VM is ballooning on host with 141 GB of free memory, why?

Similar Questions

• Target ball and ball over Active on the host with a lot of free memory

  Just at the moment where I think I understand the management of the ESX memory, something confuses me once again.  My understanding of the balloon pilot, is that when an ESX host has need of more memory to be allocated to other virtual machines as the VMKernel steals memory back from the virtual computer running.  However, I have a virtual machine which seems to indicate that the target of ball is well above what is memory 'Active', and the host ESX virtual machine running on a 6 GB of unused memory.  I think that this might have led this virtual machine to go into a State of forgetfulness once already, because I found it does not, only to discover that she used only 30% of CPU and 2 GB Active out of 6 GB of memory allocated, but making a ton of swap memory.  What gives?

  Why in the world of the balloon pilot would target memory on a VM running busy, when there is a lot of free memory available on the host, and why would be a beginning of comments Exchange when it uses only a fraction of the memory allocated to it and the host is sitting on the free memory pletty?  None of my other virtual machines have memory usage charts that look like this.

  This can happen even in the absence of resource pools. Check your settings of VM, "resources" tab and make sure that the unlimited checkbox that she checked on the memory and CPU...

• V21 Firefox cannot find the files if file *.htm used "style windows" path slashes to the java scripts (on Unix, hosting with apache)

  Hello. I find a strange behavior of Firefox v21. If the 'windows' path used *.htm file slashes to the java scripts (on Unix, hosting with apache). Firefox says 404, but other browsers (such as Opera, Chrome, IE10) works properly with this.

  You should use slashes in links so that they work properly in Firefox, otherwise Firefox will escape them that 5% C as you can see if you would them hover in the case of a real link (this is obviously not possible in this case).

• syslogging to 2 host with different levels of severity?

  I have a PIX that sends the logs to a host with Cisco syslog installed. It is configured to send messages of level 3 (as required by the internal standards) of gravity.

  I have another host which collects the theses papers too, but I want some info - level 5 or 6.

  ? is it possible to config the PIX to send logs on 2 different hosts with different levels of severity?

  you configure the pix to send to 7 on a syslog server. configure you this server to send events to the level 3 to another server. I think that it should be possible

• How to restore backup rman to the new host with the structure of the various files on windows?

  Here are the steps that I follow...

  1 oradim create new instance

  2 startup nomount pfile = 'C:\inittest.ora'

  3. restore conrolfile of 'c:\name.bak '.

  I found this file control point to the oradata folder as directed by the original host

  4. Alter database mount

  Mounted DB.

  5 restore database completed with error

  switch to the previous backup

  RMAN-00571: ==================================================

  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.

  RMAN-00571: ==================================================

  RMAN-03002: failure of the restore command at the 12:17:55 08/12/2014

  RMAN-06026: some targets not found - abandonment of restoration

  RMAN-06023: no backup or copy of the file 4 found to restore

  RMAN-06023: no backup or copy of datafile 3 found to restore

  RMAN-06023: no backup or copy of datafile 2 found to restore

  RMAN-06023: no backup or copy of datafile 1 found to restore

  So, it is possible to restore and recover the rman backup to the new host with the structure of different file...?

  Hello

  Try to rename the log file to roll forward online to the desired path on the destination server, and then open with resetlogs.

  SQL > alter database file rename ' G:\APP\ORADATA\HOOP\REDO01. JOURNAL "to"";

  Kind regards

  Shivananda

• count the number of targets, devices and paths by hba for each host with powercli 5.5

  Hi all

  I'm writing this Question again in the community, was not able to found the answer I was looking for in the nets:

  https://communities.VMware.com/thread/516226?start=0 & tstart = 0

  https://communities.VMware.com/thread/293531

  I went through the scripts provided in the community, but seems that t not work on powercli 5.5.

  ///

  # The target account, devices and paths for each host

  Get-Cluster $cluster | Get-VMHost | Sort-Object-property name. {ForEach-Object

  $VMHost = $_

  $VMHost | Get-VMHostHba-type FibreChannel | Sort-Object-property device | {ForEach-Object

  $VMHostHba = $_

  $ScsiLun = $VMHostHba | Get-ScsiLun

  If {($ScsiLun)

  $ScsiLunPath = $ScsiLun | Get-ScsiLunPath | `

  Where-Object {$_.} Name - like "$($VMHostHba.Device) *"} ".

  $Targets = ($ScsiLunPath |) »

  Group-object - property SanID | Measure - Object). County

  $Devices = ($ScsiLun |) Measure - Object). County

  $Paths = ($ScsiLunPath |) Measure - Object). County

  }

  Else {}

  $Targets = 0

  $Devices = 0

  $Paths = 0

  }

  $Report = "" | Select-Object - property VMHost, HBA, target devices, paths

  $Report.VMHost = $VMHost.Name

  $Report.HBA = $VMHostHba.Device

  $Report.Targets = $Targets

  $Report.Devices = $Devices

  $Report.Paths = $Paths

  $Report

  }

  }

  ///

  I went through the script LucD posted below: but it's not exactly what I'm looking for.

  LucD : can you please change the same for me please.   to count the number of paths per hba for each host with powercli 5.5, devices and targets.

  //

  $esx = get-VMHost < host name >

  foreach ($hba to (VMHostHba Get - VMHost $esx - type "FibreChannel")) {}

  $target = ((get - see $hba. VMhost). Config.StorageDevice.ScsiTopology.Adapter | where {$_.} Adapter - eq $hba. Key}). Goal

  $luns = get-ScsiLun - Hba $hba - LunType 'disk '.

  $nrPaths = ($target | % {$_.}) Lun.Count} | Measure - Object - sum). Sum

  Write-Host $hba. Device ' target: ' $target. County "devices:" $luns. County ' path: ' $nrPaths

  }

  //

  I'll be grateful for any help.

  Tarun Gupta

  Try something like this

  {foreach ($esx in Get-VMHost)

  foreach ($hba to (VMHostHba Get - VMHost $esx - type "FibreChannel")) {}

  $target = ((get - see $hba. VMhost). Config.StorageDevice.ScsiTopology.Adapter | where {$_.} Adapter - eq $hba. Key}). Goal

  $luns = get-ScsiLun - Hba $hba - LunType "disk" - ErrorAction SilentlyContinue

  $nrPaths = ($target | % {$_.}) Lun.Count} | Measure - Object - sum). Sum

  $props [ordered] = @ {}

  VMHost = $esx.name

  HBA = $hba. Name

  Target = $target. County

  Device = $luns. County

  Path = $nrPaths

  }

  New-object PSObject-property $props

  }

  }

• vMotion between two hosts with different version?

  
  Hello

  Can I vMotion VM between two hosts with different version? For example, a virtual machine of esxi 6.0 5.5 esxi and esxi esxi 6.0 5.5 vMotion.

  Thank you.

  Yes, it will avoid you to move.

• vMotion (without shared storage) between two hosts with different vSwitches standard s VM

  I can VM vMotion (without shared storage) between two hosts with different vSwitches standard? The vSwitches on these two hosts have different names and number of NIC, but have the same name of portgroup.

  
  Thank you.

  Yes, you can in a Storage VMotion to one Esxi to another with no shared storage...

  It is important the portgroup bears the same name... This is the reference for the VM...

• Best way to migrate virtual machines to new ESXi hosts with a different configuration of networking.

  We currently have a cluster of 6 Dell machines that move VM 120 or so on ESXi 5.5. Networking consists of a dVswitch with 6 uplinks that map to 6 x 1gig physical NETWORK adapters on each host ESXi... These are grouped into 3 trunks / gal, so we have 2 GB by SHIFTING of the bandwidth and failover. The LAGS are configured as follows:

  LAG1

  Traffic to our (NFS) shared data store and management of ESXi. (vmk0)

  LAG2

  Access to the network of the VM - groups of different ports on different VLANS

  LAG3

  vMotion (vmk1)

  We just bought 6 new Dell machines that have connections fiber 2 x 10gig each. The idea is to bring together in a 20Ggig LAG by machine that will handle all of the traffic, so traffic NFS data store, network traffic from the VM, vMotion and traffic management ESXi.

  
  

  What is the best way to migrate the virtual machine in the cluster existing to the new machines? Many of the virtual machine are essential, so we cannot stop time.

  
  

  Can I create the new DELAY on existing dVswitch and put it as an auxiliary card? I'll be able to do it for all 3 of the above both above LAGS?

  
  

  Thanks for the info Alistar, it's almost here. The bit that was me stumping was that after establishing a new dVswitch I received messages that the new dvSwitch was not reachable on the old host, when I tried to migrate the virtual computer.

  The solution (with the permission of VMware Support) was that I need to add the two dVswitches to one of the old hosts, then assign a physical alternative on this host uplink to a temporary 3 uplink on the new dVswitch. I then tag all the relevant VLAN that uplink.

  This way I can first migrate the virtual machine to the new dVswitch networking so I can move the computer to the new cluster.

  Obviously I can only migrate from old to new cluster with a single host. each virtual machine must be a double for the host with 2 switches, and then to the new cluster. I also had to have the two dVswitches to the same version.

  Fortunately, I had a physical rising spare on the hosts of the old, otherwise I would have had to break one of the trunks and re-use of one of these uplinks.

  Oh and be aware, impossible to use the same names for groups of ports 2 DV switches, they must be unique!

  Nick

• Fixed host with Virtual SAN - host of recovery objective chess

  Hello community,

  I have a question about the failures of the host with Virtual SAN.

  We have 6 x ESXi hosts with (a 1 SSD and 7 hard drives) virtual SAN enabled. It is therefore possible to lost 2 ESXi hosts without a failure. Correct?

  Is it possible to define a rule based on the host that defines who is ESXi host target failure?

  
  

  For example:

  When host 1/6 breaks down, host 4/6 should start the VMs.
  

  When the 2/6 host goes down, host 5/6 should start the VMs.

  When the 4/6 host goes down, host 1/6 should start the VMs.

  When the 5/6 host goes down, host 2/6 should start the VMs.

  Thanks a lot for your help.

  Best regards from the Germany!

  Yes & No :-)

  To lose 2 ESXi hosts and continue to run all your virtual machines, you must define the number of failures to tolerate = 2 VM storage strategy and set this policy to the virtual machines.

  There is no way to create areas of the fault to VSAN currently however. Then VSAN will determine where them reconstructed objects when the failure of a host - this isn't something you can order now.

  HTH

  Cormac

• Update / fixes ESXi single host with VCSA running on it.

  We have just updated since the free version of ESXi 5 by ordering the Kit of Essentials. In the process, of course, we were able to deploy vCenter Update Manager (which has been the main motivation, like patching manually the free version seemed to be quite a pain). We have a Windows host that has installed the Update Manager, and we have a VCSA running on the single instance of ESXi, we deployed to date. We were able to install the plugin on the viClient and scan the ESXi host for patches with success (a need a good number). But, we're stuck trying to install the patches... us when trying to restore the host, we get: "the operation is not supported on the selected inventory items. Check events for the objects selected for the operation.

  The host says:

  The 10.0.1.80 host has a virtual VMware vCenter machine

  Device server with VMware vSphere update

  Manager or VMware vCenter Server installed. The

  Virtual MACHINE needs to be moved to another host for the

  consolidation to proceed.

  error

  28/08/2014-16:21:32

  Clean up the entity

  10.0.1.80

  root

  Looking at this logic, for the host switch to maintenance mode, the VCSA should have been suspended (apparently, actually moved), which would avoid in turn patches / updates to the deployment. Am I correct in thinking, that we have a second host and networked storage? Made a Kit Essentials / license provide for the possibility of vMotion from one host to another? It wasn't really clear... Yes, the web site says "vMotion" is included in most Essentials, but find it me hard to imagine any point to a facility in ESXi stream without the ability to vMotion between one system and another guest...

  Welcome to the community,

  And since you have only an ESXi host and your vCSA is running on this host, you cannot patch this host with Update Manager for reason that already know you: the host should be the maintenance mode.

  Add about another host, with Kit Essentials without vMotion functionality, you can try using a data store of shared storage between two hosts, stop the vCSA on first host and switch the vCSA on the second host... and finally the first host patch.

• Purple screen on HOST with Esxi 5.5

  Good afternoon

  I'm upgrading from 5.1 to 5.5 esxi servers.

  I've updated 5 servers without problems but the sixth, I had a problem.

  This morning when I've updated a host everything worked fine. Suddenly, an hour later, host was unreachable and when I checked in the server room

  I got a purple screen. I manually restarted the host, and everything seems to work.

  In my research, one user has this kind of problem, and it's cyclical. I'm doing research in this type of case a way to fix this bug.

  I took a quick glance at 2 positions:

  VMware KB: ESXi hosts with larger workloads can generate partial core dumps

  VMware communities: All content-all communities

  Is that someone already had this kind of screen? Understand what the purple screen?

  Kind regards

  Definitely check your versions of firmware on your server, BIOS, NETWORK card, card Raid array.  What type of server is it?

• Can we do vmotion between hosts with load balancing in network cards grouping different strategy.

  Hello

  We are implementin new host in our Infra and do some vmotions between different groups. A group a host with balancing 'route based on IP hash. " Can we do Vmotion to another cluster where the hosts have different load balancing policies. In addition, if we change the policy on new hosts in furture of load balancing is loss of downtime or a package?

  Kind regards

  Vikram Kumar

  First of all, no problem from VMs one ESX to another, they use different political Multipathing.

  Second, if you use IP hash, it means you are using etherchannel (or LACP) on the physical layer - configuration/cancellation of the configuration that almost always require that the uplinks are not used by any VMs, so you have time to synchronize the physical and virtual configuration. It is not a time out, but you need to plan accordingly - all depends on your physical network capacity.

• VMotion between 2 Hosts with different processors

  Hallo,

  I have a server in my homelab I use for studies of VMsphere and I want to get a second machine to test advanced features such as HA, Vmotion etc.

  My first server has a map of X9SCM-9 with processor Xeon E3-1230V2 (Ivy Bridge). I am currently running ESX 5.0 U2 on it.

  I was counting on obtaining of material quality for the second host consumers because I don't want to spend a lot on it, so I thought that I need any CPU generation Ivy Bridge.

  I tested today with a second machine with a Pentium G2020 Ivy Bridge, but when I try to Vmotion he says "the virtual machine requires physical features that are not available or disabled on the host... ».

  So I figure there must be a difference in architecture regardless of the CPU beeing an Ivy Bridge CPU.

  When I try to activate CVS it just me are going to the 'Nehalem' Version so my virtual machines running on host 1 would not be able to use all the features of the CPU since the Xeon.

  So the question boils down to: what is the processor less which offer compatibility VCA L5?

  Or is there something else to consider in VMotion between two hosts with different processors?

  I got a shared storage, etc. of the same network configuration.

  Thank you

  Benjamin

  Hi Benjamin,

  For compatibility VCA L5 want to look at one of the following models, Intel CPU:

  • Ivy Bridge: Series E3-1200-v2
  • HASWELL

  I would recommend mental health that before rushing and a CPU purchase, check if a cheaper option would be, as Jon mentions in suppresss some fo the CPU compatibility checks so obviously, this is not desirable in a production environment, and must be applied in your study of VMware Lab to the home address.

  Hope this helps,

  Simon

  TechHead

  http://techhead.co

• Stop reboot request when fast vMotioning to a host with a different processor

  I need help to find a better way to deal with a prompt to reboot my users get due to a vDesktop in model (Dell 710) host vMotioned to another (Dell 720). The difference between the two is the processor models. The 710 is a Xeon X 5670 and the 720 is a Xeon E5-2667. The two processors use the same .inf (cpu.inf) file and apparently the drivers. Simply, it triggers the installation process of the material causing the Plug and Play Manager to require a restart of the operating system after the installation. The result, I'm wanting just needs that this reboot window displays don't not weither would it by slaughtering the hardware installation process, hide the window of dialogue itself, or any other possible way to configure the system which does not show users. I've confirmed that you don't have to restart the machine remain operational. Here is some additional information on the operating system and all.

  Windows 7 32 bit

  1 vCPU

  2G of RAM

  vSphere 5.0.0 623860

  Discover Constable's 5.0.0 481677

  Any suggestions or thoughts are greatly appreciated.

  Tanner

  I encountered the same problem and the only solution I've found so far is to start the master on both hosts.  When he started on the two hosts, the message no longer appears when moving the virtual machine back to a host with a different model of CPU.  I have two clusters currently, so I have not tested vMotioning fast machine between hosts with different CPU while it is running.  When I had the chance to test, I'll post my results.

  -Chris