Custom Query (431 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (91 - 93 of 431)

Ticket Resolution Summary Owner Reporter
#446 fixed IZ2335: load_avg data missing from "qstat -f -xml" Dave Love <d.love@…> craffi
Description

[Imported from gridengine issuezilla http://gridengine.sunsource.net/issues/show_bug.cgi?id=2335]

        Issue #:      2335             Platform:     All      Reporter: craffi (craffi)
       Component:     gridengine          OS:        All
     Subcomponent:    qmaster          Version:      6.1u3       CC:    None defined
        Status:       NEW              Priority:     P3
      Resolution:                     Issue type:    DEFECT
                                   Target milestone: ---
      Assigned to:    rayson (rayson)
      QA Contact:     ernst
          URL:
       * Summary:     load_avg data missing from "qstat -f -xml"
   Status whiteboard:
      Attachments:

     Issue 2335 blocks:
   Votes for issue 2335:  13


   Opened: Tue Jul 24 07:19:00 -0700 2007 
------------------------


Summary: qstat should produce the same output regardless of output format type.
The XML output option ("-xml") currently does not show load_avg data.


Easy to reproduce:

chrisdag-aliased:~ dag$ qstat -f
queuename                      qtype used/tot. load_avg arch          states
----------------------------------------------------------------------------
all.q@chrisdag-aliased         BIP   0/1       0.26     darwin-x86



chrisdag-aliased:~ dag$ qstat -f -xml
<?xml version='1.0'?>
<job_info  xmlns:xsd="http://www.w3.org/2001/XMLSchema">
  <queue_info>
    <Queue-List>
      <name>all.q@chrisdag-aliased</name>
      <qtype>BIP</qtype>
      <slots_used>0</slots_used>
      <slots_total>1</slots_total>
      <arch>darwin-x86</arch>
    </Queue-List>
  </queue_info>
  <job_info>
  </job_info>
</job_info>

   ------- Additional comments from craffi Mon Mar 17 10:48:32 -0700 2008 -------
Is it possible to get a status update or target milestone added to this issue? I
may be biased (!!) but getting qstat output consistent is pretty important.

   ------- Additional comments from rayson Tue Feb 2 20:13:44 -0700 2010 -------
The difference comes from the code:

Normal qstat
============
- clients/qstat/qstat.c

   /* load avg */
   if (!summary->has_load_value) {
      if (summary->has_load_value_from_object) {
         sprintf(to_print, "%2.2f ", summary->load_avg);
      } else {
         sprintf(to_print, "---  ");
      }
   } else {
      sprintf(to_print, "-NA- ");
   }

qstat XML
=========
- clients/qstat/qstat_xml.c

   /* load avg */
   if (summary->has_load_value && summary->has_load_value_from_object) {
      xml_append_Attr_D(attribute_list, "load_avg", summary->load_avg);
   }


For normal qstat output, load_avg is displayed when summary->has_load_value is FALSE and summary->has_load_value_from_object is TRUE .

For qstat XML output, load_avg is displayed when summary->has_load_value is TRUE and summary->has_load_value_from_object is TRUE .

This explains the difference in behavior, and I will read the code and see if the logic for qstat XML makes sense.



   ------- Additional comments from rayson Fri Feb 12 18:58:29 -0700 2010 -------
A simple fix:

Index: clients/qstat/qstat_xml.c
===================================================================
RCS file: /cvs/gridengine/source/clients/qstat/qstat_xml.c,v
retrieving revision 1.43
diff -C 4 -r1.43 qstat_xml.c
*** clients/qstat/qstat_xml.c   4 Sep 2009 11:30:14 -0000       1.43
--- clients/qstat/qstat_xml.c   13 Feb 2010 01:55:51 -0000
***************
*** 1014,1022 ****
     xml_append_Attr_I(attribute_list, "slots_resv", summary->resv_slots);
     xml_append_Attr_I(attribute_list, "slots_total", summary->total_slots);

     /* load avg */
!    if (summary->has_load_value && summary->has_load_value_from_object) {
        xml_append_Attr_D(attribute_list, "load_avg", summary->load_avg);
     }

     /* arch */
--- 1014,1022 ----
     xml_append_Attr_I(attribute_list, "slots_resv", summary->resv_slots);
     xml_append_Attr_I(attribute_list, "slots_total", summary->total_slots);

     /* load avg */
!    if (!summary->has_load_value && summary->has_load_value_from_object) {
        xml_append_Attr_D(attribute_list, "load_avg", summary->load_avg);
     }

     /* arch */

Now qstat -f XML displays the load information:

hccgalaxy:~> /tmp/s/source/LINUXAMD64_26/qstat -f -xml
<?xml version='1.0'?>
<job_info  xmlns:xsd="http://www.w3.org/2001/XMLSchema">
  <queue_info>
    <Queue-List>
      <name>all.q@hccgalaxy</name>
      <qtype>BIP</qtype>
      <slots_used>0</slots_used>
      <slots_resv>0</slots_resv>
      <slots_total>4</slots_total>
      <load_avg>1.03000</load_avg>
      <arch>lx26-amd64</arch>
    </Queue-List>
  </queue_info>
  <job_info>
  </job_info>
</job_info>

While standard qstat misses it:

hccgalaxy:~> qstat -f -xml
<?xml version='1.0'?>
<job_info  xmlns:xsd="http://www.w3.org/2001/XMLSchema">
  <queue_info>
    <Queue-List>
      <name>all.q@hccgalaxy</name>
      <qtype>BIP</qtype>
      <slots_used>0</slots_used>
      <slots_resv>0</slots_resv>
      <slots_total>4</slots_total>
      <arch>lx26-amd64</arch>
    </Queue-List>
  </queue_info>
  <job_info>
  </job_info>
</job_info>
#503 fixed IZ2542: Allow users to limit concurrent array job task execution craffi
Description

[Imported from gridengine issuezilla http://gridengine.sunsource.net/issues/show_bug.cgi?id=2542]

        Issue #:      2542             Platform:     All           Reporter: craffi (craffi)
       Component:     gridengine          OS:        All
     Subcomponent:    scheduling       Version:      current          CC:    None defined
        Status:       STARTED          Priority:     P2
      Resolution:                     Issue type:    ENHANCEMENT
                                   Target milestone: Maintrunk
      Assigned to:    andreas (andreas)
      QA Contact:     andreas
          URL:
       * Summary:     Allow users to limit concurrent array job task execution
   Status whiteboard:
      Attachments:

     Issue 2542 blocks:
   Votes for issue 2542:


   Opened: Sun Apr 6 07:17:00 -0700 2008 
------------------------


Summary:
 - This RFE requests a way for users (who do not have administrative access to
policy and RQS settings) to self-throttle the number of concurrent array tasks
executing at any given time.


In relation to these existing Enhancement issues:
http://gridengine.sunsource.net/issues/show_bug.cgi?id=2147
http://gridengine.sunsource.net/issues/show_bug.cgi?id=1276

The desired behavior is the following:

- Allow a user to limit the number of concurrent executing array tasks

Issue:

- Users have been trained to use array tasks to reduce SGE load
- Users who do this typically submit very large array jobs via qsub
- Users want to throttle concurrent execution of tasks to a fixed value
- This "self throttle" control is desired at the user level (not policy or RQS
level)

A "task" target for resource quotas has already been written up in #2147 - this
RFE asks for an extention to the "-t" switch of qsub:

 Existing array task syntax:

   -t n[-m[:s]]

 Proposed array task syntax:

  -t n[-m[:s]] -tc <INT>

.. where "-tc" is shorthand for "task concurrency". The default value should
place no constraint on executing tasks. If present, <INT> will specify the upper
bound of tasks that are allowed to execute simultaneously.

   ------- Additional comments from andreas Mon Apr 7 00:34:00 -0700 2008 -------
Good RFE!

It should be possible to reuse our existing RQS code infrastructure for it.

I raise it to P2 as this was repeatedly requested by the community. It will
improve SGE resource control significantly.

   ------- Additional comments from andreas Mon Apr 7 00:39:41 -0700 2008 -------
Target milestone obviously must be past 6.2.

   ------- Additional comments from templedf Mon Dec 22 23:43:05 -0700 2008 -------
Expanded platform to all.
#629 fixed IZ2899: bad RQS syntax can crash/hang qmaster process craffi
Description

[Imported from gridengine issuezilla http://gridengine.sunsource.net/issues/show_bug.cgi?id=2899]

        Issue #:      2899             Platform:     All      Reporter: craffi (craffi)
       Component:     gridengine          OS:        All
     Subcomponent:    qmaster          Version:      6.2u1       CC:    None defined
        Status:       NEW              Priority:     P3
      Resolution:                     Issue type:    DEFECT
                                   Target milestone: 6.2u3
      Assigned to:    ernst (ernst)
      QA Contact:     ernst
          URL:
       * Summary:     bad RQS syntax can crash/hang qmaster process
   Status whiteboard:
      Attachments:

     Issue 2899 blocks:
   Votes for issue 2899:


   Opened: Thu Jan 29 20:19:00 -0700 2009 
------------------------


Reproduce:

Make 2 project objects:

name testProject
oticket 0
fshare 0
usage NONE
acl NONE
xacl NONE

name testProject2
oticket 0
fshare 0
usage NONE
acl NONE
xacl NONE

The following BAD RQS syntax will hang or crash Grid Engine.

On Linux/x86_64 I was able to crash the qmaster process, on Mac OS X the system
seems to hang and clients will eventually see this error: "ERROR: failed
receiving gdi request response for mid=4 (got syncron message receive timeout
error)."


This will cause the problem:


{
  name         testRQS
  description  will this crash a qmaster?
  enabled      TRUE
  limit        projects !{testProject,testProject2} queues * hosts * to slots=10
}

   ------- Additional comments from crei Thu Feb 26 02:37:18 -0700 2009 -------
When I try to create the projects in current 62u2 (maintrunk) system I get the
following error message:

error: unknown attribute name "usage"
error: error reading file: "/tmp/1782-VUVTwk"
cant read project

But setting a incorrect resource quota set still can crash qmaster:

qconf -arqs
{
   name         abc
   description  will we crash the qmaster?
   enabled      TRUE
   limit        projects !{testProject,testProject2} queues * hosts * to slots=10
}
1024   7629    worker000     worker000 takes packet from priority queue.
(packet_queue->counter = 0; packet_queue->waiting = 1)
  1025   7629    worker000     GDI ADD resource quota set (host2/qconf/1)
(user1/0815/rad/10)
  1026   7629    worker000     got new resource quota set
  1027   7629    worker000     got new description
  1028   7629    worker000     got new enabled
  1029   7629    worker000     !!!!!!! sge_resolve_host: WARNING call with old
lStringT data type,
  1030   7629    worker000     !!!!!!! this data type should be replaced with
lHostT data type in
  1031   7629    worker000     !!!!!!! the future! Nevertheless, just a warning!
Function works fine!
=> crash !!!

Priority P3 is ok since you have to be manager user to setup invalid rqs!
Note: See TracQuery for help on using queries.