Thanks Greg for the quick respond and cc'ing me.

Hi Tanzeeb, great to talk with you.
Please allow me to comment inline and I hope it will clear things out for you.
Basically the solution we are discussing about is the site-so-site solution for oVirt-DR
The official documentation for it is at 
[1] https://access.redhat.com/documentation/en-us/red_hat_virtualization/4.2/pdf/disaster_recovery_guide/Red_Hat_Virtualization-4.2-Disaster_Recovery_Guide-en-US.pdf

I recomment that you take a look into that documentation since it might answer many of your questions.

Please feel free to contact me on anything that is still unclear

Regards,
Maor

On Sat, Jul 7, 2018 at 12:38 AM, Greg Sheremeta <gshereme@redhat.com> wrote:
Hi Tanzeeb,

Unfortunately googling for 'ovirt disaster recovery' doesn't achieve a great result. Search for things developed primarily by Maor [cc'd] recently.

On Fri, Jul 6, 2018 at 12:59 PM <samee095@gmail.com> wrote:
Hi
I've looking for some ideas about designing disaster recovery planning with ovirt 4.2. Since 4.2 we have options to have intergration with disaster recovery site also. I went through some videos at internet and want to share if those are correct or not. Thus you can help me regarding what other things I should be having ideas during planning and designing more about this.

1. We've to keep both sites on and already created datacenter, cluster and one site with storage domain attached and other site with no storage domain. 

That is correct
 
2. We've to keep latency of 10ms at maximum between both sites. 

That depends on your replication configuration process, sounds good to me.
 
3. Have to configure virtual machines with affinity group.

You don't have to do that to support the DR site-to-site
 
4. The VM's which are configured as high-available vm, will be migrated first.

Not exactly, all the VMs will be migrated, regardless whether those are HA or not, the HA VMs will run first (if those were running before on the primary setup)
 
5. There's an Ansible script while have to run to fail over and fail back.

Indeed, and also a python script that make it more user friendly.
 

Now, please could you help me out regarding this.
1. Where can I find this ansible script? Is this https://github.com/oVirt/ovirt-ansible-disaster-recovery.git?

yes. Note the youtube links on that.

@Maor for the rest of your questions.
 
2. Can this script run at rhv also ? or just for ovirt? Or what other changes I have to make for rhv to have this functionalities?

Theoratically, the script should also run on RHV as well, but I would not recommend to mix between the two in one setup.
 
3. How ovirt/rhv works with the OVN compared to NSX? Is there's any challenges ?

I don't think that is related to the oVirt DR solution.
@Dan perhaps you can shed more light on that subject.
 
4. Can you share me some high and low level diagram related to rhv disaster planning?

Sure, I suggest that you look into the following documentation (see [1]) at Figure 3.1. Active-Passive Configuration and Figuire 3.2:

I think that what is more important for you are the steps to prepare your setup to support DR, this should also be in the documentation mentioned above.

5. How the storage migrates on backend ? During the script ?


The storage is not migrated, it should be replicated on the secondary site's storage server and the secondary site should import those storage domains.
The storage replication is managed by the admin.
 
6. Where do I run this script? During fail over at primary site and during fail back at secondary site?

The script can run on any host, while that host has python-oVirt SDK installed (see https://www.ovirt.org/develop/release-management/features/infra/python-sdk/).
You can also run this script on the primary site or the secondary site.
My suggestion is that you will have a seperate machine/VM/container which will be the the one which will run it.
 
7. Basically since only the storage migrates and all other components are on already so basically backend it's only dealing with storage migration according to defined policies ?

The storage domains does not get migrated, they are being replicated.
It is a DR solution based on the storage servers, since those contain the OVF_STORE disk which should help to recover the VMs/Templates in the setup.

 

Please could you help me out with some designing and ideas what're the best architecture during planning on disaster recovery at RHV.

I suggest that you take a look at [1].
There should be an explanation of the steps and also several usecases how to test it (see APPENDIX B. TESTING THE ACTIVE-PASSIVE CONFIGURATION)
 

If you are a Red Hat customer, a Red Hat Consulting engagement could also help.
 
Thank you.

_______________________________________________
Users mailing list -- users@ovirt.org
To unsubscribe send an email to users-leave@ovirt.org
Privacy Statement: https://www.ovirt.org/site/privacy-policy/
oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/3BNWV3MGHYMAP2PQG2ZSASUIFBAENUM7/


--

GREG SHEREMETA

SENIOR SOFTWARE ENGINEER - TEAM LEAD - RHV UX

Red Hat NA

gshereme@redhat.com    IRC: gshereme


_______________________________________________
Users mailing list -- users@ovirt.org
To unsubscribe send an email to users-leave@ovirt.org
Privacy Statement: https://www.ovirt.org/site/privacy-policy/
oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/
List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/5OLXCN3QLNJDIVAYLI7UDOAC4B23NRRQ/



[1] https://access.redhat.com/documentation/en-us/red_hat_virtualization/4.2/pdf/disaster_recovery_guide/Red_Hat_Virtualization-4.2-Disaster_Recovery_Guide-en-US.pdf

Regards,
Maor