VMs migration of a non - CVS to EVC cluster cluster

Similar Questions

• EVC cluster creation / migration hosts him / minimal disruption VM

  I am reshape my somewhat environment and need a bit of advice.  Currently, I have 2 ESX hosts in a non - EVC cluster and built a new blade server.  The new blade and 2 hosts will work together in a cluster active CVS.  I know that I have to run my EVC cluster "Intel Xeon Core 2" mode.  All 3 hosts will support this mode.

  The News / 3rd cluster is sitting in his own vCenter, a VCA-activated cluster (in mode above).  Has no running VMS, it is already configured with network access and storage that matches that of the 2 existing/production hosts, which are currently managed by another instance of vCenter (but I am moving them out of this instance of vCenter in this new EVC-activated, to join the new blade.

  All hosts running ESX 3.5 U4.  vCenter is 2.5 U4.

  BIOS on guests must be modified to memory XD in so that they can be used in an EVC cluster, so I know I need to stop at SOME point to get them in the cluster active CVS.

  I have an idea on how to address these requirements, with a big goal is NO interruption of service for running virtual machines.  VMs can be migrated to a new cluster, EVC-activated, coming from an old cluster VCA not compatible?  Most all current VMs based on Intel some running already in reference "Intel Xeon Core 2" EVC mode or I'm just kidding me and must arrest all of them at some point after they "seized" a VCA-activated cluster?

  How would you do to this migration between vCenters and Clusters?  I think I can use 'Disconnect' the first vcenter (to at least avoid having to go into Maint Mode and then 'delete' of the first cluster/vCenter database)... also, I'm looking for the most logical get "here" to "there".  Thank you.

  I see that the problem of different vCenters/Clusters is resolved.

  What you need now is to know if you can VMotion between two ESXs with different processors. Before existing VCA, I have had... and had NOT a few difficulties doing this. It is not a problem when DIFFERENT pCPUs have the SAME GAMES of INSTRUCTIONS (SSE3, SSSE3, SSSE4 and so on...) Even if the sets are different, I was able to mix the MASK of CPU management options, which are by VM (thing that I didn't like... There should be an option to change this on a same CLUSTER basis or an ESX!).

  I recommend you to test for differences between the sets of instructions of the Jepp of two ESX with this tool, I think it is the best out there:

  http://www.run-virtual.com/?page_id=155

  If the sets of instructions are different, you have a problem, because you have to find the bits change in the MASK of CPU options in each virtual machine... and who needs a reboot, too. I think they will be different, because more recent pCPUs with FlexMigration (VCA compatible Intel processors), simply because they are new they surely have a new instruction set (perhaps SSSE4)

  Good luck!

• Active EVC cluster will not allow guests to reach...

  Recently, I installed a new vCenter Server 5 (vCenter 4.x running on a 32 bit OS then upgrade was not an option-fresh install)

  I had 1 main cluster in the datacenter with HA and active CVS. All 10 guests were very well in it.

  However, when the new vCenter Server has been up - and after having removed the old vCenter data center, when I went to add the host to the cluster (same as before active EVC mode) 6 guests joined very well, while the other 4 could not.

  Everyone knows this? Just a reminder:

  new vCenter Server 5

  all 10 guests were in an EVC cluster well

  4 of 10 guests will not join the same cluster...

  Thoughts?

  Hi xxx,.

  XXX wrote:

  Another question regarding VCA - if I point the old vCenter DB, the new vCenter server will this cause problems because a new database created for the new vCenter? How should I approach the issue?

  You can't do it now, you can not connect vCenter 5 dB in the version of vCenter 4, different schema, tables etc. as I said in my previuos reviews, you must migrate to vSphere with DB 5

  1. Start vCenter old (I hope that you did not remove or disconenct any server from there, right?)

  If you do not make any changes to the OLD vCenter vCenter should be able to connect to the return ESXi hosts and apply all the settings (including the CTS)

  And Yes, I take off (unplug the old vCenter servers... so I guess that reconnect the old Server vCenter is not possible.)

  If you simply disconnected the old vcenter ESX hosts (not deleted) you should be able to connect to vCenter old with old settings, but as I said earlier, I don't know what will happed with a VCA base (because you creted clusters with different CVS on new vCenter.)

  Artur

• vMotion & e EVC Cluster

  Salve,

  HO a couple of request riguardo a quanto in oggetto.

  1. vMotion: E' meglio sulla rete delle VMs o nella rete della SAN abilitarlo? Currently ho 2 vSwitch; UNO by the SAN by the VMs. Su quale e meglio abilitare uno e he vmotion by turned out the best performance? -Preventing che ho 4 nic brother by it vSwitch delle 2 nic brother by the SAM e VMs... all a 1 GB
  2. Cluster: Host of Posso creare a cluster con 2 con 2 modelli diversi di (uno e altro INTEL AMD) CPU? MI è sembrato di di understand if... sempre's abilito VCA... corretto?
  3. EVC: Dove if errors?

  Grazie thousand per aiuto

  Ciao,.

  You'll find more previous post a riguardo, my lo rispiego:

  -lascio iscsi isolato per evitare problemi di congestione del che essere bear area has brain delle vm dei blocchi

  -Moussa rete vmotion insieme a lan, dato che hai 3 nic in e no e previsto rates vmotion in continuazione of load balancing

  -nella configurazione standard, vmotion usa una scheda che no usata e dalla lan, in modo che non so infastidiscano a story

  -management of the crea UN area hotel, visit the Super della struttura

  varies-reti permettono Eve, is not as funzionare, di usare una what appunto rimaste in mode standby. Degradano performance, continued to platform of the my at least a funzionare

  By vmotion crei UN port group apposito di tipo vkernel, abilitando the "vmotion" spunta, e assegni due ip di una rete separata, che siano ovviamente tra gli host compatibility.

  Ciao,.

  Luca.

  --

  Luca Dell'Oca

  http://www.vuemuer.it

  @dellock6

  vExpert 2011

  [Assegnare punti a useful risposta e una a modo di say thank you]

• Changing the Mode of EVC Cluster for guests of Intel

  We cross a turnover of rental for our ESX host that puts out some old CPU architectures. Once we get rid of these hosts, we will be able to raise our EVC Cluster mode.

  (1) once ready, are there special precautions to take before changing the Mode of EVC Cluster? Then I just change the setting?
  (2) when the VMs will begin taking advantage of the new architecture CPU (a reboot will or VMs must they be turned off)?
  (3) any question with VM do not be put off for long periods of time after the VCA mode change?

  (4) all other things to worry about?

  If I raise the EVC mode, virtual machines will be able to access the new features available in this mode?

  Is that you turn off and then turn on the virtual machines. A reboot of the guest operating system is not enough. A virtual machine determines which functions are available to turn on and cannot access all the new features that are added until the power is turned off.

  http://KB.VMware.com/kb/1005764

• VMotion between EVC cluster and native cluster

  Is vmotion between EVC cluster in Nehlem mode, and a standard cluster with Nehlem processors had problems?

  given that computers virtual are presented Nehlem CPU mode in clusters botyh as ok, it should work.

• migrate to esxi free host in the cluster paid with minimal downtime

  Please excuse my ignorance, I am very new to VMware and I am familiar with the concepts of other virtualization software, I used, but not some of the termininology, and maybe that's why my research are not turning to the top of what I need.

  We had been using the free esxi hypervisor based to host some server hosts.  These servers are running a copy of our database and are used for reporting applications.  They are also setup with files VMDK provisioned very large and thick (> 300 GB!).  They were spread great for a year now.  However, we just bought the VMWare "vSphere 4 Essentials Plus" kit for the creation of a cluster of 3 machine, with a few redunancy and other features we wanted.  We also have a new san to go with it.  We are very pleased with the product and move to her!

  On the other invited VM running on this host, (and several other hosts, its amazing to see how they arise) I stop the virtual machine and the vCenter Converter allows you to migrate the virtual computer to the new cluster.  At the same time, I would like to convert it to the thickness of a provisioning. However, because these clients have very, very big thickness provisioned HD, its research as they would take five hours or more to migrate. (other VM moved to essentially 1 GB/min)  These report servers can go without updates the master DB for several hours, but servers should always be available. I can't have them down for 5 hours. There are a lot of historical reports are refreshed by the end users.

  Is it possible for me to do what follows somehow? I can't understand:

  • disable the DB update
  • closing comments, take a quick snapshot and then launch the guest.
  • While his copy running (and not updated) main and VMDK files file to the cluster (therefore only changes to files in the snapshot file)
  • on the new cluster, once copied (or during the copy) convert the VMDK to a thin provisioned (given that only about 85 GB of data is used in fact to this topic).
  • Finally, bring up new comments on the cluster at the same time I get comments on the old host (and ignore the changes that have taken place on the snapshot file, since its mainly a readonly report server)? and re - activate the DB update.

  For end users who would be about 5 minutes total of downtime (basically, two reboots) which would be much easier to deal with than the 5 hours (each!) of how we are past other virtual machines would.

  If the stand-alone host has the same type of CPU and family, you will be able to add it to the cluster otherwise just add to vCenter. You will always have access to other features.

• VMs migration from one to another on ESXi 5.1 data store (free license)

  I'm under ESXi in my lab (for the Cisco voice servers and a few Windows virtual machines). Virtual machines are on a datastore iSCSI hosted by a Netgear ReadyNAS Pro Business. Due to major problems with iSCSI on the ReadyNAS platform and the inability of Netgear to resolve problems (blocking NAS requiring a hard reset), I decided to buy a QNAP TS - 569 L and use it to host my ESXi data store. Now, I'm migrating my VMs from the iSCSI data store hosted on the ReadyNAS to the iSCSI data store hosted by QNAP.

  My VMs are supplied thin, and I want to preserve thin provisioning after migration. The data store Navigator ESXi can move (or copy) machines virtual data to another store, but thin-set in service virtual machines become thick-set service once they have been moved (or copied). Somebody advised Veeam to that end, I've installed and configured. Unfortunately, Veeam-error with the message that the current license does not have the migration of virtual machines from a data store to another. Just today, I learned that he must have at least the ESXi Essentials ($500) license to allow this type of migration of Veeam.

  I also tried OVFTool (the MMware command line utility), using the following syntax:

  . / ovftool - ds QNAP-iSCSI - dm = vi thin: / / vi [email protected]/CUCM-Pub: / / [email protected]

  where:

  QNAP-iSCSI is the name of the data store on the QNAP

  192.168.200.10 is the IP address of my box of ESXi that is connected to two data stores iSCSI (the one organized by the ReadyNAS and the other by the QNAP).

  CUCM-Pub is the name of the virtual machine on the ReadyNAS iSCSI data store I'm migrating to the QNAP iSCSI data store.

  However, I get the following error message:

  Error: Unexpected Option: vi: / / [email protected]/CUCM-Pub

  This is probably due to the same issue - free ESXi license does not allow this type of VM migration. So, what are my options with free ESXi license? I'm not willing to pay $500 for the license of the essential at this stage.

  Thank you!

  A few options, I can think of are:

  • export to the FVO and re - import
  • use VM explore Trilead and backup/restore virtual machines
  • Use the vmkfstools command to copy the virtual disks

  André

• HA cluster - vms migrate

  I turned on my cluster HA and off a multitude and none of the virtual machines on this host migrated to one of the other guests. Why?

  How do you turn off the host?  When entering maintenance mode or simply the pulling power?

  You have activated the admission control?  Do you have enough resources on the other host?  HA is no guarantee he will start guests on another host, but he does the best he can.

• VMs migration 3.5 ESX vSphere cluster cluster

  We need migrate approximately 100 virtual machines of the old cluster of ESX 3.5 to a cluster running vSphere. Separate environments of vCenter. What is the best way to do this? Servers in both groups use the same SAN storage. His favorite make migration as quickly as possible with downtime to do online for a longer time. As I see the options are either to present the LUN 3.5 to 4.0 servers and browse the data store and select Add to the inventory or to use the Vmware Converter and do a v2v conversion.

  Is this a correct observation? What happens when you use the option Add to the inventory? Do you need to do a lot of configuration later? I have no ESX environment to test this right now. I guess we can do a Storage Vmotion to "correct" LUNS after virtual machines, we added to the cluster 4.0?

  Converter will probably take a lot of time because the data must be copied.

  Yes is good (remember to answer I_moved the first time you turn on the computer, virtual).

  But you wrote "LUNS are not shared between guests today.

  If new guests can see the LUNS from the old host, that a simple solution (VMS must be turned off and on again after adding it to the inventory).

  André

• Thick VMs migrate NFS = thin?

  We have all our virtual machines to disk local vmfs.  They are all the VMs 'thick' right now.  If we their migration to a new NFS NetApp, are they automatically becoming thin provided?  Or when I migrate I them need to select the option to make them slim?

  I know with regular VMFS to conver them to thin.  But I've read here in the forums that NFS is configured 'automatically' thin.

  Follow-up question - how thin provisioning and NetAp ASIS Dedup works together?  Do we get the benefits of both?

  Thanks, Shane

  The underlying storage will determine what type of supply that you see on the physical storage itself.  If you create a thick or thin disk accrued on ESX, if you have a provisioning on your NFS storage server will be be thin provisioned.  ESX will think that he created a thick or a thin, depending on what you have chosen, but the physical storage it will treat it as a drive put in end use.  See here for more details http:kb.vmware.com/kb/1005418.

  You should still be able to use dedup on top end provisioning.  The two must be mutually exclusive.  Provisioning end only reports on the actually used disk blocks and blocks dedup duplicated on the recovery disk.

  -KjB

• ESXi 4 VMs migration from one host to another without HA

  We are facing a problem of automatic migration with VM ESXi migration between hosts in the cluster. Even though we have disabled HA and no reason apparent for migration. We have two virtual machines residing on a host, and the host does not work normally. However, I have noticed that when we have turned off or the other virtual machines and once again, he climbs to another host in the cluster.

  The host tasks log shows now that the migration has been made by the SYSTEM when the power on. And once one of the virtual machines migrated even without power and with a similar journal entry showing that this has been done by the system. This can happen due to a shortage of resources on the host computer even if HA is disabled or are there other reasons that this can happen?

  Hello

  I'm sure that's the reason for your behavior described.

  Concerning

  Andi

• Turn on CVS on existing cluster?

  I have a few questions about the activation of the VCA:

  1. is it correct that if a cluster is already existing and run several virtual machines I would be able to their or move them to activate CVS?

  2. is it true that if I have few hosts I could vMotion virtual machines on the hosts in the cluster and when empty to VCA and your choice?

  As in, if the cluster DRS has many configuration with resource pools and others I want to keep.

  3. is it always necessary to empty the cluster to enable EVC, even if the definition of the VCA operating at a level that all guests can currently use?

  4. If there is a cluster with, say, five hosts, would - it possible first to break two hosts of the cluster and vMotion all virtual machines for them, enable CVS, back the VMs and then join the other two hosts on the cluster? If do not and so rush hour now. The principle would work?

  > But in realitity then DRS depends on vMotion to work and vMotion requests CPU compability, so to really run DRS the CPU must match levels throughout the cluster, is that correct?

  Not really. If they do not match, you will be able to vMotion virtual computers between hosts that match. But if you want to be able to move to any host cluster VM - then Yes.

  > And if all processors match today, so there is no problem to activate CVS while all running virtual machines and then adding a host of new?

  No problem now. But before an update (May 4.0u1, remember) there is a strict 'No running VMs' requirement for VCA.

  > Will not take any downtime for virtual machines or need to migrate far away?

  N ° of course, if you want to enable CVS at the level that corresponds to current processors.

  > Is VCA a new feature in 4.0 or there in previous editions of ESX/ESXi?

  3.5 updated something I think.

  ---

  MCSA, MCTS Hyper-V, VCP 3/4, VMware vExpert

  http://blog.vadmin.ru

• Dynamic Cluster migration "Haswell" Intel and Intel 'Sandy Bridge' Cluster

  Hi all

  Need your help on this, on our below, we have two clusters ClusterA (EVC mode is Intel Haswell) and focus (EVC mode is Intel Sandy Bridge), the current problem we are facing right now is that we can not migrate from VM direct of clusterA to b, but successful once we stop the virtual machine and migrate to b. Webcam live migration of VM of focus to ClusterA now question. My question is, is there a way that we can migrate from VM direct ClusterA to b without changing the mode of the VCA on the two hosts?

  Details of home:

  Group A:
  CPU: Intel Xeon CPU E5-2698 V3 2.30 Ghz
  RAM: 256 GB
  EVC: Intel "Haswell" generation

  
  

  Group b:

  CPU: Intel Xeon CPU 2.6 Ghz E5-2670

  RAM: 256 GB

  EVC: Intel 'Sandy Bridge' generation

  If a processor can support level VCA XN, it can also support XN-1 to Xlevels 0. For example, a processor that supports Intel EVC baseline 'Sandy Bridge' generation has a level of EVC of L4. Therefore, it can also support VCA L0, L1, L3 and L2 levels. However, he cannot bear VCA level L5, which corresponds to the generation of Intel®'s "Ivy Bridge'. Intel EVC baselines are listed in table 1.1.

  Ref: KB VMware: best vMotion compatibility (EVC) processor support

  Thus, virtual machines that are running in your focus (EVC mode is Intel Sandy Bridge), should be able to move freely (vMotion) to b (mode of EVC is Intel Sandy Bridge) to ClusterA (EVC mode is Intel Haswell) and A to B provided that the virtual machine that is in your Cluster B vMotioned while in Group A and A is not in the power Cycle, if you put it off and on again then it will be new pickup EVC mode, otherwise it will continue to work with his original mode of VCA.

  But if you take the case of the virtual machines running within your group, those who can not be vMotioned to group B, because they are running with mode higher CVS, canoe taken to lower VCA as live migration, you can migrate cold (power off and move)

  Ideally if you want to reach vMotion compatibility between the two clusters, you must set the VCA mode for each group for Intel Sandy Bridge. But like asking you if you won't change EVC mode at all, the solution proposed would be

  Turned off of the virtual machines in your A group phase by phase, turn cold migration from Group B, there and let the fashion of these machines virtual micro Cluster B EVC, then move them again in A Cluster, in short to realize all the VM in EVC of Cluster B mode. But you will always be in the face of challenge then if VM through the feeding cycle while in Group A, it will pickup Haswell VCA mode and you will lose compatibility vMotion for this virtual machine on cluster again.

  so, what you're asking can be done using method above, but better would be to have two clusters in the VCA mode even if you don't have to worry about power cycles.

  Note: You need to use vSphere Web Client.

• Cannot migrate between identical hosts in the same cluster - CPUID level 0 x 1 'ecx' inconsistency

  I have problems of migration of virtual machines between hosts that are identical. They are the new BL 6 490 HP servers with a pair of processors x 5570 @ 2.93 Ghz and I activated the "Intel Xeon 45nm Core 2" EVC mode, there is also two other cluster hosts (HP BL480 G1 with a pair of processors E5430 @ 2 .66GHz).

  I ESX4, vCenter4 and all the latest firmware on all devices.

  When I try to migrate an identical to another or to other hosts host machines, I get the following error;

  Host CPU is not compatible with the requirements of the virtual machine to CPUID level 0 x 1 'ecx' register

  Bits of the host: 0000:0000:0000:0000:0010:0010:0000:0001

  Required: 1000:0000:0000:000 x: xxx0:0x1x:xxx0:x 001

  Inconsistency detected for these features:

  • General incompatibilities; see section 1993 of possible solutions.

  I have consulted the Article KB http://kb.vmware.com/kb/1993 and the document referred to Sub http://kb.vmware.com/kb/1008315

  All affected machines are hardware VM 7, the last VMTools are installed, have not RDM or nonstandard hardware, there is no floppy disk or CD drive mapped drive

  I tried to change such CPUID mask specified in the document 1008315, but this did not help. I also tried to change the mask of CPUID for ecx to be 0-

  ---

  -

  ---

  -

  ---

  -

  ---

  -

  ---

  -

  ---

  -

  ---

  and that did not help.

  I turned off a VM to ensure that he had picked up the CVS settings. I am able to vMotion, virtual machine when it is off, but this is not enough as we have several "stuck" on the same host machine and cannot be moved by the DRS or manually.

  Virtual machines are Windows 2003 (R2, SP2, x 32). worked well before vMotion on all hosts. Other identical machines from the same model are fast vMotioning very well, and when we were on ESX 3.5u2 wasn't a problem.

  Can someone advise me on how I can diagnose and fix this problem?

  I had the same problem after migrating from ESX 3.5 to V4.0 with almost the same HW-Configuration you have. (BL490c and BL480c). But in my environment, the article you mentioned has brought the solution:

  To manually remove the CPUID masks using the VI Client:

  1 power off the virtual machine.

  2 - Click on the VM, then click on change settings.

  3. click on the Options tab.

  4. Select CPUID mask under Advanced.

  5. click on advanced.

  6. click Reset all to default.

  7. click on OK.

  8. click on OK.

  9. start the virtual machine and migrate.

  Have you actually tried it? The default mask is OK, there is no problem with it in your environment, as I have said already, I have the same hardware as you.

  Paul