cancel
Showing results for 
Search instead for 
Did you mean: 

abnormal increase in the oracle data base of XI dev

Former Member
0 Kudos

Hi Experts,

The size of the oracle data base for SAP-XI is increased by 9% abnormally in 1 day, where as 1% increase in the oracle data base size normally takes 1 to 2 weeks.

What could be the problem? Please give me some links or some idea how to check to know what causes this problem.

Thnx

Subbu

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi Subbu,

How many messgaes are proceessing every day?

Databse size will grow because messages will store in the database.

Did you configure tretention period (to delete archive messages)?

Regards

Ben

Former Member
0 Kudos

Hi Ben,

Thanks alot for your quick responce.

Tretention period is i guess for every 15 days.

But i can say, so many users are interacting with Xi Dev server.

apart from that i dont have any clue.

Actually my problem is , why this abnormal increase? and how to pin point this problem? Are there any transations to see this abnormal increase?

Thnanks

Subbu

Former Member
0 Kudos

Hi Rama

You should look out for big tables. You can do this in transaction DB02 -> Detailed Analysis -> Size / Kbyte > 100000

This will give you all segments larger than 100mb. In an XI system the messages are stored in table XI_AF_MSG. It is also possible that your largest segment is a lob segment. You will then see something like SYS_LOB0000039303C00020$$.

To see which table it is just use this SQL:

[code]SQL> select table_name from dba_lobs where SEGMENT_NAME = 'SYS_LOB0000039303C00020$$';

TABLE_NAME

-


XI_AF_MSG[/code]

In DB02 you can also select the segment and view its history to make sure it is responsible for the database growth.

Best regards, Michael

Former Member
0 Kudos

Hi Subbu,

By default the max number of entries for the table SXMSPMAST is 1.000.000 entries, considered to be XI messages.

A DROP_MAX_TABLE_LOAD value of 90% meaning 900.000 messages are stored in the system until the switch and deletion is triggered.

Example:

Retention Period = 30 days

Throughput = 10 000 messages / day

 first time for triggering a table switch will be after 90 days (as fill level is reached).

Any message of age 30 days and older will be marked for deletion. Messages of age 0 – 29 days are kept, which means 61 days worth of messages will be deleted.

From that point on a “delta” growth of 61 days until the next deletion occurs.

This results in the following peak space usage considering 1Mb/entry: 900.000 MB + 290.000 MB = 1,19 TB for copy and dropping procedure

Regards

Ben

Answers (2)

Answers (2)

Former Member
0 Kudos

self resloved.

Former Member
0 Kudos

Hi Subbu,

use following reoprt to check the number of messages

RSXMB_SHOW_REORG_STATUS

Regards

Ben