Hello,
You are giving two conflicting requirements:
1) you are not allowed to delete old jobs
2) you must set keep clause -> this WILL delete old jobs
If you want to keep history without having to impact performance you will have to archive it.
Regards Gerben
Hello,
Sorry to interrupt the joy here, but this is the best way to make your system crash as soon as possible if it is already loaded to the top. You will need to do a more sophisticated way of deleting the old jobs, one by one, or at least in SMALL (100-1000) batches.
Note that the deleteFinalJobs will delete ALL jobs in history for this Job Definition. It will not keep the last 20 days, nothing.
When you are worried about performance impact, I would suggest to listen what SAP has to say about this and follow their command. In my opinion this forum is for asking regular questions and posting problems you face. You cannot put the safety of your production instance in the hands of a stranger, without all the insights and information required to make a professional judgement on the situation, here on the forum.
Regards Gerben
Hi Vishnu,
My intention is not to crash your production system.
It was my assumption that you will test this in your QA first and more importantly that you are aware that deletfinaljobs() will delete the complete past history of the job definition, given the nature of the method that you are not passing any keep values, if you were not I apologize.
So, one easy alternate i can think of is submitting System_RemoveOldJobs with the required age or date or id.
So loop through the same set of job definitions, but instead of deleting all past history, schedule this System job for each job definition, with specific age or date(for example, job A older that Jan 31, 2012). In this case, you might have to run this System job multiple times just for a single definition. So it requires meticulous planning, extensive testing in QA for all test cases (all frequencies for example). Keep in mind, you have to either schedule the system jobs in future date or wait for first System_RemoveOldJobs to complete before starting the second one.
The Key here is to determine the optimum age or date that you can repeat for each category of frequency but still not cause any performance issue.
For example, Job A, you choose Age as 47 months (deleting 1 month of data from 2012) and you run the System_RemoveOldJobs. You wait for that job to complete and then run it again for Job A for 46 months.
Of Course, create a message with SAP in parallel as suggested above for a standard solution.
thanks
Nanda
Hi Nanda,
Thanks for the clarification and there is no need for apologies in learning phase.
However, as suggested will contact SAP for standard solution because using system_removeOldjobs is not a feasible solution it seems as we have 5.3k jobs in production and out of which we have 3.5k jobs are of daily frequency nature.
Many thanks Nanda and Blom for constructive solutions.
Please give pointers if you come across any solution.
Hi Shwetha Houde and Praveen Sada SIVAM,
Are the credentials in the same partition as the process servers ?
Regards,
HP
Hi Gerben. Thanks for the response.
Yes perhaps, client initially was not willing to delete the jobs. And also, was not in option for Archive also.
Hence we persuaded and made them informed about the criticality of the old jobs and the impact on performance of the system.
So, now we just want to go with Keep Clause setting options with deletion of oldest jobs accordingly.
And, in case of any further possible inputs from you, will keep you posted on it.
Thanks.
Hi Nanda, Thank you.
Simultaneously, we have received this below information from SAP for similar suggestions on CPS for above based information about our client.
1. Alter your CPS Registry so that this entry is smaller than 365:
/configuration/history/jobhistory
Note: Note setting this to 30 will keep 30 days of Job History.
2. Execute job System_Aggregate_History - this job will then begin cleaning up your JobHistory tables based on the registry setting /configuration/history/jobhistory.
3. Configuration > jcs >core >monitorUpdateInterval # default is 30 and should be increased to 900
4. Configuration>requestQueueSize>GLOBAL>action- usaully just left at default unless request by redwood to change
5. Configuration>user>ui>configuration>Job>JobMonitorFields>Filter # usaully just left at default unless request by redwood to change
(note: for registry parameters that are not listed the default is always used)
Redwood always suggests disabling 'App server Load balancing' in your SAP definitions. To do this deselect this option while editing one of your SAP definitions - save and close - then stop and restart the SAP system's CPS process server so the change will take affect.
Can you please explain about this ''App server load balancing '' and where it is located to perform the above mentioned task.
Looking for a hopeful response.
Thanks.
Thanks. I already saw that post Nanda.
Hello Nanda,
Thank you. We have some job definitions with the below parameter/value I tried to search these job definitions with the below parameter/value in the search job definition field, but it didn't display anything. Am I doing some thing wrong?. Please advise.
p:DESTINATIONFILENAME=cashout.txt
p:JOBNAME==BI_I117_IN_ECC_CASHOUT_02
p:ABAP_PROGRAM_NAME=ZBBL_117_INTERFACE_CASHOUT
Regards,
Tinku
Hello All,
I have been trying to define a job type mail in my test environment but I receive this error message.
I don't know because in production environment I don't have this error.
Any help will be greatly appreciated.
Thanks in advance for your kind support.
Regards,
Giovanni
Hello Nanda,
Thank You for your reply. I have been trying to set/change the status but without success.
I have followed these steps
I find this context
Thanks in advance for your kind support.
Regards,
Giovanni
Hi Nanda,
Thank you for your response. Very helpful answer. I also found out that only Webi and Crystal reports can be imported into Cronacle. I found a way around it by duplicating the job SAP_BObj_ReportRun and creating the new job name and adding the report CUID. Then, I was able to schedule the report which was a Publication report.
I will mark this thread as answered.
Best Regards,
Rick
Hi,
We need to move our redwood system from one SID to another SID. We did the homogeneous backup restore method. Could share any document which talks about the specific post system copy procedure in redwood scheduler UI.
Our intention is to clear old data and create new queue entry.
Regards
Pratheep S
Hi Folks,
We have a production CPS system with 5000+ jobs running. We have some performance issue related to Garbage Colelctor Values, below is the system technical information.
We are on Version 8.0
Build Id: M33.92-68971
Database Driver Version: Microsoft SQL Server JDBC Driver 3.0 3.0.1301.101
RAM = 32 GB
Cluster on Server0
Java version=1.6.0_51
OS=Windows Server 2008 R2
Memory info:
totalMemory: 4116054016
freeMemory : 1752108128
maxMemory : 4116054016
-XX:PermSize=1024m
-Xmx4096m
-XX:MaxNewSize=1365m
-XX:MaxPermSize=1024m
-Xms4096m
-XX:NewSize=1365m
-XX:+UseConcMarkSweepGC
-XX:TargetSurvivorRatio=90
-verbose:gc
-XX:+DumpDetailedClassStatisticOnOutOfMemory
-XX:+PrintGCDetails
-XX:MaxErrorQueueLength=200
-XX:+PrintGCTimeStamps
-XX:+DisableExplicitGC
-XX:SoftRefLRUPolicyMSPerMB=1
-XX:-StringInternTableInPermGen
-XX:SurvivorRatio=6
-Xss2m
-XX:+HeapDumpOnOutOfMemoryError
-XX:HeapDumpPath=OOM.hprof
-XX:-TraceClassUnloading
Can you please suggest the recommended changes and values which are need to be adopted for fine tuning the garbage collection in our landscape from above information?
Hi Gerben,
We are giving Below parameters in "In Parameter Mappings" "Text Tab" below.(Even not saving as it is in Text Tab) But Output is not getting in Email as it is.It is getting in one line continuously.PFA. Could you please provide solution..P.F.A.
We need Email output as below:-
Name:2000_FCST_SAT_PIR_DELETE STEP-1
Job:2000_FCST_SAT_PIR_DELETE STEP-1
Program: RM60RR20
Variant: 2000_PIR_DELBK
Triggered from Redwood messaging.
But getting Email output as below:-
Regards,
Sudarshan
Hi Gerben,
We are giving Below parameters in "In Parameter Mappings" "Text Tab" below.(Even not saving as it is in Text Tab) But Output is not getting in Email as it is.It is getting in one line continuously.PFA. Could you please provide solution..P.F.A.
We need exact Email output as below:-
Name:2000_FCST_SAT_PIR_DELETE STEP-1
Job:2000_FCST_SAT_PIR_DELETE STEP-1
Program: RM60RR20
Variant: 2000_PIR_DELBK
Triggered from Redwood messaging.
But getting Email output as below:-
Regards,
Sudarshan
Hi Gerben,
We are giving Below parameters in "In Parameter Mappings" "Text Tab" below.(Even not saving as it is in Text Tab) But Output is not getting in Email as it is.It is getting in one line continuously.PFA. Could you please provide solution..P.F.A.
We need exact Email output as below:-
Name:2000_FCST_SAT_PIR_DELETE STEP-1
Job:2000_FCST_SAT_PIR_DELETE STEP-1
Program: RM60RR20
Variant: 2000_PIR_DELBK
Triggered from Redwood messaging.
But getting Email output as below:-
How to schedule the Powershell script in Redwood?
Please let me know whoever tried this.
I am getting an error like ""ps1 cannot be loaded because the execution of s cripts is disabled on this system. Please see "get-help about_signing" for more details. "
Hi Prasad,
Are you getting any OOM dumps when system is slow
Please chekc recommendations in below notes
723909 - Java VM settings for J2EE 6.40/7.0
http://service.sap.com/sap/support/notes/1289240
1095476 - Information for troubleshooting of the AS Java
Regards,
Murali
Hi All,
In Redwood environment is it possible to find where an Event is used?
More exactly let's suppose to use the same Event in two Chains. Is there the possibility to display the name of the two Chains?
Thanks in advance for your kind support.
Regards,
Giovanni