VM HA (VM monitoring - response to the failure of data store)

Similar Questions

• How to monitor the growth of data store

  I want to see what others are using to monitor the growth of data store. Everyone prefers Foglight HP SIM, Plugins and custom scripts?

  Mike

  Hello

  I played with them all, it's 6 of these, 6 of those. Nagios also has a plugin as well. All you need is something that will send/page when there is a problem. This is crucial.

  Best regards
  Edward L. Haletky
  VMware communities user moderator
  ====
  Author of the book ' VMWare ESX Server in the enterprise: planning and securing virtualization servers, Copyright 2008 Pearson Education.
  Blue gears and SearchVMware Pro items - top of page links of security virtualization - Security Virtualization Round Table Podcast

• Vcenter/VSphere/ESX 5.0.0 question: could not download /opt/vmware/www/build/Tenacity.iso for the san-core8 data store. == > Build fails

  I built this different ways now, and I still get this same exact error. I spin as 'Admin' and I have so any available privilege.

  There are no special characters in the password, I tried this file generation and he got away on my computer than this (v-soft bootable with interface).

  Please help me get to the next step, I'm stuck here.

  Everything seems fine until:

  09/10/2013 18:00:25 [info] validation available disk space...

  09/10/2013 18:00:25 [info] prepares the operating system installation files.

  09/10/2013 18:00:25 [info] looking for /opt/vmware/www/ISV/ISO/CentOS-6.4-x86_64-bin-DVD1.iso

  09/10/2013 18:00:25 [info] Validating md5 checksum for /opt/vmware/www/ISV/ISO/CentOS-6.4-x86_64-bin-DVD1.iso (may take a few minutes)

  09/10/2013 18:01:15 [info] verification of repositories of request for the installation of the packages packet...

  09/10/2013 18:01:16 [info] creating package repositories...

  09/10/2013 18:01:16 [info] create the virtual machine model.

  09/10/2013 18:01:17 [info] VM carrying from the Studio to the host configuration. This may take a few minutes...

  09/10/2013 18:02:20 [warn] cannot download /opt/vmware/www/build/Tenacity.1/Tenacity.1_provstart.iso for the s-core9 data store.

  09/10/2013 18:02:20 [warn]

  %3 %7% 11% 15% 19% 23% 26% 30% 34% 38% 42 failure: (22, ' the requested URL returned error: 401')

  09/10/2013 18:02:20 [error] met a unrecoverable build error!

  I am able to use this model for Centos on my computer actually install and build a paralytic with centos on that - unfortunately I do a

  hacks to make it work properly and I cannot get it to install other real applications (perhaps there is a new version for workstation 10?)

  So I went to vsphere, and I KNOW my generation model is ok, at least to get the installed operating system.

  What information can I give you or what can I try, thanks!

  Is it really "sort of" loading ie 42% or which is a total failure?

  OK, I have a build that WORKED, now I see what you mean about watching build/deploy the VM guest.

  Here is the solution to my problem: our vcenter uses names with spaces, we also use the domain names for our connections. In addition, each admin password was 'special characters', that I've debugged meticulously by ALL of them and much more. Basically, it did not matter that I have "URL" code things, used new tools or edited different scripts to encode things one way or another.

  What finally gave us successes was to go to a 'clean' Vcenter implementation that

  (a) use/did not require a domain name (not / credentials)

  (b) all had the "BACK original' legal names, IE no spaces, no special characters, etc..

  (c) the passwords contained NO "special characters".

  (because I changed this way, 'URL' encoding didn't work, tried it)

  Moral of the story: building a new, simply configured Vcenter or get one and use it.

  (Note: I even downloaded a new ovftool and tried every way that I could but it did not success)

• How can I get the list of data stores in a cluster data store?

  How can I get the list of data stores in a cluster data store? I mean the command line option.

  Hello

  by command line, you mean PowerCLI?

  If so, you can display data with this warehouses:

  Get-DatastoreCluster-name DSClustername | Get-Datastore

  Tim

  Edit: Moved the thread to the PowerCLI community

• Download VMDK files from the browser to the Web-based data store

  Hello

  I want to download the virtual computer VMDK file using the browser of the Web-based data store. I know how to do, but there is something I don't understand.

  I have configured my VM with a provisioning. From the VI client, I see that the size of the vm.vmdk is 4 GB.  The size of proviosned is 20 GB.

  Technically speaking, if I download the VMDK file, it should be 4 GB, however, from the browser on the vm - flat hard size Web-based data store is 20 GB... What is the standard behavior? Is there a way to download only the actually used 4 GB?

  Hello.

  This is the expected behavior, and as much as I know there is no work around if you use the browser to store data.

  Good luck!

• Search for the text value data store

  Hello, I have a PDO of data to store, and I would like to search for a text value in a quantity of measure.  In each of my test series, there is a quantity of measure with the name TCNAME and stores the cell name of test in the string format.  I would like to get all the TCNAMEs that contain "TC4" for example.  Currently, I have to load TCNAME in the portal and display it in an array of string.

  Any ideas how I can do this?

  I assume you are using a PUMA testbed that stores this kind of goods on a unique value chain chains.

  There is no direct search method, so I guess that a script is the only solution we can offer.

  Datafinder you can search for the minimum and maximum of a channel. That would be too much work in the store of ASAM.

  "PDO / cpp"
  
  protected tcnameMQs: set tcnameMQs = store. GetElementList ("MeasurementQuantity", "name = TCNAME", true)
  
  protected tcnameMQ: for each tcnameMQ in tcnameMQs
  
    MsgBox tcnameMq.References ("local_columns"). Elements (1). Properties ("values"). Value
  
  Next
  
  ' AOP5 / Corba
  
  protected tcnameLCs: set tcnameLCs = store. GetElementList ("LocalColumn", "measurement_quantity.name = TCNAME", true)
  
  protected tcnameLC: for each tcnameLC in tcnameLCs
  
    MsgBox tcnameLC.Properties ("values"). Value
  
  Next
  
  

  I have added the example which should be the shortest path from my point of view. In AOP5, you can search LocalColumn and spend BA which is less overhead.

  It is perhaps a good idea to post-processing of the data base and crush the decription "Test name" by

  "The test cell name == NAME. You can search not post-processed by Searchin' for.

  protected tcnameMQs: set tcnameMQs = store. GetElementList ("MeasurementQuantity", "description = name of the test cell", true)
  
  protected tcnameMQ: for each tcnameMQ in tcnameMQs
  
    tcnameMQ.Properties ("description"). Value = "name of the test cell is" & tcnameMq.References ("local_columns"). Elements (1). Properties ("values"). Value
  
  Next
  
  Store.Save

  Thereafter, you would be able to search for it.

  Greetings

  Andreas

• Report on the use of data store based on cluster (not data center)

  Hello

  I want to create separate HTML reports for each cluster I have in my virtual Center. I've created a script, but it doesn't seem to work. This script creates outputs separated from HTML based on clusters, but all the files have the same data, i.e. all data that are available in the Vcenter stores. How can I separate them with regard to the cluster in which they are assigned to the place?

  ===========================================================

  # Functions for math operations.

  usedspace {} function

  Param ($datastore)

  [math]: round (($datastore.)) CapacityMB - $datastore. (FreeSpaceMB) / 1024,2)

  }

  function dscapacity {}

  Param ($datastore)

  [math]: Round ($datastore. CapacityMB/1024,2)

  }

  freespace {} function

  Param ($datastore)

  [math]: Round ($datastore. FreeSpaceMB/1024,2)

  }

  function {percentage

  Param ($datastore)

  [math]: Round ((($datastore.)) FreeSpaceMB/1024) /($datastore.) CapacityMB/1024) * 100) / 1.2)

  }

  #Connect to Vcenter

  to connect-viserver-Server < myservername >

  # CSS stylesheet

  $a = '< style >.

  $a = $a + "BODY {background-color: Gainsboro ;}}.

  $a = $a + "TABLE {border-width: 1px;}. border-style: solid; border-color: black; border-collapse: collapse ;} »

  $a = $a + "TH {border-width: 1px;}. padding: 5px; border-style: solid; border-color: black; "{background-color: Blue}".

  $a = $a + "TD {border-width: 1px;}. padding: 5px; border-style: solid; border-color: black; "{background-color: PaleTurquoise}.

  $a = $a + ' * {do-family: Verdana, Arial, Helvetica, without serif;} '. font size: small ;} »

  $a = $a + ' < / style >.

  # get a list of clusters

  $clusters = get-cluster

  # Create HTML report for each cluster

  foreach ($cluster in $clusters)

  {

  $datastores = get-Datastore. where {$_.name - notcontains 'local'} | Sort the name

  $Report = @)

  {ForEach ($datastore to $datastores)

  $row = "" | Select-object Datastore, Datacenter, CapacityGB, UsedGB, FreeSpaceGB, PercentFree

  $row. Data store is $datastore. Name

  $row. Datacenter = $datastore. Data Center

  $row. CapacityGB = dscapacity $datastore

  $row. UsedGB = usedspace $datastore

  $row. FreeSpaceGB = freespace $datastore

  $row. PercentFree = % $datastore

  $Report += $row

  }

  $Report | Tri-objet-property PercentFree | ConvertTo-Html-head $a | Set-Content "D:\VMware\Scripts\Reports\Storage\$cluster.html".

  }

  ===============================================================

  To recover the cluster data warehouses, you must change the line:

  $datastores = get-Datastore. where {$_.name - notcontains 'local'} | Sort the name

  in:

  $datastores = $cluster | Get-Datastore. where {$_.name - notcontains 'local'} | Sort the name

• Error input/output of the VMFS5 local data store

  Hello

  I have 3 locally on the hard drive on my server from IBM. 1 250 GB drive, I used to create the data store Linux (datastore1) and other 2 disks that I created the Windows data store (datastore2)

  Couple of VMS is deployed on the two databases. Over the weekend, suddenly vms Windows Data Store (datastore2) went down. When I checked Monday concluded that data store is visible to ESXi host but I am unable to navigate to virtual machines.

  If I Directory VM cd I get below error:

  ~ # cd/vmfs/volumes/Windows\ Data\ store/Win - 2 k 8-2.

  SH: getcwd: input/output error

  # (unknown)

  ~ # esxcfg - scsidevs - l

  MPX.vmhba32:C0:t0:l0

  Device type: Direct access

  Size: MB 1911

  Full name: USB Direct-access Local (mpx.vmhba32:C0:T0:L0)

  Multichannel plugin: NMP

  Device console: /vmfs/devices/disks/mpx.vmhba32:C0:T0:L0

  Devfs Path: /vmfs/devices/disks/mpx.vmhba32:C0:T0:L0

  Seller: SMART model: USB - IBM Revis: 0000

  : SCSI 2 is Pseudo: false status: on

  : RDM Capable false is removable: true

  Is Local: true is SSD: false

  Other names:

  VML.0000000000766d68626133323a303a30

  VAAI status: unsupported

  MPX.vmhba33:C0:t0:l0

  Device type: CD-ROM

  Size: 0 MB

  Full name: Local USB CD-ROM (mpx.vmhba33:C0:T0:L0)

  Multichannel plugin: NMP

  Device console: /vmfs/devices/cdrom/mpx.vmhba33:C0:T0:L0

  Devfs Path: /vmfs/devices/cdrom/mpx.vmhba33:C0:T0:L0

  Seller: KVM model: vmDisk-CD live again: 0.01

  : SCSI 2 is Pseudo: false status: on

  : RDM Capable false is removable: true

  Is Local: true is SSD: false

  Other names:

  VML.0005000000766d68626133333a303a30

  VAAI status: unsupported

  NAA.5000c500478aee93

  Device type: Direct access

  Size: 286102 MB

  Full name: IBM-ESXS Serial Attached SCSI disk (naa.5000c500478aee93)

  Multichannel plugin: NMP

  Device console: /vmfs/devices/disks/naa.5000c500478aee93

  Devfs Path: /vmfs/devices/disks/naa.5000c500478aee93

  Seller: IBM-ESXS model: ST9300605SS Revis: B556

  SCSI level: 6 is Pseudo: false status: gradient

  RDM Capable is: true is removable: false

  Is Local: the SSD of is wrong: false

  Other names:

  VML.02000000005000c500478aee93535439333030

  VAAI status: unknown

  NAA.5000c500478b8d3b

  Device type: Direct access

  Size: 286102 MB

  Full name: IBM-ESXS Serial Attached SCSI disk (naa.5000c500478b8d3b)

  Multichannel plugin: NMP

  Device console: /vmfs/devices/disks/naa.5000c500478b8d3b

  Devfs Path: /vmfs/devices/disks/naa.5000c500478b8d3b

  Seller: IBM-ESXS model: ST9300605SS Revis: B556

  SCSI level: 6 is Pseudo: false status: gradient

  RDM Capable is: true is removable: false

  Is Local: the SSD of is wrong: false

  Other names:

  VML.02000000005000c500478b8d3b535439333030

  VAAI status: unknown

  NAA.5000c500478d724f

  Device type: Direct access

  Size: 286102 MB

  Full name: IBM-ESXS Serial Attached SCSI disk (naa.5000c500478d724f)

  Multichannel plugin: NMP

  Device console: /vmfs/devices/disks/naa.5000c500478d724f

  Devfs Path: /vmfs/devices/disks/naa.5000c500478d724f

  Seller: IBM-ESXS model: ST9300605SS Revis: B556

  SCSI level: 6 is Pseudo: false status: gradient

  RDM Capable is: true is removable: false

  Is Local: the SSD of is wrong: false

  Other names:

  VML.02000000005000c500478d724f535439333030

  VAAI status: unknown

  ~ # esxcfg - scsidevs - m

  NAA.5000c500478d724f:1 /vmfs/devices/disks/naa.5000c500478d724f:1 4f83af28-52abeaf2-6475-5cf3fcba46f2 0 Linux Data Store

  NAA.5000c500478b8d3b:3 /vmfs/devices/disks/naa.5000c500478b8d3b:3 4f7cfc39-4d17993a-a08c-5cf3fcba46f2 0 Windows Data Store

  Data store of Windows NAA.5000c500478aee93:1 /vmfs/devices/disks/naa.5000c500478aee93:1 4f7cfc39-4d17993a-a08c-5cf3fcba46f2 1

  Here is the output of dmesg: I suspect to be a hard drive problem

  (2014 06-16 T 10: 46:25.385Z cpu5:382346) WARNING: UserObj: 675: could not crossdup fd 8, / dev/ptyp0 CHAR type: busy

  (2014 06-16 T 10: 46:25.385Z cpu5:382346) WARNING: UserObj: 675: could not crossdup fd 10, / dev/ptyp0 CHAR type: busy

  (2014 06-16 T 10: 46:25.385Z cpu2:382348) WARNING: UserLinux: 1340: unsupported: (void)

  (2014 06-16 T 10: 46:52.510Z cpu16:4112) NMP: nmp_ThrottleLogForDevice:2318: Cmd 0 x 28 (0x412440db9c40) to dev 'naa.5000c500478b8d3b' on the way 'vmhba2:C0:T0:L0' failed: H:0 x D:0 x 2 P:0 valid 0 x 0 sense-data: 0x4 0x32 0x0. Bill: NO

  (2014 06-16 T 10: 46:52.510Z cpu16:4112) ScsiDeviceIO: 2316: Cmd (0x412440db9c40) 0 x 28, CmdSN 0x2a to dev 'naa.5000c500478b8d3b' has no sense H:0 x D:0 x 2 P:0 valid 0 x 0-data: 0 x 4 0 32 x 0 x 0.

  (2014 06-16 T 10: 47:34.631Z cpu18:4114) NMP: nmp_ThrottleLogForDevice:2318: Cmd 0 x 28 (0 x 412441635000) to dev 'naa.5000c500478b8d3b' on the way 'vmhba2:C0:T0:L0' failed: H:0 x D:0 x 2 P:0 valid 0 x 0 sense-data: 0x4 0x32 0x0. Bill: NO

  (2014 06-16 T 10: 47:34.631Z cpu18:4114) ScsiDeviceIO: 2316: Cmd (0 x 412441635000) 0 x 28, CmdSN 0x1c to dev 'naa.5000c500478b8d3b' has no sense H:0 x D:0 x 2 P:0 valid 0 x 0-data: 0 x 4 0 32 x 0 x 0.

  (2014 06-16 T 10: 47:41.476Z cpu18:4114) ScsiDeviceIO: 2316: Cmd (0x412440da3500) 0 x 28, CmdSN 0 x 21-dev 'naa.5000c500478b8d3b' has no sense H:0 x D:0 x 2 P:0 valid 0 x 0-data: 0 x 4 0 32 x 0 x 0.

  Please guide me to solve this problem. Thank you all in advance

  /virtu4

  This problem is solved.

  Thanks to the communities for not responding, that made me to fix this myself

  /virt4

• Fill the CSV file with the host name and the name of data store

  Hello

  How can include the hostname in my output csv file?

  The following code adds the name of the data store but not the host name:

  import-csv hosts | % {get-vmhost-name $_.name} | Get-Datastore. SELECT name.

  Export-csv ESX - Datastore.csv - notypeinformation - useculture

  How to watch your entry CSV file?

  If it's something like

  Name

  ESX1

  ESX2

  Do it like this

  &{foreach($esx in (Get-VMHost -Name (Import-Csv host.csv -UseCulture | Select -ExpandProperty Name))){    $esx | Get-Datastore |    Select @{N="VMHost";E={$esx.Name}},Name}} | Export-Csv report.csv -NoTypeInformation -UseCulture
  

• Use the error report data store capacity

  Hello world

  When I try to get the report for a capacity of data store, I got an error like home but he had no explanation for the error.

  Does anyone know why I get this error? I've updated vCOps from 5.0.3 to 5.6. Maybe it has to do something with that.

  Thank you in advance.

  Thanks for the replies;

  I mean by object parent exactly your situation - this is the under(inside?) folder.

  If you can, try to move a data store on the file and reports should work for it.

  in any case, it is a known problem that will be fixed in the next version of update.

• Disassemble the help of data store, modify the existing function

  I would like to change the below function to pass a single host settings and a data store single instead of doing all THE hosts connected to the lun.  I would like to be able to specify the news.

  Function Remove Datastore { }

  [CmdletBinding()]

  Param

  (

  [Parameter (ValueFromPipeline

  = $true )]

  $Datastore

  )

  Process of

  {

  if (-not $Datastore) {

  Write-Host "No data not defined as an input store"

  Output

  }

  Foreach ($ds in $Datastore) {

  $hostviewDSDiskName = $ds. ExtensionData.Info.vmfs.extent [0]. DISKNAME

  if ($ds. ExtensionData.Host) { }

  $attachedHosts = $ds. ExtensionData.Host

  Foreach ()$VMHost en $attachedHosts) {

  $hostview = Get-View $VMHost. Key

  $StorageSys = Get-View $HostView. ConfigManager.StorageSystem

  Write-Host 'Disassembly $($DS.) VMFS Datastore. The name) host $($hostview ). «' «Name)...»»»

  $StorageSys . UnmountVmfsVolume ($DS.) ExtensionData.Info.vmfs.uuid);

  }

  }

  }

  }

  I tried to change it to this, but don't think I have it correct.  I would run like this preference:

  $esx = get-vmhost hostname

  Disassemble-Datastore - VMHost $esx - data datastorename store

  Function

  Remove Datastore {

  [CmdletBinding()]

  param (

  [

  [ VMware.VimAutomation.ViCore.Impl.V1.Inventory.VMHostImpl ]$VMHost

  [

  string ]$Datastore

  )

  Process of {

  if (-not $Datastore) {

  Write-Host "No data not defined as an input store"

  Output

  }

  $hostview = Get-View $VMHost. Key

  $StorageSys = Get-View $HostView. ConfigManager.StorageSystem

  #Write - Host "disassembly $($DS.) VMFS Datastore The name) host $($hostview.) (Name)... »

  $StorageSys . UnmountVmfsVolume ($Datastore.) ExtensionData.Info.vmfs.uuid);

  }

  }

  Try it like this

  function Unmount-Datastore {
    [CmdletBinding()]
    param(
      [VMware.VimAutomation.ViCore.Impl.V1.Inventory.VMHostImpl]$VMHost,    [string]$Datastore  )
  
    process {
      if (-not $Datastore) {
        Write-Host "No Datastore defined as input"      exit    }
  
      $ds = Get-Datastore -Name $Datastore    $hostview = Get-View $VMHost    $StorageSys = Get-View $HostView.ConfigManager.StorageSystem    #Write-Host "Unmounting VMFS Datastore $($DS.Name) from host $($hostview.Name)..."
      $StorageSys.UnmountVmfsVolume($ds.ExtensionData.Info.vmfs.uuid)  }
  }
  

  Given that the data store is defined as [string] I suppose you pass the datastorename.

  So you'll need to get the data in $ds store object

• Increase the capacity of data store, which with the block size?

  Hello

  I am trying to expand my datastore 2 TB, I added a new lun 2 TB of the FC SAN TB 4 even and there is no problem adding this in the existing data store.

  But the existing data store is formatted in VMFS3 with a block size of 4 MB for a maximum file size of 512 GB. But when I try to extend the data store to select the size of the block is set to 1 MB and disabled.
  The measure automatically takes on the size of the original block or wil he get a block size of 1 MB? There are some > 256GB files on the existing data store and I think I'd get a strange behavior, if half of the data store is 4 MB and the other half is 1 MB blocksize.

  Can anyone confirm that the blocksize wil be the same for the measure?

  I can also make a second warehouse 2 TB one manually fracture the virtual machine on 2 data warehouses. is there any advantages or disadvantages to this approach?

  Don't you worry, add an item to an existing data store certainly will not change the size of the block. In any case, if it is an option that you can use two data warehouses, I would. With two distinct data warehouses that you will not only be able to spread the load on each of these LUNS but also avoid the complexity. With ESXi 5 (VMFS-5), a single LUN can grow up to 64 ~ to without needing extensions.

  BTW, a 4MB block size allows 1 TB less than 512 bytes.

  André

• Draw attention to the view of data store

  Hello

  I'm intrigued by this alert, shows a red! on the same data store you I have more than 50% disk free space. I remove a 300 GB VMDK to a VM about 2 months ago...

  Please see photo.

  Thanks, Julien

  Hello.

  Turn off this alarm and then immediately turn it on again. To clear the alert.

  Good luck!

• find the hard disk data store

  Hello:

  1. is it possible to find a data store where the hard drive (I know, there should be :)?

  2. How do I find all the drives on VM (s) (if there is more than 1) and correlation of data warehouses?

  I need to run storage Vmotion (2 times), but the drive should end up the original data store is...

  Thank you very much

  Of course, try this

  Get-VM | %{
      $_.HardDisks | Select @{N="VM";E={$_.Parent.Name}},
          Name,
          @{N="DS";E={$_.Filename.Split(']')[0].TrimStart('[')}}
  }
  

• How to select the cluster/resourcepool data store?

  Hello

  as you know others my son, I'm working on a workflow to create several identical virtual machines on a VC-cluster.

  At this time a virtual computer is created like this:

  task = vmFolder.createVM_Task( configSpec, vmCluster.resourcePool);
  

  Context is being filled with the configuration of the virtual machine. vmCluster is a workflow input parameter and contains the cluster.

  It works well, but I still have to specify the data store to store files on a different input parameter, because the cluster has several shared storage units.

  Note also that our clusters have only a single pool of resources, the default.

  Let the user select a data store is not a very nice solution because of possible errors and mistakes (the user can select the local drive of the VMHost, lack of space on the storage selected etc..). But I, as a developer, also can not predefine the storage to use (several storage units, unit of different names etc.).

  Now, what I've been thinking about (and trying to implement) was to get the vmCluster data warehouses or the resourcePool, check if they are put in place for the virtual machine files and if they have enough space for the virtual machine.

  I know how to check the size and which storage are to be used for virtual machines, but I don't know how to get out them of the cluster/resourcepool.

  Can someone help me with this?

  Thank you and

  Concerning

  Andreas

  PS: Is it just me or I really make things complicated with orchestrator here?

  Hi Andreas,

  One way to get the value of storage of data is to use VcSdkConnection.getAllDatastores

  For more information, see here: http://www.vmware.com/support/orchestrator/doc/vco_vsphere41_api/html/VcSdkConnection.html#getAllDatastores

  Personally, I have not played with it, but it seems very powerful.

  There is discussion on the use of xpath in such a case in this topic (to get the value of VMs): http://communities.vmware.com/message/1673575

  Other than that you could:

  -use VcSdkConnection.getAllClusterComputeResource (gets all clusters) or VcSdk.getAllResourcePools

  -in case you work with clusters - each cluster's data store property that returns an array of data for this cluster warehouses

  -in case you use pools of resources, you must check that is the parent of each resource pool and get its data warehouses.

  Kind regards

  -Martin