Share the data store between clusters

Hello

I want to add a second group to our existing data center and want to use the same data store (iSCSI) as for the first group. Is it safe to share between clusters within the same vCenter/datastore? All hosts are running the same version of vSphere 5.1 and managed by vCenter device 5.1. If additional information is required, ask.

Yes you can share between clusters. You will share it on hosts. It won't matter for storage if the hosts are in a different cluster or a cluster, as you do at the host level.

Tags: VMware

Similar Questions

  • HA will only work if I share NFS data store in the cluster

    Hi all

    Please help me with my question below?

    HA work if I share NFS data store in the cluster? Even if if works is it advisable?

    Thank you

    Lucette

    Yes, but a VM restarts on a host within the same cluster VMware only.

  • Target iscsi ESXI 5.0, active path connected, mounted, are not not in the data store data store

    HI -.

    I am looking for help to point me in the right direction.  This problem occurred after a reboot of the system.

    I'm in vmware esxi essentials 5.0.0 821926

    I use the starwind software as my iSCSI target

    I use ISCSI to connect to my server from storage to the esxi hosts.

    I have 2 warehouses of data showing inactive under clusters of data storage and the data store.  I have a third-party data store on the same server that is loading properly.

    I currently have the same behavior on three esxi hosts:

    under configuration - storage adapters is the ISCSI path is active. Data warehouses appear under devices and mounted.  In data warehouses, they are inactive. On the same storage server

    on the starwind Server - I have

    built a new target

    removed and added the databases to the new target

    Changed the IP address of the target.

    On Esxi hosts, I removed the ISCSI server, discover places BOF dynamic and static and added under the same or IPs news discovery

    I can restart esxi hosts & storage servers and get to that same point.

    Whenever I find myself in the same place - course assets, devices mounted, inactive in data warehouses

    I don't know what else to share, let me know what you need to know to help out me.

    Thank you

    Dylan

    OK, incase someone crosses my ramblings, it may help.  My original question was warehouses of data that would not rise.  They were visible as devices in the area of storage under devices and as devices in the scsi adapter.

    When I would try to force mount them, add storage, keep the signature, mount they wouldn't mount.  I did some research and my question was after a failed mount attempt strength as got data attached as snapshots warehouses.  I then mounted force them and he tells will data warehouses and I got them back.

    I followed this KB to identify and resolve the problem

    manipulation of vSphere detected as snapshot LUN LUN (1011387)

    When it was over, I tried to add vms back and found a new show.  Power of vms would expire.  the vmkernel.log showed;

    2014 05-27 T 07: 20:40.010Z [verbose 'Vmsvc' 564D7B90] Vmsvc: filtering Vms: ignored not ready to request power state vm vim. VirtualMachine:75

    AND

    (2014 05-27 T 03: 45:47.821Z cpu4:2052) NMP: nmp_PathDetermineFailure:2084: SCSI cmd RESERVE failed on the vmhba35:C0:T1:L0 path, booking on eui.cff548d1952b1e4c of device status is unknown.

    I had latency huge read write showing upwards, 3K and more

    After several searches, I had in the ESXShell and found that there is no conflict of booking.

    On a whim, I took inventory all virtual machines that are now inaccessible.  I then added a DOS virtual machine. Alto! the latency down to version 1.2 a.7 ms for all data warehouses.

    ultimately the instructions said you may need to add virtual machines in the inventory, but does not remove all virtual machines first.  I was grabbing vms that were not in stock, so I didn't remove the old virtual machines in the inventory.

    A recruit Mennonites, Yes.

  • How do you get the path of the data store or the URL of a virtual disk in vRealize Orchestrator?

    I need to eager zero a virtual disk while building a virtual machine. I found the eagerZeroVirtualDisk_Task() under VcVirtualDiskManager function, which takes in 2 parameters:

    name - (String) the name of the disk, is a path to the data store or a URL pointing to the virtual disk that needs to be inflated.

    Data Center - (VcDatacenter) if the name is a path of data store, the data center for the path of the data store. Not necessary when called directly on ESX. If not specified on a call to VirtualCenter, the name must be a URL.

    How can I get the correct path of data store given the virtual disk? I intend to iterate through the features of the virtual machine to get the disks.

    Thank you!

    Actually figured out how to get the path of the data store, and how a drive zeros. In case it is useful to share the code here:

    var vimHost = vm.vimHost; VM instanceof VCVirtualMachine

    var vDiskMgr = vimHost.virtualDiskManager;

    for each disk

    var features = vm.config.hardware.device;

    for each {(dev var in devices)

    dev.backing contains a lot of useful things...

    If (dev instanceof VcVirtualDisk) {}

    a disc!

    var dsPath = dev.backing.fileName;

    System.log ("path of disk:" + dsPath);

    Data var Center =

    System.getModule('com.vmware.library.vc.datastore').getDatacenterForDatastore (dev.backing.datastore);

    var vcTask = vDiskMgr.eagerZeroVirtualDisk_Task (dsPath, datacenter);

    wait

    var waitResult =

    System.getModule('com.vmware.library.vc.basic').vim3WaitTaskEnd (vcTask, true, 5);

    }

    }

  • Migration of virtual machines from a data store, and then delete the data store?

    Hello

    I have a future deployment this month and my Director wants to build a new RAID 6 array, create a store of data with the new table and then move all the VMS to the newly created data store.

    Then he asked me to take the old data store and remove it and the available space between the deleted data store and other stores of data in our shared environment. The question is when the new raid is created and vm has migrated to the new data store, which is the best way to remove the data store empty and make sure that the space available for other data stores? Any help would be appreciated...

    Greg ~.

    If all controls are met, you can go ahead and take it apart without any problem.

    I have re-iterate below checks:

    -No virtual machine is in the data store

    -The data store is not part of the cluster data store

    -The data store is not managed by the DRS storage

    -IGCS is disabled for this data store

    -The data store is not used for vSphere HA heartbeat.

    Especially highlighted 2 controls. Please note that SIOC can be enabled on the data without SDR photo store

  • What best practices to remove a host server data store without emptying the contents of the data store

    Hi all

    This question was probably asked a million times. the best practice to remove a host server data store without emptying the contents of the data store?

    I have two clusters and I moved host group A to group B. host A is now to see data from Cluster Both A store and Group B datastore. We want that host A to store data in Cluster B. The Cluster a data store is not and can not be empty coz VM A Cluster using data warehouses. So I just unpreset the data LUN to host A store and Reanalyze the data store? Or there is a better way to do it?

    Thank you all

    Yes that's what I meant - should have specified where virtual machines are running.

  • Technical question from the data store

    Hello everyone.

    I have a question that you can maybe answer:

    "Is it possible to have a WARNING message (check) before formatting a LUN (EMC)?"

    Let me explain:

    We currently have a datastore composed of 12 LUN (of 236 GB), and even if this isn't in the best practices of VMware, I made a mistake (und) when creating a new data store.

    Indeed, I used a logic unit number which was attached directly to a virtual machine (in Raw Device) pour set up my new data store.

    Would you have a solution that makes a data control (on the logic unit number) and which could see us the partition proposed pour extend the data store contains data!

    Thanks in advance for your answer!

    J. Gaspoz

    Hi Julien

    Nice to see you on communities

    The vCenter will list all LUNS that are not already in the "Add storage" wizzard data warehouses, and by definition, a RDM en of is not formatted VMFS, since the lun belongs to the VM and it met with filesystem clean so son.

    There is therefore no direct way to pour do what you're asking (UN ESX knows nothing about the data on a non-vmfs LUNs).

    Personally, plug-in a simple way: the LUN ID. All the LUNs that will be used in RDMS are presented to my ESX with UN LUN ID between 50 and 70 (f.ex.). So I know whether or not a lun is RDM.

    Otherwise maybe third-party products (rv maybe tools) can get out a report listing the VM using the RDM.

    I hope that helps!

    Best regards.

    Stéphane Grimbuhler

    Senior consultant virtualization & storage (VCP / VCAP-DCA)

    VMware instructor (VCI)

    My Blog: www.virtualgeek.ch

  • Recommended minimum on the data store when free disk space you use connection vRanger 5

    I was wondering if anyone can tell me what the recommendation is on the left of free space on a data store. I use 5 vRanger connection to back up my VM to another physical server. As such, I will not take snapshots 'manual '. I'm not sure if vRanger connection, running on a separate server, creates additional files on the data store.

    I've read a few threads on this site and some are advocating to let free, 20 to 25%, while that of others, offer 10% left free may be sufficient. Is there an official number for this or if anyone can share valuable experince?

    Welcome to the community,

    I generally recommend to have about 20% of free space on each data store. However, this depends on the activity/writes to disk during backup. vRanger connection - as well as any other backup software - create a snapshot, back up the virtual machine and then remove / commit the snapshot again. This is necessary in order to access the virtual disk files. Usually the backups are made during hours, so snapshots should not grow too much. However, you must also take into account the fact that in the case of a mistake - if the snapshot cannot be deleted - it will grow until it is deleted manually.

    André

  • Cloning into a subfolder of the data store

    I have a powershell script that creates daily snapshots and clones on Sunday.  I would like to have clones stored in a subfolder on a NFS share.  I can't figure out how to set the data store in a subfolder like "NFS_NAS_SHARED\CLONES".  The following code fails indicating "'NFS_NAS_SHARED\CLONES' is not found, using the specified filters."  Is it possible to have the destination clone as a subfolder of a data store?

    Start-Transcript-Path "C:\VMWare_Scripts\LOGS\SnapshotsTranscript.log".

    #Get the VMware PowerCLI
    Add-pssnapin VMware.VimAutomation.Core - ErrorAction SilentlyContinue

    # Import the backup information
    #File layout: MasterVM (vmname), BackupDS (DS to store clone), BackupFolder

    #$backupinfo = import-Csv C:\VMWare_Scripts\Snapshot_backups.csv
    $backupinfo = import-Csv C:\VMWare_Scripts\Snapshot_backups_TESTING.csv

    Date format #Set to name
    $datetime = get-Date-Format "YYYYMMDD-hh: mm".
    $date = get-Date-Format "YYYYMMDD".

    #Connect to vCenter
    SE connect-VIServer "localhost".

    #Get the day of the week
    switch ((get-date).dayofweek.toString ())
    {
    Sunday {$dow = "SUN"; break}
    Monday {$dow = 'MY'; break}
    Tuesday {$dow = "KILL"; break}
    Wednesday {$dow = "WED"; break}
    Thursday {$dow = "THU"; break}
    Friday {$dow = "FRI"; break}
    }

    {foreach ($customer to $backupinfo)

    $vm is get - VM $customer. MasterVM
    $days = $customer. DaysToKeep

    Write-Host $vm. Name
    #Write - Host $days

    #Delete snaps more $days
    Get - VM $vm | Get-Snapshot | Where {$_.} Created - lt ((Get-Date).) AddDays(-$days))} | Remove-Snapshot - confirm: $false

    If ($dow - eq 'MY') {}
    #Create a clone

    #Create a new snapshot for cloning
    $cloneSnap = $vm | New-Snapshot - name "clone_snap" - suspend: $true

    # Get the managed object display
    $vmView = $vm | Get-View

    # Get the managed folder object reference
    $cloneFolder = $vmView.parent

    # Construction specification clone
    $cloneSpec = new-object Vmware.Vim.VirtualMachineCloneSpec
    $cloneSpec.Snapshot = $vmView.Snapshot.CurrentSnapshot
    $cloneSpec.Location = new-object Vmware.Vim.VirtualMachineRelocateSpec

    #set the destination clone
    $backupDestination = $customer. BackupDS + $customer. BackupFolder
    $cloneSpec.Location.Datastore = (get-Datastore-name $backupDestination |) Get - View). MoRef
    $cloneSpec.Location.Transform = [Vmware.Vim.VirtualMachineRelocateTransformation]: sparse

    $cloneName = "$vm - $datetime".

    # Create clone
    $vmView.CloneVM ($cloneFolder, $cloneName, $cloneSpec)

    # Write newly created VM to stdout as confirmation
    Get - VM $cloneName

    # Remove snapshot created for clone
    Get-Snapshot - VM (Get - VM $customer. MasterVM) - name $cloneSnap | Remove-Snapshot - confirm: $False


    }
    else {}
    #Take a snapshot
    New-Snapshot - name $datetime - description "scripted Snap: Snapshots.ps1" - vm $vm - suspend: $true
    }

    }

    VCenter #Disconnect
    Disconnect-VIServer-confirm: $false

    Stop-Transcript

    I'm afraid that the CloneVM_Task method you cannot specify subfolders on a data store.

    He wants a data store and use the displayname property of the virtual machine to create a folder on this data store.

    If you clone a computer virtual called XYZ to the store of data DS1, the clone will be located in the "[DS1] XYZ" folder.

    If you want to move the clone to a different folder, you can use the cmdlet Copy-DatastoreItem thereafter.

  • Help display information from the data store

    Hi, I'm trying to format a table with information from various commands, and I can't find out how to do it.

    Basically I want the table to have the following information:

    ESX host - DatastoreName - CanonicalName - CapacityMB - MultipathPolicy

    The problem is the get-scsilun cli annoys me only the canonical ability and multiple trips, but I want to add the other get-vmhost and get-datastore information

    The command I'm ussing is:

    Get-vmhost | Get-ScsiLun | WHERE-object {$_.} Seller - eq "EmC" - and $_. LunType - eq 'disc'} | Format-Table-property CanonicalName, CapacityMB, MultipathPolicy-auto | Out-String-width 120

    No idea how to do that?

    Thanks in advance!

    Pablo. -.

    Try something like this

    $report =@()
    
    foreach($esx in get-vmhost){
        foreach($lun in (Get-ScsiLun -VmHost $esx -LunType disk | where-object {$_.Vendor -eq "EMC"})){
            foreach($ds in (Get-Datastore -VMHost $esx)){
                $ds.ExtensionData.Info.Vmfs.Extent | %{
                    if($_.diskName -eq $lun.CanonicalName){
                        $row = "" | Select Host,DS,CanonicalName,CapacityMB,MultiPathPolicy
                        $row.Host = $esx.Name
                        $row.DS = $ds.Name
                        $row.CanonicalName = $lun.CanonicalName
                        $row.CapacityMB = $lun.CapacityMB
                        $row.MultiPathPolicy = $lun.MultipathPolicy
                        $report += $row                }
                }
            }
        }
    }
    
    $report | ft -AutoSize
    

    The link between the data store and the logic unit number is done by comparing canonicalname the LUN with each canonicalname of the extent of the disk of the data warehouses.

  • Method to move the virtual machines in the data store temporary so I can rebuild pool recommended NFS

    Our current data store is a mounting NFS on Nexenta (zfs), which has a pool, I need to rebuild (migration of raidz2 to mirror striped for performance). Initially, I was relying on the judgment of all virtual machines, create a new pool, copy (cp - r) of the Nexenta console all virtual machines to the new pool, recreating the pool mirror and then copy (cp - r) all the data back.

    I have seen several discussions on this topic and seems like the simplest method would be to use the feature to "Migrate" so that the virtual machines are not fragmented. However in some of my pre-test when I copy a virtual machine between the basins of the Nexenta console it appears (from the data store Explorer) to initially create the vmdk provisioned and then once completed it copy looks too thin.

    I hope that I can simply copy the Nexenta console, using the option "Migrate" in vCenter on each of my 40 MV looks like it would be * quite * tedious compared to the simple issuance of an order of 'cp - r' and leave all the copies of VMs (1.5 to) during the night and then adding them to each host.

    Does anyone have experience with this? Or maybe a way to generate scripts easily (?) the option "migrate" in vCenter?

    Thank you all!

    You can select multiple VMs in vCenter regularly using CTRL + click or shift + click and both migrate them as long as they have the same storage destination.  You can also schedule it as a task.    Add manually each computer virtual back in hand is not less painful than migrating individually, anyway.

  • EagerZeroedThink! Why read i/o on the data store while inflation

    Hi all

    I got a call from a customer of 2 weeks ago asking why there reading activities on its data store when it swells a vmdk. I didn't either, and did in my Infrastructure to justify.

    Whenever I created a vmdk for EagerZeroedThick, with the Virtual Machine Wizzard (functions of Clustering Support etc.) create or inflate the context. Prezeroeing starts at 1/3 of the activity is read from the data store and the other 2/3 is in writing. I expect 100% writing to the data store since I thought that prezero a full vmdk involves only write operations (write zeros). The data store does not support vAAI and tried since with EMC and Netapp subsystems, Fibre Channel and iSCSI. Its always the same. With EMC its even more read than write operations.

    Anyone have an explanation for the behavior of ths. I looked around and didn't find anything.

    Thank you

    Harry Mitschele

    VCI and senior Consultant

    The way I understand it, the process of resetting (no matter if it has run for a newly created or an existing vmdk) zeros only blocks that are not already reset to zero. That's why he should read each block before eventually writing on it.

    Take a look at KB http://kb.vmware.com/kb/1011170 (section "vmkfstools-t0") which deals with the topic of zero setting.

    André

  • VMDK is not displayed when you add the drive but shows on browse the data store

    This may be a simple solution but my forehead hurts to hit my head against the wall and need your expert help.

    Environment: running ESX 4.0 with 4 nodes in the cluster.  Each node in the cluster has only one HBA card with double fiber connections that connect to both sides of FC SAN.  There is not a fiber between nodes and SAN switch, so direct connect.  Built two systems to support Exchange 2007 using the CCS.  Using a FC SAN to shared disk space and iSCSI for disk space not shared.

    Problem: have added 5 player glow spaces on System #1 using RDM and they appear.  Try to load the same readers in the System #2.  I am able to load 4 readers in the System #2 but the 5th disc (which is in fact the reader 3rd in the series) refuses to show when I added the player.  The change settings - & gt; Add Hard Drive - & gt; Use the existing - & gt; Open the data store and the folder where the virtual machine from the system files and the folder where the LUN mapping lie as they are stored with the system.  But at the opening of this file the LUN mapping in question is absent.  All others appear.  However when I right click to browse the data store outside of the settings and go to the folder where the LUN mapping for the virtual drive and it appears.

    Attempts to resolve:

    (1) have removed the drive and delete system #1 data store and we readded to the system = & gt; nothing.

    (2) removed the disk without delete system data store #1 and it added (thus creating two virtual disks in the file for the system to the same LUN... xxx_3.vmdk and xxx_7.vmdk) = & gt; without success.  (xxx_3.vmdk, nor xxx_7.vmdk appear when trying to add a player)

    (3) tried to use a system not intended within the CSC Exchange cluster, attemtped to add the xxx_3.vmdk or xxx_7.vmdk and both appear as an option.

    I am at a total loss.  Do I need to trash the system no. 2 and rebuild?  Is there a problem of possible system configuration #2 that does not allow him to see the two virtual drives?  The #2 system configuration is corrupt?  Do I have to power System #2 power on and off and then try to add readers?  I can the virtual cycle vCenter system?  Or I am much more of a noob that the answer is with myself and I can't see it?  Any help would be welcome.

    When you want to use RDM on hosts, please make sure lun the same number on hosts.

    for example you have assigned the lun 3 host A storage, make sure that it is also the lun 3 to host b.

    Let me know if it does not

    Binoche, VMware VCP, Cisco CCNA

  • Possible to remove the data store shared without deleting content?

    Is it possible to delete a data store shared a host, without actually deleting content?

    If there are several hosts who share a store of data and for some reason any I delete the data of one of them, store connection can I do this without a deletion of all the virtual machines on the data store, as shown in the warning above?

    You can delete single LUN on the side storage (for example by hiding the it).

    Before that... do not forget that without virtual machines are running on it.

    After that... do a rescan storage adapters.

    André

  • Renamed the VM inventory are not not in the data store?

    I created a virtual machine and then renamed from the inventory Panel.

    Later, I noticed that there is a record in the data store more with the name of the original virtual machine. When I went to rename the file to match the name of the virtual machine in the data store, I get an error that the object is not found. There are no files listed in this folder in the data store.

    The virtual machine seems to operate normally, but I can't find its files anywhere in my data store?

    Hello

    To rename a virtual computer, there are 2 steps: remove the inventory of your virtual machine and rename it on the ESX.

    • You must turn off the virtual machine.

    • Remove inventory VM concerned. You must do this because the link between VC and the file will be lost.

    • To rename on the ESX, you should know where all the files stored one. Connect you to the ESX Server with ssh. Rename the files with the correct name, and to be clean, rename the directory too.

    • Editing with vi vmdk and vmx files to change the name and path to the files of the virtual machine.

    • Once completed, the VC, browse your LUN to find your vmx file and add it to the inventory.

    It's the only way I found to properly rename a virtual machine. It's not complicated

    http://communities.VMware.com/message/822788;JSESSIONID=F44C51342768E9B02B25A12A26A4571D

    * If you found this information useful, please consider awarding points for

    "Correct" or "useful."

Maybe you are looking for

  • Disabled Firewall - now starts like new computer

    Based on a recommendation of the Apple Genius Bar, I disabled my firewall.  It took about 20 minutes and then says to restart when he did.  When I rebooted, it starts as a new computer asks me to select a language and then restore from time machine,

  • Keithley 2100 and Labview 7.1?

    HelloI own (happy?) a new 2100 Keithley and all I found drivers for Labview 8.0 or later version. I want to use this device on a computer Windows XP running Labview 7.1 (on another machine, I have a LV8.5 but I can't save this crimp screw for prior t

  • Trouble installing the 2012 E - tax to the ATO.

    Error message "this installation package could not be opened. Contact the seller to check if this is a valid windows installer package. Troubleshooting installation E-tax 2012 ATO after I install msi etax 2011 that I downloaded on the internet. I tri

  • Remove 8.1 components from a secondary drive system

    Hello I had another question on a new installation of 64-bit Windows 8.1 on two SSD.  My computer has an ASUS PZ778-V pro motherboard running an i5 - 3570 k and 32gbs of memory chip.  (Once again thanks for the solution to the question of the double

  • FWSM and multiple-vlan-interface command

    Hello I am due to configure a FWSM in an IOS 6513 running, I suppose that the example configurations are quite similar, when you use an MSFC, but I have a question. I have 3 inside VLANs, configure 3 LAN interfaces VIRTUAL with IP addresses (as a def