Backup Of volume before doing reverse-resync or resync to avoid data loss

[TechiS]

Dear Team,

We have often been doing  NAS volumes reverse resync tasks, but we need to to know what is recovery option available if some does resync instead of reverse resync?

If resync is performed instead of reverse resync then data will be over written resulting in data loss and there is no option to recover that data.

Thanks

[Sergey_Osipov]

Possible recovery options are external ndmp backups, fan-out\cascade snapmirror.

Bu generally you should be aware that resync is overwriting the destination of resync and there is a warning message

[TechiS]

Thanks for the reply! I have been successfully doing reverse resync since past 6 years without any issue and i know the difference between reverse resync and resync. But wanted to know what if someone does it wrongly. I had also given feedback to NetApp Team to avoid keeping both option of reverse resync and resync just next to each other but know success.

I was thinking about having clone backup of volume before doing reverse resync and resync, but what is fan-out\cascade snapmirror?

Thanks

[Ontapforrum]

Following articles covers 'cascade' concept: 
https://docs.netapp.com/us-en/ontap/data-protection/supported-deployment-config-concept.html#how-fan-out-deployments-work
https://kb.netapp.com/onprem/ontap/dp/SnapMirror/How_to_configure_SnapMirror_cascade_relationship

[TechiS]

We have setup as below

DR <---30 minutes Schedule----<DC>-----30 Minutes Schedule-----> NDR

So do you mean to say, If we are LIVE at DC side and want to reverse resync from  DC to DR, we should first  do snapmirror update from DC to NDR manually , so that we can have additional copy from DC to NDR so even if in worst case if we do resync from DR to DC and overwrite contents of DC with DR, we will still have additional copy of DC at NDR correct?

Answer:

Is this called cascade deployments?

Thanks

[Ontapforrum]

I am only confused with terms and your setup. But, your understanding of the fact that you have another copy of the destination volume as Territory copy and you can use that to recover if anything happens to source/dest is correct.

Normally, I would assume, this way:

Resync

Source                   Dest.
DC (live) -------> DR

Reverse-resync

Source                  Dest.
DC (live) <------- DR

Cascade is basically:
DC                                                DR                                        TC
Source (DC:DR) ----> destination ----> (DR:TC) Territory copy

[TechiS]

Thanks for the reply, this seems to be just solution for one site either DC or DR, for both sites I think below will work

NDR <------------<DR<-----------------------<DC (LIVE Site)>--------------->NDR

NDR <------------<DR (LIVE Site)>----------------------->DC >--------------->NDR

Kindly note here NDR is Same NetApp Cluster  AFF 700.

In this case before doing resync or reverse resync from DC or DR site

we first just need to update  NDR volumes from DC or DR site depending on from where we are doing resync or reverse resync operation DC or DR

so even if we wrongly do reverse resync or resync we will still have copy of data at NDR.

Please let me , if I am correct or not?

Thanks

[TechiS]

Dear All,

Greetings for the day!

Could you please update on this.

Thanks