fanlongmengjd opened a new issue #3603:
URL: https://github.com/apache/incubator-doris/issues/3603


   **Describe the bug**
   Create routine load job with desired_concurrent_number =5 , running for a 
while , and there is only 1 -2 running tasks. 
   
   **To Reproduce**
   Steps to reproduce the behavior:
   1. Create routine load job with desired_concurrent_number =5 ,
   2.  go to web console and see routine load 
   
   
   **Expected behavior**
    there are 5 running tasks
   
   **Screenshots**
   
   
   **Desktop (please complete the following information):**
   
   
   **Smartphone (please complete the following information):**
    - Version 0.11.33.1
   
   **Additional context**
   we checked the JRF
   
   
[fe.jfr.tar.gz](https://github.com/apache/incubator-doris/files/4636825/fe.jfr.tar.gz)
   
   sounds like DescriptorTable cost most of time and DescriptorTable be 
increased by  the method plan in (
   fe/src/main/java/org/apache/doris/planner/StreamLoadPlanner.java) every time
   `
       // create the plan. the plan's query id and load id are same, using the 
parameter 'loadId'
       public TExecPlanFragmentParams plan(TUniqueId loadId) throws 
UserException {
           // construct tuple descriptor, used for scanNode and dataSink
           TupleDescriptor tupleDesc = 
descTable.createTupleDescriptor("DstTableTuple");
           boolean negative = streamLoadTask.getNegative();
           // here we should be full schema to fill the descriptor table
           for (Column col : destTable.getFullSchema()) {
               SlotDescriptor slotDesc = descTable.addSlotDescriptor(tupleDesc);
               slotDesc.setIsMaterialized(true);
               slotDesc.setColumn(col);
               slotDesc.setIsNullable(col.isAllowNull());
               if (negative && !col.isKey() && col.getAggregationType() != 
AggregateType.SUM) {
                   throw new DdlException("Column is not SUM AggreateType. 
column:" + col.getName());
               }
           }
   `
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to