How to continue after handling chain interruption / sub chain error removed from scheduling

1. How to continue after handling chain interruption

First, there is a table :
RSPCPROCESSLOG At a glance, I know that there is process chain Of log.

There are many in your processing chain DTP, But there is one DTP Something went wrong , Then its state is R 了 .
If this chain has to be executed after the correct running above , Then this chain will be stuck here .
Automatically repair after , The chain will run down by itself .
But if you repair it manually , After that, how to make this chain continue to run ？

The chain is a step One step Running down .
When you deal with the wrong chain DTP Right click on message.
stay Chain You can see it in there Variant and Instance：

Go to the log Look up in the table , Can see TYPE wait .
After manual repair , Run this FM :
RSPC_PROCESS_FINISH

Change the state of this chain to G, Let this chain continue trigger Other chains at the bottom .

To sum up , If you fix one manually DTP, In order to make this chain from this DTP Run down again , Just use the above FM.

2. Subchain removed from scheduling

What's the problem ？
When the chain is executed , There is one backstage job BI_PROCESS_CHAIN Start .
Then I went into this chain . There are many in this chain DTP, The next step is to perform job BI_PROCESS_DTP_LOAD 了 .

You can see ,DTP_OLAD It was not implemented .
Why? ？ The normal process is ：

1. remove from scheduling: I don't know why remove First , But it is a whole process
2. The next step is to start many DTP_LOAD Of job
3. Then is activate and schedule This chain, And the first step to form a complete process . The next time I run, I will still be triggered by the main chain .
   The problem is , Why the bottom DTP_LOAD Of these job Not triggered ？
   Is there no executable process in the background at that time ？
   This is not to say that in the sub chain DTP Execution error , It's because it didn't start at all , Even one log None .
   

The general situation is , This processing chain has not been changed .
But maybe inside DTP Due to the change of conversion , Then it doesn't activate . These are on the watch RSTRAN and RSBKDTP See the corresponding OBJSTAT Just go . Is it right? INA
It may be converted A and M Version inconsistency ？

What should I do at this time ？ Fix inconsistent A and M edition ： SE38 RSTRAN_ROUT_RSFO_CHECK
Activate M And DTP：RSDG_TRFN_ACTIVATE

Then put the problem processing chain remove from schedule, Again reschedule once . This solves the problem .

But if it's not the above problem . I also see an explanation .
An explanation of this long history , I'm skeptical .
But it seems to make some sense , To be verified .

The saying is , Because every backstage JOB There is one. ID. This ID It's a timestamp and two numbers 00 - 99
Only time is counted ID, Dates are not counted .

If you have a processing chain that runs periodically at a fixed time every day , That backstage job It may occupy the same time 00-99
It's full for a long time , Then there is no way to generate new job 了 .
Then you have to let basis Delete the completed background on a regular basis job ID .

I cut a picture and found that it was not what I said , There are Z 了 , There are also characters in the description . That could be 00-00 and AA-ZZ？ There is no time to check , Check it later .
however reschedule It should be solved in a moment . Well .
Let's check ：


Forget it , Not this one ：
Take a look at the following two ：
TBTCO - Job Status Overview Table
TBTCD - Job Log Directory

Why so many ？1296 How did it come from ？
0-9 yes 10 A digital ,A-Z yes 26 Letters . That's it 36*36=1296 ah
That is to say, from 00:05 branch 11 All that starts in seconds ID Have been used up ？？ This seems to be in the past . I instantly feel that what I said is reasonable . I took a look , It's full before and after . How to delete it ？


test I found so many . Tick off test One more time, it will be deleted .

And then it's all right .

Or ,SM37 Delete what you want to delete . Find that job, Select... On the command line job->delete.