IZ1759: scheduler ignores load_thresholds with pe jobs

[sgrell]

[Imported from gridengine issuezilla ​http://gridengine.sunsource.net/issues/show_bug.cgi?id=1759]

        Issue #:      1759             Platform:     All      Reporter: sgrell (sgrell)
       Component:     gridengine          OS:        All
     Subcomponent:    scheduling       Version:      6.0         CC:    None defined
        Status:       NEW              Priority:     P3
      Resolution:                     Issue type:    DEFECT
                                   Target milestone: ---
      Assigned to:    sgrell (sgrell)
      QA Contact:     andreas
          URL:
       * Summary:     scheduler ignores load_thresholds with pe jobs
   Status whiteboard:
      Attachments:

     Issue 1759 blocks:
   Votes for issue 1759:


   Opened: Thu Aug 18 07:51:00 -0700 2005 
------------------------


Hello,

in case of multiple queues on the same host with different load_threshold
setings. In this case the scheduler computes the max slots, which would set all
queues into alarm state.
The scheduler should evaluate the load threshold for each queue individiualy.

scheduling info:            queue instance "test1.q@scrabe.workgroup" dropped
because it is overloaded: np_load_avg=5.070000 (= 0.100000 + 1.00 * 4.970000
with nproc=1) >= 5.0
                            queue instance "test2.q@scrabe.workgroup" dropped
because it is overloaded: np_load_avg=5.070000 (= 0.100000 + 1.00 * 4.970000
with nproc=1) >= 1.0


Stephan