☰

Encontrar todas as máquinas virtuais com RAM inflada ou trocada no cluster VMware?

tag-icon • tag-icon vmware-esxi memory vmware-vsphere vmware-vcenter
Encontrar todas as máquinas virtuais com RAM inflada ou trocada no cluster VMware?

Se eu olhasse para um cluster ou pool de recursos do vSphere e visse um certo nível de aumento ou troca de memória, como identificar quais VMs são afetadas?Eu sei o que fazer com eles quando os encontrar :)

No exemplo abaixo, existem 65 máquinas virtuais no pool de recursos. Há pouco menos de 32 Gigabytes de RAM inflada. Como posso identificar quais VMs têm uma parte de sua RAM aumentada ou trocada?

insira a descrição da imagem aqui

Responder1

Powercli é ótimo para coisas como esta:

Obter-VM | onde {$_.BalloonedMemory -ne '0'}

Muitos mais exemplos estão aqui:

http://www.getshifting.com/wiki/powerclinotes

Particularmente, dê uma olhada em "VMs com balão e troca" http://www.getshifting.com/wiki/powerclinotes#resource_pools_with_ballooning_and_swapping

Copiado do link acima:

$myCol = @()
foreach($vm in (Get-View -ViewType VirtualMachine | Where-Object `
  {$_.Summary.QuickStats.BalloonedMemory -ne "0"})){
    $Details = "" | Select-Object VM, `
    SwappedMemory ,BalloonedMemory

    $Details.VM = $vm.Name
    $Details.SwappedMemory = $vm.Summary.QuickStats.SwappedMemory
    $Details.BalloonedMemory = $vm.Summary.QuickStats.BalloonedMemory

    $myCol += $Details
  }
$myCol

Responder2

Espero que isso seja um pouco de ajuda.

Desculpe, foi rápido, então você gostaria de fazer classificações adicionais nesta saída de informações, bem como agrupar por recursos (host, armazenamento de dados, cluster), executei em todo o vCenter.

#!/bin/env python
__author__ = 'dladner'

import os
import sys
import pysphere

from pysphere import VIServer, MORTypes
from pysphere.resources import VimService_services as VI

s = VIServer()
s.connect("vcenter_hostname_or_ip", "username", "password")
vmlist = s.get_registered_vms(status="poweredOn")
count = 0
for vmpath in vmlist:
    pm1 = s.get_performance_manager()
    vm = s.get_vm_by_path(vmpath)
    mor = vm._mor
    ballooned = pm1.get_entity_statistic(mor, 77)
    swapped = pm1.get_entity_statistic(mor, 61)
    count += 1
    print "VM Number %d is %s and Ballooned Memory Amount on this VM is : %s" % (count, vmpath, ballooned)
    print "VM Number %d is %s and Swapped Memory Amount on this VM is  : %s" % (count, vmpath, swapped)

print "NOTICE: This statistic refers to VMkernel swapping and not to guest OS swapping."
s.disconnect()

Isso lhe dará o resultado abaixo, onde você já pode modificar e classificá-lo da maneira que desejar, do valor maior em balão/trocado para menor (+ gostaria de converter KB em MB)

Eu modifiquei a saída (nomes de host substituídos e agrupados por balões/trocados)

SWAPPED:

VM Number 2 is [vm11_rr] hostname and Swapped Memory Amount on this VM is  : [<vm-303427:swapped(61):Swapped::107336:kiloBytes:2013-08-31 04:35:33.561569>]
VM Number 9 is [vm04] hostname and Swapped Memory Amount on this VM is  : [<vm-17451:swapped(61):Swapped::7896:kiloBytes:2013-08-31 04:36:25.791696>]
VM Number 10 is [vm05_nq] hostname and Swapped Memory Amount on this VM is  : [<vm-266882:swapped(61):Swapped::816:kiloBytes:2013-08-31 04:36:33.266692>]
VM Number 11 is [vm06] hostname and Swapped Memory Amount on this VM is  : [<vm-4550:swapped(61):Swapped::-1:kiloBytes:2013-08-31 04:36:40.313133>]
VM Number 14 is [vm05] hostname and Swapped Memory Amount on this VM is  : [<vm-158642:swapped(61):Swapped::12532:kiloBytes:2013-08-31 04:37:02.907590>]
VM Number 23 is [vm07] hostname and Swapped Memory Amount on this VM is  : [<vm-437276:swapped(61):Swapped::135272:kiloBytes:2013-08-31 04:38:10.633787>]
VM Number 24 is [vm14_rr] hostname and Swapped Memory Amount on this VM is  : [<vm-74703:swapped(61):Swapped::183436:kiloBytes:2013-08-31 04:38:17.985025>]
VM Number 25 is [vm01] hostname and Swapped Memory Amount on this VM is  : [<vm-156212:swapped(61):Swapped::13268:kiloBytes:2013-08-31 04:38:25.224494>]
VM Number 33 is [vm14] hostname and Swapped Memory Amount on this VM is  : [<vm-156661:swapped(61):Swapped::828:kiloBytes:2013-08-31 04:39:25.827377>]
VM Number 39 is [vm23_rr] hostname and Swapped Memory Amount on this VM is  : [<vm-391487:swapped(61):Swapped::10020:kiloBytes:2013-08-31 04:40:10.795514>]
VM Number 48 is [vm15] hostname and Swapped Memory Amount on this VM is  : [<vm-446333:swapped(61):Swapped::-1:kiloBytes:2013-08-31 04:41:20.224237>]
VM Number 49 is [vm14_rr] hostname and Swapped Memory Amount on this VM is  : [<vm-395852:swapped(61):Swapped::91664:kiloBytes:2013-08-31 04:41:27.588455>]
vm number 51 is [vm03] hostname and swapped memory amount on this vm is  : [<vm-47780:swapped(61):swapped::8:kilobytes:2013-08-31 04:41:43.539276>]
vm number 54 is [vm29_rr] hostname and swapped memory amount on this vm is  : [<vm-430048:swapped(61):swapped::1200:kilobytes:2013-08-31 04:42:07.398927>]
vm number 56 is [vm03] hostname and swapped memory amount on this vm is  : [<vm-45758:swapped(61):swapped::19560:kilobytes:2013-08-31 04:42:22.790720>]
vm number 59 is [vm25_rr] hostname and swapped memory amount on this vm is  : [<vm-267833:swapped(61):swapped::55836:kilobytes:2013-08-31 04:42:46.202327>]
vm number 60 is [vm01] hostname and swapped memory amount on this vm is  : [<vm-279859:swapped(61):swapped::4289160:kilobytes:2013-08-31 04:42:53.200330>]
vm number 61 is [vm04] hostname and swapped memory amount on this vm is  : [<vm-174860:swapped(61):swapped::26288:kilobytes:2013-08-31 04:43:01.326958>]
vm number 62 is [vm25_rr] hostname and swapped memory amount on this vm is  : [<vm-384185:swapped(61):swapped::8180:kilobytes:2013-08-31 04:43:09.043827>]
vm number 63 is [vm12_rr] hostname and swapped memory amount on this vm is  : [<vm-434605:swapped(61):swapped::25780:kilobytes:2013-08-31 04:43:16.270494>]
vm number 64 is [vm01] hostname and swapped memory amount on this vm is  : [<vm-372871:swapped(61):swapped::21840:kilobytes:2013-08-31 04:43:23.820811>]
vm number 67 is [vm09_nq] hostname and swapped memory amount on this vm is  : [<vm-184134:swapped(61):swapped::716:kilobytes:2013-08-31 04:43:46.909753>]
vm number 69 is [vm31_rr] hostname and swapped memory amount on this vm is  : [<vm-434653:swapped(61):swapped::148:kilobytes:2013-08-31 04:44:02.652494>]

BALLOONED::
VM Number 23 is [vm07] hostname and Ballooned Memory Amount on this VM is : [<vm-437276:vmmemctl(77):Balloon::2638188:kiloBytes:2013-08-31 04:38:08.631993>]
VM Number 24 is [vm14_rr] hostname and Ballooned Memory Amount on this VM is : [<vm-74703:vmmemctl(77):Balloon::1335560:kiloBytes:2013-08-31 04:38:16.188130>]
VM Number 49 is [vm14_rr] hostname and Ballooned Memory Amount on this VM is : [<vm-395852:vmmemctl(77):Balloon::1249812:kiloBytes:2013-08-31 04:41:25.789048>]
VM Number 56 is [vm03] hostname and Ballooned Memory Amount on this VM is : [<vm-45758:vmmemctl(77):Balloon::-1:kiloBytes:2013-08-31 04:42:20.445191>]

Se precisar de ajuda adicional, posso tentar torná-lo mais sensato amanhã, pois hoje é bastante tarde.

FONTES:

http://pubs.vmware.com/vsphere-50/index.jsp#com.vmware.wssdk.apiref.doc_50/memory_counters.html

https://code.google.com/p/pysphere/

Espero que isto ajude.

Saúde.

PS Não tenho certeza do que -1:kiloBytes significa na API.

Responder3

Esta é uma adição muito tardia a um thread antigo, mas você pode torná-lo mais rápido mantendo-o em um pipeline e evitando um loop for each. Experimente:

    get-vm | where{ $_.swappedmemory -ne '0'} | get-view | Select Name,@{Name="Swapped”;Expression={$_.summary.quickstats.swappedmemory}},@{Name="Ballooned";Expression={$_.summary.quickstats.balloonedmemory}}

informação relacionada