Wonderful. Thank you Pete.
I see most of the entries are from Agents and Work processes are occupying the RT table. Is there anyway we can make sure that the agent and work processes doesnt store its logs in the DB but stores it in the OS level ?
Original Message:
Sent: 12-02-2019 04:27 PM
From: Pete Wirfs
Subject: RT table increasing
When I want to know which objects are writing the largest reports to RT, I use this (SQLServer) query.
Note that I also use ah_timestamp1 as a filter to limit how far back in time it will search. I use this to speed up the query.
select oh_name as job_name --, ah_timestamp1 as activation_time , dateadd(hour, datediff(hour, getutcdate(), getdate()), ah_TimeStamp2) as Start_time --, ah_runtime as runtime , CONVERT(varchar, DATEADD(ms, ah_runtime * 1000, 0), 114) as runtime , count(*) as report_sizefrom uc4.dbo.rt , uc4.dbo.ah , uc4.dbo.ohwhere ah_timestamp1 > cast('20191201 00:00:00:000' as DATETIME) -- how far back in timeand ah_oh_idnr = oh_idnrand rt_ah_idnr = ah_idnrand not oh_name in ('APPUTILK', 'AE_PROD#WP001') -- Objects to exclude from the selectiongroup by oh_name, ah_timestamp1, ah_timestamp2, ah_runtimehaving count(*) > 500 -- MAX_REPORT_SIZE is set in UC_HOSTCHAR_DEFAULTorder by 2;
------------------------------
Pete
Original Message:
Sent: 12-02-2019 11:04 AM
From: Franz Zalubil
Subject: RT table increasing
Please also double check if client 0000 is in the automic reorg definitions.
- ucybdbar
- ucybdbun
- ucybdbre
if you dont need and run archiving of old statistics please change in the ucybdbun.ini and ucybdbar.ini
no_Archive_Check=0 to no_Archive_Check=1
Missing Client 0000 in Reorg or no need to run archiving but forget to change the ini files is often the root cause why A*, R* and MELD Tables are conitously growing.
If you have allowed SQLI Variables you can define a Variables object what is giving you back automatically all Clients in the System:
Select OH_Name from OH where OH_OType = 'CLNT' and OH_NAME <> 'CLNT'
With a for each workflow you can refrence this variables object. It will automaticall publish client to define Jobs for each client defined in your System.
For example you publish ist as &Automic_Client#
UCYBDBRE Job flag reords to be delted would look like this:
uybdbre -B -S&Automic_Client#
UCYBDBUN Job delting all flaged records:
ucybdbun -BREORG -C&Automic_Client#
This workflow you can run on a daily base, to limit the numbers of records deleted in one run. This helps to reduce the load on the db to compare it with a weekly or monthly based execution.
For sure if the reorg never was running with sucess before, see my comments why this might happen, I would recommend to change the values in the ucybdbre definitions (run ucybdbre or for experts UC_UTILITY_REORG Variables in each client) to higher values (older than days). To reoganice and delete records of a too huge time frame will make a huge load on your DB and might slow down the db repsons time to the Automation Engine.
Hope this helps,
Franz
------------------------------
Managing Consultant Enterprise Studio
HCL Technologies Austria GmbH
Original Message:
Sent: 11-25-2019 10:38 AM
From: uc4_team
Subject: RT table increasing
Hello,
We are experiencing huge increase in our DB.
Two months ago we upgraded to version 12.2.
We keep logs 25 days back. Tho our RT table, increasing insanely (contains more than 280,000,000 rows!!)
It doesn't make sense and of course, affect our system performance.
Does anyone knows this issue? Do you know if it is a known bug?
Many thanks,
Ella