Opened 12 years ago

Last modified 9 years ago

#1112 new enhancement

IZ171: Memory leak detection with parallel scheduling

Reported by: andreas Owned by:
Priority: normal Milestone:
Component: testsuite Version: current
Severity: Keywords: tests
Cc:

Description

[Imported from gridengine issuezilla http://gridengine.sunsource.net/issues/show_bug.cgi?id=171]

        Issue #:      171             Platform:     All           Reporter: andreas (andreas)
       Component:     testsuite          OS:        All
     Subcomponent:    tests           Version:      current          CC:    None defined
        Status:       NEW             Priority:     P3
      Resolution:                    Issue type:    ENHANCEMENT
                                  Target milestone: milestone 1
      Assigned to:    issues@testsuite
      QA Contact:     joga
          URL:
       * Summary:     Memory leak detection with parallel scheduling
   Status whiteboard:
      Attachments:

     Issue 171 blocks:
   Votes for issue 171:     Vote for this issue


   Opened: Fri Jan 18 07:56:00 -0700 2008 
------------------------


DESCRIPTION:
Issue

http://gridengine.sunsource.net/issues/show_bug.cgi?id=2187

has shown there is a need to improve memory leak detection with parallel
scheduling. E.g. in 6.1u3 a memory leak can be detected when scheduler is run
under dbx control with

   # check -leaks -frames 10 -match 10
   # ignore INT

dbx reports after qconf -ks:

       182      14      -      _strdup < lCreateListHash < lCopyListHash <
parallel_maximize_slots_pe < sge_select_parallel_environment <
select_assign_debit < dispatch_jobs < scheduler <
event_handler_default_scheduler < main
       168      14      -      _strdup < dup_func_string < sge_htable_store <
cull_hash_create_hashtables < lCopyListHash < parallel_maximize_slots_pe <
sge_select_parallel_environment < select_assign_debit < dispatch_jobs < scheduler

when a job like

> qsub -pe 'pe*' 5 -b y sleep 10

is submitted into a cluster consisting of

> qconf -sp pe1
pe_name           pe1
slots             4
user_lists        NONE
xuser_lists       NONE
start_proc_args   /bin/true
stop_proc_args    /bin/true
allocation_rule   $pe_slots
control_slaves    FALSE
job_is_first_task TRUE
urgency_slots     min
> qconf -sp pe2
pe_name           pe2
slots             4
user_lists        NONE
xuser_lists       NONE
start_proc_args   /bin/true
stop_proc_args    /bin/true
allocation_rule   $pe_slots
control_slaves    FALSE
job_is_first_task TRUE
urgency_slots     min

> qconf -sq all.q |grep pe_list
pe_list               mytestpe make pe1 pe2

Change History (0)

Note: See TracTickets for help on using tickets.