
Reducing Temporary Space

Determine if there are any applicable temporary storage PTFs using support document 706651,
Temporary Storage PTFs

 
O On the operating system command-line, type the following:

WRKSYSACT

Press the Enter key. Then, press PF11 three times and sort by allocated space.
O On the operating system command-line, type the following:

WRKACTJOB

Press the Enter key. Sort by CPU %.
O If excessive IFS space is tied up in this directory: /QIBM/UserData/webquery/ibi/srv76/wfs/edatemp, stop and restart WebQuery using
ENDWEBQRY, followed by STRWEBQRY. Note: Consult your DBWEBQ Support team with any questions.
O To end the subsystems, on the operating system command-line type the following:

ENDSBS

Press the Enter key.
O To end specific jobs, on the operating system command-line type the following:

ENDJOB

Press the Enter key.
O To free up all temporary space by jobs currently on the system, IPL the system.
O Force an MSD, IPL.
O Get iDoctor and / or DIRDUMP on the system. Note: A consulting agreement is required for analysis of iDoctor and DIRDUMP.
If the system is not up and will not IPL in NORMAL mode, do the following:
O Do a manual IPL. Then, clear incomplete job logs, clear out queues (note that all spooled files in all output queues will be deleted), and clear job queues from the IPL Options screen.
O During a step mode IPL, reduce the size of all LIC Logs and the Product Activity Log.
O Validate Work Management Job Table System Value:

WRKSYSVAL SYSVAL(QTOTJOB)
O Go into DST. Start a service tool, take the Main storage dump manager, and work with copies of Main Storage dumps. Then, delete the selected dump.

[{"Type":"MASTER","Line of Business":{"code":"LOB57","label":"Power"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SWG60","label":"IBM i"},"ARM Category":[{"code":"a8m0z0000000CRjAAM","label":"Disk Information"}],"ARM Case Number":"","Platform":[{"code":"PF012","label":"IBM i"}],"Version":"7.1.0;7.2.0;7.3.0;7.4.0"}]

Historical Number

N1019466

Document Information

Modified date:
20 April 2022

UID

nas8N1019466