Monitoring Disk Drive Usage

To detect problems with disk drive space, keep track of disk drive usage by executing the WHAT command regularly. This command lists, among other things, the amount of available disk drive space. An example of the top three lines of output of the WHAT command follows.

The example above displays the amount of available disk drive space. There are 51006 frames currently available. If you find you are running out of disk drive space, you have two options:

  1. You can replace your present disk drive with a larger one.

    or:

  2. You can remove various unnecessary files from the disk drive and reorganize those that remain.

The most obvious first step is to ask all users to delete any unnecessary files, especially word processing files. (Word processing files can take up a lot of space.) Files and accounts that are not currently needed on line can be backed up to tape. You can restore them when you need them.

Available frames that are randomly scattered in small blocks on the disk drive can result in longer response time and may even inhibit the creation of new files if there are no contiguous blocks of available space large enough to allow new files or accounts to be created. Use the POVF command to learn how the available space is distributed.

The POVF command displays the Overflow Table (POVF stands for Print OVerFlow). The example below shows what the Overflow Table might look like after a period of normal system operation.

6505-

6505 :

1

6513-

6527 :

15

6529-

6537 :

9

6539-

6542 :

4

6544-

6548 :

5

6550-

6584 :

35

6586-

6589 :

4

6593-

6605 :

13

6610-

6619 :

10

6621-

6622 :

2

6625-

6625 :

1

6627-

6637 :

1

6639-

6640 :

2

6642-

6646 :

5

6651-

8124 :

1474

8143-

8144 :

2

8488-

8499 :

12

8519-

8520 :

2

8583-

8583 :

1

8628-

8629 :

2

8636-

8653 :

18

8771-

8777 :

7

8779-

8780 :

2

8784-

8786 :

3

8814-

8820 :

7

8822-

8822 :

1

8826-

8830 :

5

8856-

8857 :

2

8859-

8863 :

5

8868-

8884 :

17

8886-

8915 :

30

8917-

8923 :

7

8926-

8930 :

5

8943-

8944 :

2

8946-

8948 :

3

8951-

8957 :

7

8959-

8960 :

2

8962-

8970 :

9

8972-

8979 :

8

8981-

8994 :

14

8997-

9002 :

6

9004-

9008 :

5

9010-

9012 :

3

9015-

9015 :

1

9017-

9023 :

7

9025-

9029 :

5

9031-

9033 :

3

9035-

9035 :

1

9038-

9039 :

2

9042-

9045 :

4

9047-

9056 :

10

9058-

9073 :

16

9075-

9078 :

4

9080-

9089 :

10

9091-

9093 :

3

9095-

9098 :

4

9100-

9109 :

10

9111-

9123 :

13

9125-

9134 :

10

9136-

9137 :

2

9150-

9161 :

12

9163-

9184 :

22

9186-

9199 :

14

9201-

9207 :

7

9209-

9211 :

3

9213-

9227 :

15

9229-

58063 :

48835

 

 

 

 

 

 

ADDITIONAL FRAMES

 

 

:

152

 

 

 

 

 

 

 

 

 

 

 

TOTAL NUMBER OF AVAILABLE FRAMES

:

50977

 

 

The Overflow Table can list up to 150 blocks of contiguous available frames. If there are more than 150 blocks of contiguous available space on the disk drive, all other available frames are linked together one at a time as they become free. These linked frames are known as linked overflow. These linked frames will not be included in the TOTAL NUMBER OF AVAILABLE FRAMES line of the POVF report, but they will be included in the OVRFLW total shown in the WHAT report.

The POVF report shown above gives the frame IDs of the first and last frames of each block of contiguous frames, followed by the total number of frames in the block. The additional frames are prelinked frames set aside for use as stacked workspace by the EXECUTE statement. The number of prelinked frames are always divisible by 152, since 152 frames are set aside for each EXECUTE statement. Finally, the total number of available frames is given (minus the linked overflow frames). No linked overflow is listed in the report shown above.

When there are linked overflow frames, they are listed in the first line of the report, before the contiguous frames are listed. The frame ID of the first frame of linked overflow is shown, followed, in parentheses, by the total number of linked frames.

CAUTION

If there are many frames listed in linked overflow, the disk drive is seriously fragmented. An overly fragmented disk drive slows down the system, and makes it more likely that GFEs will occur. A fragmented disk drive should immediately be reorganized by performing a filesave and restore (see Backing Up Files in the mvBase Guide to Files and Accounts).

The example below shows what the Overflow Table might look like immediately after a file-restore, when all available disk drive space has been consolidated into a single block of contiguous available space:

>POVF

   8454-  58079  :                  49626

ADDITIONAL FRAMES                                            0

TOTAL NUMBER OF AVAILABLE FRAMES   49626

System Files That Grow

Certain system files automatically grow if they are not watched and periodically cleared. Among them are the ACC file, the SECURITY-LOG file, and the SYSTEM-ERRORS file.

Accounting history items in the ACC file are automatically updated whenever designated users log on and off the system. Designated users are those with a U (Update) in line 9 of their Account Definition items or user-ID items. Values are added to each user’s accounting history item each time they log off the system. Approximately 1000 logon/logoffs are allowed per item. If the account history item for a user exceeds the available workspace, the user will be logged off, the ACC file will not be updated, and this message will be displayed:

[338] ACCOUNT FILE STATISTICS WERE NOT UPDATED DUE TO EITHER:

1. INSUFFICIENT WORK-SPACE TO CONTAIN THE ACCOUNT FILE ITEM, OR

2. SYSTEM DICTIONARY CHANGED WHILE YOU WERE LOGGED ON.

Items in the SECURITY-LOG and SYSTEM-ERRORS files are similarly added to whenever a security violation or system error is detected. (The SECURITY-LOG and SYSTEM-ERRORS files are discussed in Error Recovery and Troubleshooting.)

To avoid overflowing these logs, they should be deleted or cleared periodically. Accounting files can be deleted or cleared after the programs that generate summary data have been run; other system log files can be printed out or backed up to tape before regular deletion or clearing.

See Also

Monitoring System Activity

Monitoring Response Time

Monitoring and Controlling Processes

Monitoring Errors