So what I've ended up doing is just re-snapmirror to tape, FedEx, load into a new volume. Now I need to switch all my clients to the new volume. Easier said than done, since a bunch of long running user processes are doing to slow this all down.

Then once that's all done, I then blow away the old volume and re-create it, re-snapmirror it, and then switch everyone back. Mostly because I want to keep the volume in the same aggr as before, but I didn't quite have room for a second tools volume. Bummers.

Luckily, it's a low churn volume, so it should be ok.

Joel> Wonder if you can put the source volume into restricted mode Joel> first and migrate will the open read or write simply continue if Joel> the FSID are identical. It should but I have never tried it.

Joel> Ed Wilts wrote:

...

On Wed, Dec 9, 2009 at 7:56 AM, Joel Krajden <joelk@encs.concordia.ca mailto:joelk@encs.concordia.ca> wrote:

You might want to look at

Solution ID: kb40272 Last updated: 30 SEP 2009

Sample procedure using ' snapmirror migrate ' that avoids remounting in a NFS environment

Unfortunately, it only works if all NFS access has stopped. If you have just one host out there that you forgot about that is doing I/O to the volume, it's remount time and users grumbling about stale handles.

.../Ed

Ed Wilts, RHCE, BCFP, BCSD, SCSP, SCSE ewilts@ewilts.org mailto:ewilts@ewilts.org Contact Me Linkedin http://www.linkedin.com/in/ewilts

Joel> -- Joel> | Joel Krajden | Rm: EV-7105, Tel: 514 848-2424 3052 | Joel> | Core Technologies Mgr | | Joel> | Engineering & | Email: joelk@encs.concordia.ca | Joel> | Computer Science | | Joel> | Concordia University | In a circus, the clowns are supposed | Joel> | Montreal, Canada | to make you laugh, not cry. |