ESXi

 View Only
Expand all | Collapse all

Need answers for 2 interview questions

  • 1.  Need answers for 2 interview questions

    Posted Mar 30, 2015 08:38 PM

    Hi Friends, Need answers for 2 interview questions

    1. Virtual Machine is not manageable / connectable after successful vMotion. what is the reason ?

    2. A company has 2 ESXi hosts, if one of the ESXi host goes down, the VMs from the failed ESXi host is not moving to the live ESXi host. what may be the reasons.

    Regards,

    Sathish Vijay



  • 2.  RE: Need answers for 2 interview questions

    Posted Mar 30, 2015 09:32 PM

    1. as per my exp i seen below issues

    (a) Check out the NIC configuration.may be nic is not connected.

    (b) sufficient resource are available to run the vm.

    (c) CPU utilization must be check. it can be jumped after vmotion.

    2. Check out if your HA is configured with Strict Admission control. its the reason sometime ha doesnt work if host goes down or you keep any host in maint. mode.

    Have a look on it. : http://www.vmwareminds.com/troubleshooting-maintenance-mode-process-get-stuck-at-2/



  • 3.  RE: Need answers for 2 interview questions

    Posted Mar 31, 2015 04:19 AM

    Thank you Kumar for sharing your experience.

    Friends, can anyone please share your views...



  • 4.  RE: Need answers for 2 interview questions

    Posted Mar 31, 2015 07:09 AM

    Anjan added good points and let me give my view about interesting questions:

    1. Virtual Machine is not manageable / connectable after successful vMotion. what is the reason ?


    Question says successful vMotion - which means all resource checks for compute, storage & network are performed

    However - there is possibility of couple of cases

    If you notice back ground process correctly only Computer part is transferred. So reasons are

    1)  Failed to perform clean transfer of memory pages  

    2)  Transfer interrupted by Network congestion 

    3)  Storage may stopped responding after migration

    4)  Network features causing VM unavailability - VMware KB: A virtual machine loses network connectivity after vMotion

    • The Virtual Machine Memory state is copied over the Vmotion Network from the source Host to the Target Host. users continue to access the virtual machine and potentially update pages in memory. A list of modified pages in memory is kept in a memory Bitmap on the source Host.
    • After most of the virtual machine memory is copied from the source host to target host the virtual machine quiesced no additional activity occurs on the virtual machine. In quiesce period VMOTION transfers the virtual machine device state and memory Bitmap to the destination Host.
    • Immediately after the virtual machine is quiesced on the source host, the virtual machine initialized and starts running on the target host.
    • users access the virtual machine on the target host instead of the source host.
    • The memory pages that the virtual machine was using on the source host are marked as free.

    HA:   Based on your configuration this will have different possible reasons

    1)  Network Isolation - so each host thinking they are isolated

    2)  Good design always starts with 3 hosts but if this is special case - having agent issues

    3)  Agent corruption

    4)  If you set VM isolation response is power-off

    5)  Storage is not properly configured (not shared)

    ... etc



  • 5.  RE: Need answers for 2 interview questions

    Posted Mar 31, 2015 12:22 PM

    Hi Satheesh,

    As this is a Successful vMotion,  Point 1 & 2 are not sensible.

    Point 1) Failed to perform clear transfer of Memory Pages

    Point 2) Transfer completed by Network congestion

    Also as you said, as this is a successful vmotion, all resource checks for compute, storage & network perfomed.

    Hope so..



  • 6.  RE: Need answers for 2 interview questions

    Posted Apr 01, 2015 09:44 AM

    Thanks for your feedback :smileyhappy:



  • 7.  RE: Need answers for 2 interview questions

    Posted Apr 01, 2015 04:06 PM

    Hi,

    My  thoughts to your questions:

    1) The VM has been moved to a host with incorrectly configured port group/networking. I would suggest checking the configuration on both hosts to compare and troubleshoot.

    2) I thought of the following:

    - HA is not configured correctly (settings, licensed, etc?)
    - Possible DRS Rules for "MUST RUN ON X HOST"?

    - VM's do not live on shared storage visible by both hosts?

    - VM has been disabled in the VM Monitoring status of HA cluster settings.

    - Incorrect networking configuration (DNS, IP addressing, etc?)


    This is just what I jumped too when reading your questions. Don't know if it is of any use or you have the answers?

    Regards,
    Ryan



  • 8.  RE: Need answers for 2 interview questions

    Posted Apr 07, 2015 11:56 AM

    Hi Ryan, Sorry for the delay.

    Appreciate if you could elaborate it in detail.

    1. WIll the VM get migrated to a host with incorrectly configured port group. (In what aspect port group may be incorrectly configured)

    2. will VM not get migrated if it is disabled in VM Monitoring of HA cluster. is it so ?

    Warm Regards

    Sathish Vijay




  • 9.  RE: Need answers for 2 interview questions
    Best Answer

    Posted Apr 09, 2015 09:17 AM

    Hi,

    To confirm:

    1) If you have a port group that exists on both hosts but has been incorrectly configured on a specific host, that could cause the problem. This is the problem with standard switches and port groups as they are configured individually per host, which allows room for human error. For example, you could have the port group with an incorrect VLAN ID assigned or the switch it is on could have an uplink that is different to the previous host. It is also possible that the port group might not be configured on the host at all, it is possible to be overridden.

    2) You can select the VM behavior in the event of an HA occurring. If you select the cluster and view the vSphere HA features, you have "Virtual Machine Options".  Here it is possible to change the VM Restart Priority (it could be disabled) and also how the VM behaves if the host is isolated from the network. These settings could have possibly prevented the VM from restarting during HA failover.

    I hope that makes sense!

    Regards,


    Ryan



  • 10.  RE: Need answers for 2 interview questions

    Posted Apr 09, 2015 10:14 AM

    Hi Ryan and friends.

    Need your help on these 2 questions as well.

    1. Need of Pre and Post requistics for an ESXi upgrade from 4.1 version to 5.1 or 5.5.

    2. What needs to be more concentrated while installing Vcenter Server and how many SQL databases are required for this and how many SQL instances need to be created.

    3. Which one will work inn the absence of Vcenter server - 1.DRS, 2.HA, 3.FT

    4. Which one will support SNAPSHOT .- 1. Physical RDM 2. Virtual RDM.

    Sathish Vijay.



  • 11.  RE: Need answers for 2 interview questions

    Posted Apr 10, 2015 03:37 PM

    Hi,

    1) I'm not too sure what you mean. I guess the pre and post steps to upgrade from 4.1 to 5? In which case this article from VMware explains it.

    2) The windows vCenter only needs one SQL database. So only 1 database on 1 instance is fine. Many people have a dedicated SQL server with a VMware Update Manager SQL database on the same SQL Server instance too!

    3) HA and FT (if already configured) will continue to work if vCenter is offline. DRS will not.

    4) Virtual machine snapshots ARE supported with Virtual RDM , NOT Physical.

    Regards,


    Ryan