Hi Thomas, we've had the same problem and have written such a tool and are in the process of cleaning our pools for quite some time now. it's quite a mess. I'll contact you off-list later.
-Arne On 16/08/16 21:18, Thomas Caputi wrote:
Hi all, My company has a bunch of replicated zpools that were affected by the Illumos 6370 hole_birth send bug. These pools contain a lot of data and it has been deemed too expensive to resend everything. I am currently planning on writing a little program to edit a zfs send file to inject the needed holes so that we can resend the last few snapshots, but I was wondering if there was a possibility of writing holes to a received zfs dataset without preventing future receives on that dataset since that would probably be quite a bit cleaner to implement. Does anyone have any thoughts on how I could accomplish this? Thanks
------------------------------------------- openzfs-developer Archives: https://www.listbox.com/member/archive/274414/=now RSS Feed: https://www.listbox.com/member/archive/rss/274414/28015062-cce53afa Modify Your Subscription: https://www.listbox.com/member/?member_id=28015062&id_secret=28015062-f966d51c Powered by Listbox: http://www.listbox.com
