Job Chain Restart Behaviour

If a Job Chain with 3 steps fails in the second step - we would like the restart to resume execution with the second step, bypassing execution of the first step. What are the correct settings to enable restarting the chain with the failed step? Thanks

Hi Noel,
For now, you could put an "on completed" "request restart" on the SendMessage step. If the SendMessage runs, there was an error so you would want to restart, but the restart options have some restrictions then because they don't take the step of the error but the SendMessage step as a reference now (you have the request restart on the send message step).
Last but not least: if you already inform someone, you can also simply right click a step in the monitor and choose restart, this should restart the chain but from the step you selected. This is the same as submitting a new chain but specifying the step where it should start (on the same page in the submit wizard where you specify the queue).
There also is an add-on available from Redwood called "active alerting" which allows you to do rule-based alerting (depending on source of the error send message to different teams) as well as escalation (if no reply within x minutes, send to next team) very easily.
Regards,
Anton.

Similar Messages

Restart/resubmit job chain waiting of a file event

Hi Experts,
how can I restart/resubmit a job chain waiting of a file event?
In my case the variables OriginalPath and FinalPath are used in job chain parameters as an Default Expression (waitEvents...). If I restart the job chain I get the message "Could not evaluate default value for parameter...".
We are using SAP-CPS Build M28.20-37214.
Regards
Mathias

Hi,
You probably want to restart it because there was a failure somewhere in the chain. In this case it is easier to not let the chain end in error, but to let it go to status Console using the "Request Restart" postcondition.
Then when the step is restarted, or the chain is restarted from the beginning, it is restarted within the original chain and it reuses the already determined parameter values.
Regards,
Anton.

Restart in Job chains and managing DST

Hi,
1) Given a 5 step job chain, if step 3 fails how is this handled in CPS? The following steps are chained but can the errored step be fixed and rerun to continue the job chain processing?
2) How is day light saving handled for particular jobs. Is there anywhere this is captured in the detailed job request form solman?
Regards,
Esha

Hi,
1) Given a 5 step job chain, if step 3 fails how is this handled in CPS? The following steps are chained but can the errored step be fixed and rerun to continue the job chain processing?
-- Yes, you can restart at any step of Job chain. (Not not at call level, remember one step might have multiple call inside it)
2) How is day light saving handled for particular jobs. Is there anywhere this is captured in the detailed job request form solman?
-- Use GMT timezone to submit jobs. There is nothing on redwood which can solve day light saving. It alwasy has to be at OS level. But redwood has all timezones availabe and you can submits job in any timezone.
Regards,
Sunil Rohit

How can you get the schedule of a single reoccurring job with Restart ability in Redwood to pause when it has failed...?

We need the Request Restart ability when scheduling a single reoccurring job in Redwood. Most of our jobs are scheduled in CHAINs which offers the Final Status Handler on Step level, and the schedule is "paused" when a job in the CHAIN has failed ( the next instance does not populate until the OP MSG has been Replied to) , and once the OPERATOR message is replied to, the CHAIN can resume running again ( desired ) . But we need this same ability when scheduling a single job not in a CHAIN. In our testing, we set up a job with Request Restart chosen on the Error , Killed and Unknown selections on the Restart Behavior Tab. But we found when a job is scheduled say once a day at 8AM, and when it fails, an OPERATOR msg appears allowing a Restart choice ( desired), but the next day's schedule also populates ( undesired) and we do NOT want the schedule to continue on yet because we need a chance to fix the error before the next insance runs for business reasons. So how can you get the schedule of a single reoccurring job with Restart ability in Redwood to pause when it has failed...?

Hello Fran,
You can wrap the job chain in a master job chain in the first step and set the final status handler there.
You can also set the Restart Behavior to 'Stop Submit Frame', in that case you will have to resubmit the job once you are finished troubleshooting.
Regards Gerben

Error while scheduling the Email Alert JOB chain

Hi All,
I have defined a job chain in CPS and when i am going to schedule it then it is giving me error message.We have taken the trial version.
Please find the log attached below.
11:18:31 PM:
JCS-111004: Queue ETD.sapetd00_Queue has no ProcessServer with the required JobDefinitionType/Service/Resource for Job 932 (submitted from ETD.Z_MONI_BATCH_DP copy from 2009/12/30 18:22:23,113 Australia/Sydney) (submitted from Job Definition ETD.Z_MONI_BATCH_DP (Copy from 2009/12/30 18:22:23,113 Australia/Sydney)): Job Definition Type CSH/Service PlatformAgentService/"Empty"
JCS-102064: Job 934 (submitted from System_Mail_Send copy from 2009/12/29 17:54:16,608 Australia/Sydney) is global but refers (via Job) to an object in an isolation group
JCS-102064: Job 934 (submitted from System_Mail_Send copy from 2009/12/29 17:54:16,608 Australia/Sydney) is global but refers (via Chain Step) to an object in an isolation group
JCS-102064: Job 934 (submitted from System_Mail_Send copy from 2009/12/29 17:54:16,608 Australia/Sydney) is global but refers (via Parent Job) to an object in an isolation group Show error details
Thanks
Rishi Abrol

Hi
Are you logged into the correct isolation group ?
Ensure the process server is also assigned to the queue.
Regards

Job Chaining and Quickcluster

I always get
Status: Failed - HOST [Macintosh.local] QuickTime file not found.
after the first part of the job is successful.
If I just submit with "This Computer" it works fine. Original file is ProRes 422, first job uses ProRes 422 to scale it down to 480x270. Second job compresses to h.264. I found some info on this board from 2008 saying that job chaining and quickclusters don't work together. Is that still how it is? that's really useless..
I also found this from Jan 2009
David M Brewer said:
The reason the second rendering is failing is.......this has happen to me a few times until I figured it out.....make sure you set the dimension to the video in the h.264 settings, set to the same size as the Pro Rez dimensions.
For the most part the dimensions are left blank for the second link, h.264. And don't use 100% of source. Put the physical numbers into the spaces. When you link one video to another, the second codec doesn't know the specs settings you made for the first video settings.
Also make sure you (at least check) set the audio for the second video. I usually have the Pro Res do the audio conversion and just past-through to the second video settings. Again it can happen the the audio is disable in the h.264 settings. This has happen a few time for me........... Check and double check your settings!
He doesn't mention anything about with or w/o Quickclusters, but I tried what he said and could not get it to work with quickclusters...
Anyone got any new info on this?

Studio X,
Thanks for taking the time to run some tests and post your results.
I'm finding the same results with converting ProRes422 to mp4, But...
Other codecs are giving me very different results.
I've run some random tests to try to get a grip on whats happening.
First I was playing around with the # of instances. I've read here and on Barefeats that (at least for my model MacPro) the instances should be set to (# of processors/2), so I've been using 4 for quite a while now and thought I'd test it for myself.
A single 5min ProRes422 1920x1080 29.97 file to h.264
This Computer- 15:28
2 Instances- 14:56
3 Instances- 13:52
4 Instances- 14:48
5 Instances- 13:43
6 Instances- 13:48
7 Instances- 13:58
In this case 5i was the fastest but not using a Quickcluster wasn't far off
A single 2m30s ProRes422 1920x1080 29.97 file to h.264
This Computer- 3:19
2 Instances- 3:45
3 Instances- 3:45
4 Instances- 3:45
5 Instances- 3:50
6 Instances- 4:00
7 Instances- 4:00
Interesting...not using a Quickcluster is fastest
A single 2m30s ProRes422 1920x1080 29.97 file Scaled down using original codec
This Computer- 5:20
4 Instances- 4:10
5 Instances- 4:10
7 Instances- 4:11
A single 1m30s ProRes422 1920x1080 29.97 file to mpeg-2
This Computer- 2:12
5 Instances- 2:10
When Quickclusters are faster, 4-5 instances does seem to be the sweet spot(again for my setup).
In the mpeg-2 test, I should have used a longer clip to get a better test but it was getting late and I was just tring to get an idea of the codecs usage of my resources. I was also monitoring CPU usage with Activity Monitor in all tests.
Now multiclip batches:
I forgot to write down the length of the clips in this first test but it consisted of 8 ProRes 422 clips. 3 about 1m long and the rest between 13s and 30s
8 ProRes 422 clips to mp4
This Computer- 11:25
4 Instances- 5:16
Same results as Studio X
Next tests with 5 clips(total 1m51s)
5 ProRes 422 clips to h.264
This Computer- 5:00
4 Instances- 4:52
5 ProRes 422 clips to mpeg-2
This Computer- 2:55
4 Instances- 3:01
5 ProRes 422 clips to DV NTSC
This Computer- 6:40
4 Instances- 5:12
5 ProRes 422 clips to Photo Jpeg
This Computer- 2:44
4 Instances- 2:46
I re-ran the last test with 7 clips because of the time it took reassemble the segmented clips
7 ProRes 422 clips to Photo Jpeg(total 3m14s)
This Computer- 4:43
4 Instances- 3:41
One last test,
A single ProRes 422 clip to Photo Jpeg(4:05;23)
This Computer- 5:52
4 Instances- 4:10
Let me start off by saying it is clear that there are many factors that effect compression times such as # of clips, length of clips, and codecs, but here are some of the things I noted:
1)Some codecs themselves seem to be "more aware" of the computers resources than others.
When I compress to h.264 w/o a cluster it will use about 80-85% of all resources
When I compress to h.264 with a cluster it will use about 90-95% of all resources
When I compress to PhotoJpeg w/o a cluster it will use about 20-25% of all resources
When I compress to PhotoJpeg with a cluster it will use about 80-85% of all resources
2)The time it takes to reassemble clips can be quite long and could effect overall speed
In the very last test, compressing a single file to photoJpeg using 4 instances took 4m10s. Watching Batch Monitor I noted that it took 2m0s to compress and 2m10s to reassemble.Wow...
It would be interesting to see how the disassemble/reassemble of bigger and larger batches using clusters effect overall time. But that would take some time.
I think the thing I will be taking with me from all of this is your workflow is your own. If you want to optimize it, you should inspect it, test it and adjust it where it needs adjusting. Now if anyone has the time and was to run similar tests with very different results I'd love to know about it...

How to Schedule a Job Chain to start automatically on SAP CPS.

Hi,
I did a job chain and i want to run automatically on sap cps Tuesday thru Saturday at 6:00 a.m., i make a calendar on sap cps with this specific options but the job chain doesn't start running. I don't know if i need to do something more, so if someone can give a little help with this i will apreciate a lot.
Thanks,
Omar

It finished ok but on the operator message i got the following message:
Unable to resubmit this job.
Details:
com.redwood.scheduler.api.exception.TimeWindowExpectedOpenWindowException: CalculateNextClose should only be called on an open time window
at com.redwood.scheduler.model.method.impl.TimeWindowMethodImpl.calculateNextCloseIntersectionInt(TimeWindowMethodImpl.java:388)
at com.redwood.scheduler.model.method.impl.TimeWindowMethodImpl.calculateNextCloseIntersectInt(TimeWindowMethodImpl.java:249)
at com.redwood.scheduler.model.TimeWindowImpl.calculateNextCloseIntersectInt(TimeWindowImpl.java:212)
at com.redwood.scheduler.model.method.impl.SubmitFrameMethodImpl.calculateNextInt(SubmitFrameMethodImpl.java:178)
at com.redwood.scheduler.model.SubmitFrameImpl.calculateNext(SubmitFrameImpl.java:176)
at com.redwood.scheduler.model.listeners.JobStatusChangePrepareListener.resubmitSubmitFrameJob(JobStatusChangePrepareListener.java:763)
at com.redwood.scheduler.model.listeners.JobStatusChangePrepareListener.resubmitJob(JobStatusChangePrepareListener.java:637)
at com.redwood.scheduler.model.listeners.JobStatusChangePrepareListener.processJobToFinalState(JobStatusChangePrepareListener.java:520)
at com.redwood.scheduler.model.listeners.JobStatusChangePrepareListener.modelModified(JobStatusChangePrepareListener.java:233)
at com.redwood.scheduler.persistence.impl.LowLevelPersistenceImpl.informListeners(LowLevelPersistenceImpl.java:728)
at com.redwood.scheduler.persistence.impl.LowLevelPersistenceImpl.writeDirtyObjectListRetry(LowLevelPersistenceImpl.java:207)
at com.redwood.scheduler.persistence.impl.LowLevelPersistenceImpl.access$000(LowLevelPersistenceImpl.java:38)
at com.redwood.scheduler.persistence.impl.LowLevelPersistenceImpl$WriteDirtyObjectListUnitOfWork.execute(LowLevelPersistenceImpl.java:79)
at com.redwood.scheduler.persistence.impl.PersistenceUnitOfWorkManager.execute(PersistenceUnitOfWorkManager.java:34)
at com.redwood.scheduler.persistence.impl.LowLevelPersistenceImpl.writeDirtyObjectList(LowLevelPersistenceImpl.java:102)
at com.redwood.scheduler.cluster.persistence.ClusteredLowLevelPersistence.writeDirtyObjectList(ClusteredLowLevelPersistence.java:59)
at com.redwood.scheduler.model.SchedulerSessionImpl.writeDirtyListLocal(SchedulerSessionImpl.java:648)
at com.redwood.scheduler.model.SchedulerSessionImpl.persist(SchedulerSessionImpl.java:626)
at com.redwood.scheduler.apiint.model.UnitOfWorkManager.perform(UnitOfWorkManager.java:32)
at com.redwood.scheduler.apiint.model.UnitOfWorkManager.perform(UnitOfWorkManager.java:13)
at com.redwood.scheduler.jobchainservice.JobChainService.childJobFinalStatus(JobChainService.java:223)
at com.redwood.scheduler.core.processserver.ProcessServerRuntime.childJobFinalStatus(ProcessServerRuntime.java:836)
at com.redwood.scheduler.core.processserver.ProcessServerRuntime.onMessage(ProcessServerRuntime.java:248)
at com.redwood.scheduler.infrastructure.work.MessageEnabledWork.run(MessageEnabledWork.java:104)
at com.redwood.scheduler.infrastructure.work.WorkerImpl.run(WorkerImpl.java:109)
at java.lang.Thread.run(Thread.java:534)

Can we schedule steps in Job Chain to run at a particular time of the Day.

Hi ,
We have created a Job chain for 3 steps.our requirement is we want to step 1 to run as per the schedule of Job chain but we want step2 to run on fri 2 gmt and step 3 to run on saturday 1 gmt.
is ther any setting in Job chain so that we can schedule subsequent steps to run at a particular time.
Regards
Rajesh

Hi,
You can add a timewindow to the jobdefinitions that you call in step 2 and 3, to restrict the start times for these jobs to the desired time.
Regards,
Anton.

Scheduling a BI job chain in Redwood

The problem I am having is we are trying to schedule a BI job chain via Redwood software and are not getting any response. Within Redwood, I have executed these jobs IMPORT_BW_CHAINS, IMPORT_BW_CHAIN_DEFINITION, IMPORT_BW_INFOPACKAGES using BI job chain 0fcsm_cm_10 which is defined in BI as a job chain. These jobs run to completion but nothing is moved into Redwood to schedule as you would see from a import of a CCMS job. When I run job RUN_BW_CHAIN using the same BI job chain ID I receive the below error. Not sure what Im missing or doing with the process to get to schedule the BI job chains with Redwood.
ORA-06502: PL/SQL: numeric or value error
ORA-06512: at "RSI.RSIEXEC", line 1638
ORA-06512: at "RSI.RSIEXEC", line 1759
ORA-06512: at "RSI.RSI_RUN_BW_CHAIN", line 21
ORA-06512: at "RSI.RSI_RUN_BW_CHAIN", line 80
ORA-06512: at line 1
ORA-06512: at "SYS.DBMS_SYS_SQL", line 1200
ORA-06512: at "SYS.DBMS_SQL", line 323
ORA-06512: at "SYSJCS.DDL", line 1085
ORA-06512: at "SYSJCS.DDL", line 1118
ORA-06512: at "SYSJCS.DDL", line 1177
ORA-06512: at line 3
JCS-00215: in statement RSOJ_EXECUTE_JOB

I am also seeing the same issue
anton the last information you requested
The following products are installed in the Cronacle repository:
Product                                  Version    Status
Cronacle for SAP solutions               7.0.3      Production
Cronacle Forecast Module                 7.0.3      Production
Cronacle Reports Module                  7.0.3      Production
Cronacle &module Module                  7.0.2      development
Cronacle Mail Module                     7.0.3      Production
Cronacle Audit Module                    7.0.2 r2.2 Production
Cronacle Process Manager for Web         7.0.3      Production
Cronacle Module Installer                7.0.3      Production
Cronacle Repository                      7.0.3.34   Production
Cronacle Monitor Module                  7.0.3      Production

Error in redwood job chain for Infopackage

Hi,
We have recently installed redwood for handling sap jobs and are able to run all the job chains with job step in abap program successfully.
However, for APO and BW job chains we have intermediate step for executing BW infopackage where the job is getting failed with the below error:
SAP/BW Error Message: rfc call failed 089: Job BI_BTC<infopackage_name>has not (yet ?) been started
The preceding abap job steps are getting executed successfully. After the infopackage step is failed the consecutive steps all fails.
This problem is common to all job chains with infopackage.
Any help is greatly appreciated.
Regards,
Sandeep.

Hello Anton,
We are facing the same problem: Same log error message.
The infopackage is correctly started and ended in BW.
Here our versions:
Redwood Explorer 7.0.4.2 SP2
BW :   SAP_BASIS 70016
           SAP_BW 70018
Do you think applying SAP CPS SP3 would solve the problem?
Or can we solve it by modifying some specific parameters?
Thanks in advance.
Regards;
Mathieu

Backing up Jobs, Chains and Programs in Oracle Job Scheduler

What is the best way to back up Jobs, Chains and Programs created in the Oracle Job Scheduler via Enterprise Manager - and also the best way to get them from one database to another. I am creating quite a long chain which executes many programs in our test database and wish to back everything up along the way. I will also then need to migrate to the production database.
Thanks for any advice,
Susan

Hi Susan,
Unfortunately there are not too many options.
To backup a job you can use dbms_scheduler.copy_job. I believe EM has a button called "create like" for jobs and programs but I am not sure about chains and this can be used to create backups as well.
A more general purpose solution which should also cover chains is to do a schema-level export using expdp i.e. a dump of an entire schema.
e.g.
SQL> create directory dumpdir as '/tmp';
SQL> grant all on directory dumpdir to public;
# expdp scott/tiger DUMPFILE=scott_backup.dmp directory=dumpdir
You can then import into a SQL text file e.g.
# impdp scott/tiger DIRECTORY=dumpdir DUMPFILE=scott_backup SQLFILE=scott_backup.out
or import into another database (and even another schema) e.g.
# impdp scott/tiger DIRECTORY=dumpdir DUMPFILE=scott_backup
Hope this helps,
Ravi.

SQL Developer 3.1: Job Chains - 2 Bugs.

Hi.
Just having a play with defining job chains in SQL Developer 3.1 and I came across these bugs...
1) When you try to define a new chain using SQL Developer you get, "ORA-01741: illegal zero-length identifier"
If you look on the SQL tab you can see that it is messing up the double quotes, placing two sets of double quotes around the chain_name.
BEGIN
    sys.dbms_scheduler.create_chain(
        comments => '',
        chain_name => '"TEST".""test_chain_3""'
      sys.dbms_scheduler.enable(name=>'"TEST".""test_chain_3""');
END;If you remove the extra quotes and run the code it works fine.
2) The job chain diagram seems to be case sensitive as far as the rule definitions are concerned.
If you follow the chain setup in SQL*Plus described here (http://www.oracle-base.com/articles/10g/SchedulerEnhancements_10gR2.php#job_chains), then check the associated diagram in SQL Developer it is displayed properly.
Now repeat the chain definition from SQL*Plus, but this time define the conditions and actions of the rules in lower case (which is valid and functions as expected).
BEGIN
DBMS_SCHEDULER.define_chain_rule (
    chain_name => 'test_chain_1',
    condition => 'TRUE',
    action     => 'start chain_step_1',
    rule_name => 'chain_rule_1',
    comments   => 'First link in the chain.');
DBMS_SCHEDULER.define_chain_rule (
    chain_name => 'test_chain_1',
    condition => 'chain_step_1 completed',
    action     => 'start chain_step_2',
    rule_name => 'chain_rule_2',
    comments   => 'Second link in the chain.');
DBMS_SCHEDULER.define_chain_rule (
    chain_name => 'test_chain_1',
    condition => 'chain_step_2 completed',
    action     => 'start chain_step_3',
    rule_name => 'chain_rule_3',
    comments   => 'Third link in the chain.');
DBMS_SCHEDULER.define_chain_rule (
    chain_name => 'test_chain_1',
    condition => 'chain_step_3 completed',
    action     => 'END',
    rule_name => 'chain_rule_4',
    comments   => 'End of the chain.');
END;
/Now check the diagram again in SQL Developer and you will see that all the steps are present, but the links (rules) are not correctly displayed. Seems like it is case sensitive...
If you need any more info, please ask.
Cheers
Tim...
Edited by: TimHall on Feb 10, 2012 4:23 PM

I have confirmed your two test cases are indeed bugs and I have raised them. The first seems to be if you have a mixture of _ and lowercase characters in the name but we will look at other cases too. The second is that the parsing of the condition/action does not uppercase unquoted ids. The second can be cirumvented by using quotes etc.

How to setup job chain with time dependant steps

Hi
I need to setup the following job chain with next job step waiting until specified time before starting
Step 1 Job A Time 07:30 if complete go to job step 2
Step 2 Job B start time 09:00 if complete go to job step 3
Step 3 Job C start time 09:30 (end job)
Have tried to use a pre-condition in step 2 and 3 but that just caused thoes steps to be skipped.
can somebody point me in the right direction
Thanks in advance
Jon

Hi Anton/Babu
Thanks for your help, I managed to set this up by embedding multiple Job Chains with their own time window in a master job chain.
As we continue our roll out not sure how scalable the solution will be as you have to create multiple job chains for essentially the same job with the only difference being a specific time window.
I guess only time will tell.
Thanks for you help again guys
Jon

Steps max limit in a Job Chain

Hello experts,
The SAP recommanded that the minimum number of steps in Job Chain for performance purpose.
Can anybody tell , Is there any max limit on the number of steps in a Job chain?
Thanks ,
Suresh Bavisetti

Can you reference the documentation where you see a minimum number of steps in a job chain recommended? I think there is no upper limit for the number of steps in a chain. How many steps are you thinking of adding?
Or are you talking about nesting job chains within other job chains? Pleas eclarify if this is the case.
Rgds,
David Glynn

Submit Multiple Job Definitions/Job Chains with same Time window/Submit frame in mass

Hi,
We have a requirement to submit multiple Job Definition/Job Chains which are part of common Time Window/Submit frame/Queue....
Ex. We have over 50+ different jobs/job chains which will runs Monday to Friday for every 2 hours on same Queue "XXX_Queue". Instead of submitting each job/job chain manually, we would like to know if we could use any script that can achieve our requirement? since we have couple of other jobs which fall under same scenarios...
We are on M33.104 version. Please let me know if any one has any scripts or alternate way of submitting multiple jobs/job chains in mass.
Thanks in advance!
Nidhi.

Hi Nidhish,
Here is some code to set some stuff on a job:
//Get the partition, for global this is not necessary as global is default
Partition part = jcsSession.getPartitionByName("GLOBAL");
//Get the job definition
JobDefinition jobdef=jcsSession.getJobDefinitionByName(part, "System_Info");
//Get the submit frame
SubmitFrame sf = jcsSession.getSubmitFrameByName(part, "SF_Every_Year");
//Get the time window
TimeWindow tw = jcsSession.getTimeWindowByName(part, "System_Week_WorkingHours");
//Set the start time
DateTimeZone dtz = new DateTimeZone(2015, 10, 18, 15, 0, 0, 0);
//Get the Queue
Queue SystemQ=jcsSession.getQueueByName(part, "System");
//Create the Job
Job infoJob=jobdef.prepare();
//Attach queue to job
infoJob.setQueue(SystemQ);
//Attach submit frame, time window, start time
infoJob.setSubmitFrame(sf);
infoJob.setTimeWindow(tw);
infoJob.setRequestedStartTime(dtz);
//Print out the jobid of the job
jcsOut.println(infoJob.getJobId());
//Submit the job
jcsSession.persist();
Regards,
HP

Job Chain Restart Behaviour

Similar Messages

Maybe you are looking for