Ben0it

updated

7 years ago

B

Ben0it

3 Posts

0

3847

August 1st, 2019 00:00

High availability failover clustering with 2xR640 + 2xSCv3020 + Windows Datacenter 2019

Hello, I am currently setting up this hardware, each half being located on the same site but in a separate building, roughly 350 meters apart from each others.

My intend is to run manufacturing application running in virtual machines (Hyper-V) and provide high availability with automatic switchover in case of single-component failure.

It is my first experience with this kind of hardware, things are currently setup as follows :

Based on this design, could someone do some recommendations and/or corrections if I did anything wrong ?

Best regards,

Benoît.

  • DELL-Sam L

    Community Manager

    •

    8147 Posts

    •

    34706 Points

    3534

    0

    Posted August 5th, 2019 11:00

    Hello Ben0it,

    Here is the link to the Administrator guide for your SCv3020 & it explains more how to configure your replication on page 520.

    https://topics-cdn.dell.com/pdf/dell-storage-manager_administrator-guide3_en-us.pdf

    Please let us know if you have any other questions.

  • DELL-Sam L

    Community Manager

    •

    8147 Posts

    •

    34706 Points

    3575

    0

    Posted August 1st, 2019 15:00

    Hello Ben0it,

    From what I can tell you have setup your SCv3020 as shown in our deployment guide.  I am not seeing any issues with how your setup has been done.

    Please let us know if you have any other questions.

  • Ben0it

    3 Posts

    3563

    0

    Posted August 2nd, 2019 03:00

    Hello DELL-Sam L, thanks for your review.

    I have another question indeed, related to automatic failover mode of the Windows Datacenter 2019 cluster :

    Suppose I create a live Live Volume with synchronous replication between my two SCv3020 for a virtual machine running on the cluster.

    How shall this Live volume be mapped on the two servers ?

    Best Regards,

    Benoît.

  • Ben0it

    3 Posts

    3498

    0

    Posted August 9th, 2019 09:00

    Hello Dell-Sam L, thanks for the link.

    I am still reading the guide, while I'm thinking about the following scenarios and how a cluster + disk area recovers :

    • scenario 1 : one of the server of the cluster fails and the other server takes over.
    • scenario 2 - one of the disk area fails and the other disk area takes over.
    • scenario 3 :  - one server and disk area of in the same server room become unavailable, the other pair in the other server room takes over.
    • scenario 4 :  - one disk area in server room 1 fails, while the server in server room 2 fails, the remaining live parts take over.

    Do you know if each of the 4 cases is recoverable ?

    Best Regards,

    Benoît.