[one-users] Help need understanding the scheduler

[Ruben S. Montero]

FYI, this should be working now

http://dev.opennebula.org/issues/172

On Mon, Nov 2, 2009 at 7:24 PM, Shi Jin <jinzishuai at gmail.com> wrote:
> Thank you.
>
>> So in this case. I do not really understand why host 2 with priority 1
>> is after host 1 with priority 0.98. It seems it is a bug.
>> Could you send as the RANK expression for VM 138 and the outout of
>> onehost show for hosts 0, 1 and 2. So we can reproduce this error.
>
> For VM 138, I used
> RANK=-FREECPU
> The output of onehost show for the 3 hosts are:
> oneadmin at onefrontend64:~$ onehost show 0
> HOST 0 INFORMATION
> ID                    : 0
> NAME                  : node1
> STATE                 : MONITORED
> IM_MAD                : im_kvm
> VM_MAD                : vmm_kvm
> TM_MAD                : tm_ssh
>
> HOST SHARES
> MAX MEM               : 33005972
> USED MEM (REAL)       : 32605476
> USED MEM (ALLOCATED)  : 7864320
> MAX CPU               : 1600
> USED CPU (REAL)       : 86
> USED CPU (ALLOCATED)  : 375
> RUNNING VMS           : 8
>
> MONITORING INFORMATION
> ARCH=x86_64
> CPUSPEED=1596
> FREECPU=1513.6
> FREEMEMORY=22476240
> HOSTNAME=node1
> HYPERVISOR=kvm
> MODELNAME=Intel(R) Xeon(R) CPU           X5570  @ 2.93GHz
> NETRX=5421702
> NETTX=492
> TOTALCPU=1600
> TOTALMEMORY=33005972
> USEDCPU=86.4000000000001
> USEDMEMORY=32605476
> oneadmin at onefrontend64:~$ onehost show 1
> HOST 1 INFORMATION
> ID                    : 1
> NAME                  : node2
> STATE                 : MONITORED
> IM_MAD                : im_kvm
> VM_MAD                : vmm_kvm
> TM_MAD                : tm_ssh
>
> HOST SHARES
> MAX MEM               : 33005972
> USED MEM (REAL)       : 14841036
> USED MEM (ALLOCATED)  : 3145728
> MAX CPU               : 1600
> USED CPU (REAL)       : 44
> USED CPU (ALLOCATED)  : 150
> RUNNING VMS           : 4
>
> MONITORING INFORMATION
> ARCH=x86_64
> CPUSPEED=1596
> FREECPU=1555.2
> FREEMEMORY=28004332
> HOSTNAME=node2
> HYPERVISOR=kvm
> MODELNAME=Intel(R) Xeon(R) CPU           X5570  @ 2.93GHz
> NETRX=135006148
> NETTX=492
> TOTALCPU=1600
> TOTALMEMORY=33005972
> USEDCPU=44.8
> USEDMEMORY=14841036
> oneadmin at onefrontend64:~$ onehost show 2
> HOST 2 INFORMATION
> ID                    : 2
> NAME                  : node3
> STATE                 : MONITORED
> IM_MAD                : im_kvm
> VM_MAD                : vmm_kvm
> TM_MAD                : tm_ssh
>
> HOST SHARES
> MAX MEM               : 33005964
> USED MEM (REAL)       : 18188228
> USED MEM (ALLOCATED)  : 1048576
> MAX CPU               : 1600
> USED CPU (REAL)       : 19
> USED CPU (ALLOCATED)  : 50
> RUNNING VMS           : 1
>
> MONITORING INFORMATION
> ARCH=x86_64
> CPUSPEED=1596
> FREECPU=1580.8
> FREEMEMORY=29569136
> HOSTNAME=node3
> HYPERVISOR=kvm
> MODELNAME=Intel(R) Xeon(R) CPU           X5570  @ 2.93GHz
> NETRX=136062812
> NETTX=492
> TOTALCPU=1600
> TOTALMEMORY=33005964
> USEDCPU=19.2
> USEDMEMORY=18188228
>
> Also, in the table from the log file
> Virtual Machine: 138
>       0.983083        1
>       1       2
>       0.946115        0
> Are they already sorted by the priority? Maybe there is something
> wrong with PRI=1.
> At the time of deploying VM 138, there was no VM running on HID=2.
>
> Please let me know if you need more information to debug this.
> Thanks a lot.
>
> --
> Shi Jin, Ph.D.
>



-- 
+---------------------------------------------------------------+
 Dr. Ruben Santiago Montero
 Associate Professor
 Distributed System Architecture Group (http://dsa-research.org)

 URL:    http://dsa-research.org/doku.php?id=people:ruben
 Weblog: http://blog.dsa-research.org/?author=7

 GridWay, http://www.gridway.org
 OpenNebula, http://www.opennebula.org
+---------------------------------------------------------------+