| Author |
Message
|
| premkumar |
 Posted: Wed Mar 24, 2004 7:57 am Post subject: Urgent! The Workflow's runtime database is growing |
 |
|
Apprentice
Joined: 24 Feb 2004
Posts: 40
|
Dear All,
In the production, Workflow runtime database filesystem is getting increased drastically and the filesystem is getting up filled..
For past 20 days the database usage is increased to 4GB which is 10% of the file system.
As i have analysed and found file SQL00008.DAT size is 26GB in the directory /var/fmc/rt_db/db2inst1/FMCDB/ACTWI/.
Find the files and their size respectively which are growing...
/var/fmc/rt_db/db2inst1/FMCDB/ACTWI>>ls -ltr
total 66976008
-rw------- 1 db2inst1 db2iadm1 512 Dec 10 2002 SQLTAG.NAM
-rw------- 1 db2inst1 db2iadm1 4096 Feb 28 07:02 SQL00005.DAT
-rw------- 1 db2inst1 db2iadm1 12288 Mar 24 08:20 SQL00004.INX
-rw------- 1 db2inst1 db2iadm1 4096 Mar 24 08:20 SQL00004.DAT
-rw------- 1 db2inst1 db2iadm1 4096 Mar 24 08:20 SQL00003.DAT
-rw------- 1 db2inst1 db2iadm1 12288 Mar 24 15:40 SQL00003.INX
-rw------- 1 db2inst1 db2iadm1 3889864704 Mar 24 20:14 SQL00006.DAT
-rw------- 1 db2inst1 db2iadm1 800972800 Mar 24 20:15 SQL00006.INX
-rw------- 1 db2inst1 db2iadm1 3371008 Mar 24 20:22 SQL00002.INX
-rw------- 1 db2inst1 db2iadm1 1179648 Mar 24 20:49 SQL00002.DAT
-rw------- 1 db2inst1 db2iadm1 268648448 Mar 24 20:52 SQL00007.INX
-rw------- 1 db2inst1 db2iadm1 26979168256 Mar 24 20:52 SQL00008.DAT
-rw------- 1 db2inst1 db2iadm1 566595584 Mar 24 20:53 SQL00008.INX
-rw------- 1 db2inst1 db2iadm1 1781657600 Mar 24 20:53 SQL00007.DAT
/var/fmc/rt_db/db2inst1/FMCDB/ACTWI>>
Note:
1. I have checked and Confirmed the AUDIT_TRAIL is OFF.
2. The number of records in the WORK_ITEM table is 2555 only.
3. I have tried to run the command db2 reorgchk for getting the report. After One hour the report was generated and found some of the tables requires reorg.
Does reorg will help in solving the problem?
Please help me on this issue and it is very critical.
Thanks and Regards,
Premkumar.S
_________________
Thanks and Regards,
Premkumar.S
Last edited by premkumar on Wed Mar 24, 2004 7:28 pm; edited 1 time in total |
|
| Back to top |
|
 |
| hos |
| Posted: Wed Mar 24, 2004 9:20 am Post subject: |
|
|
Chevalier
Joined: 03 Feb 2002
Posts: 470
|
Hi,
did you ensure that the cleanup server is running on a regulary base? Check the system and the error log.
You may also check that you retention intervalls for various objects
like workitems or log entries has reasonable values. |
|
| Back to top |
|
|
| premkumar |
| Posted: Wed Mar 24, 2004 7:26 pm Post subject: |
|
|
Apprentice
Joined: 24 Feb 2004
Posts: 40
|
| hos wrote: |
Hi,
did you ensure that the cleanup server is running on a regulary base? Check the system and the error log.
You may also check that you retention intervalls for various objects
like workitems or log entries has reasonable values. |
Hi,
The cleanup server is running for every one hour.
There is no error found in db2 log, workflow logs and other error logs.
As a whole the application has only 2500 workitems...
Thanks and Regards,
Premkumar.S
_________________
Thanks and Regards,
Premkumar.S |
|
| Back to top |
|
|
| olan022 |
| Posted: Thu Mar 25, 2004 5:28 am Post subject: |
|
|
Acolyte
Joined: 17 Apr 2002
Posts: 50
|
Connect to your runtime database via DB2 and run the following:
select count(*) from fmc.process_inst where state=128
Note the number of records it shows and increase your cleanup server schedule if needed. The records returned indicates how many processes still need to be cleared out by the cleanup server. I personally am running it for 6 hours a day and find that may be too little. |
|
| Back to top |
|
|
| premkumar |
| Posted: Thu Mar 25, 2004 9:59 pm Post subject: |
|
|
Apprentice
Joined: 24 Feb 2004
Posts: 40
|
| olan022 wrote: |
Connect to your runtime database via DB2 and run the following:
select count(*) from fmc.process_inst where state=128
Note the number of records it shows and increase your cleanup server schedule if needed. The records returned indicates how many processes still need to be cleared out by the cleanup server. I personally am running it for 6 hours a day and find that may be too little. |
There is no records for the above query. And i have scheduled the cleanup server for every 30 minutes.
_________________
Thanks and Regards,
Premkumar.S |
|
| Back to top |
|
|
| hos |
| Posted: Tue Mar 30, 2004 7:18 am Post subject: |
|
|
Chevalier
Joined: 03 Feb 2002
Posts: 470
|
Hi,
the mentioned tablespace contains three tables, which are all related to activity work items. Since the ACTIVITY_WORK_ITEM table contains a similar number of records as the WORK_ITEM table a reorg on the three tables: ACTIVITY_WORK_ITEM, LOC_W_I_STUBS_SEG, REM_W_I_STUBS_SEG should help.
Beside that check your workflow model whether you specified KEEP_WORKITEMS FOREVER
in one of your active processes. |
|
| Back to top |
|
|
| sshaker |
| Posted: Tue Mar 30, 2004 12:16 pm Post subject: |
|
|
Disciple
Joined: 20 Sep 2002
Posts: 185
|
| Quote: |
Does reorg will help in solving the problem?
|
yes, if u have deleted lots of data using fmcsclad or cleanup server etc.
it will reclaim the storage used by those deleted.
_________________
shaker |
|
| Back to top |
|
|
|
|
