Subscribe

Volume SnapMirror Issues with Disk Space

[ Edited ]

Hi All

 

I hope someone could steer me in the right direction with a volume snapmirror issue we currently have as I'm relatively new to NetApp SAN's?

 

We have an older setup of 2 x NetApp FAS2020 SAN that we are currently migrating data from to our newer 2 x NetApp FAS2240-4.  Mostly we are doing okay using snapmirror's to copy the volumes over to the new system although one critical volume (our Exchange Mail store) is causing us problems.  Please see below the current size allocation of the volume:

 

VolExchange - 614.15 GB Used (Total Volume Size=1TB)

 

So around 385 GB free

 

When we try and snapmirror the volume to the new SAN it breaks our Exhange.  We suspect this because when the snapmirror happens the snapshot created on the source volume doesn't have enough space to complete and obviously the volume (VolExchange) runs out of space and Exchange errors.

 

Can someone please advise me if my understanding is correct (cause of disk usage with snapshot in same volume)?

And if so what advise can anyone give me to resolve this space issue during a snapmirror to our newer system to ensure completion?

 

Please any recommendations are very welcome as we must migrate this volume as soon as posible due to the older SAN is becoming unstable and problems with hardware too.

 

Many thanks

 

Mark

 

Re: Volume SnapMirror Issues with Disk Space

Not sure if the reason you stated is the reason this fails.

 

you are right in the process, SnapMirror creates a snapshot and replicates that point in time copy of the volume, so the volume will grow as it keeps any changed blocks, however i would say that there would be a lot of change data (although i suppose it depends how long the initialisation is taking).

 

do you know that the volume fills and offlines breaking exchange.

 

what happens at the server level with this?

 

what you are trying to do, should work with very little problem, if the vol is filling, however just make the volume bigger.

 

but I would be tempted to look elsewhere, unless you know for definite that the volume offlines during this process.