On Mon, 22 Oct 2012 16:20:05 +0200 (CEST), you wrote:

>Hi,
>we have a software we made to automate zfs replica of pools into separate 
>slow/high-capacity disks, for backup.
>When the pool is single, we have no problem automating the zfs send/receive 
>into a new pool of the destination disk, because the destination is the entire 
>disk.
>Sometimes we would like to automate the sending of more pools, including the 
>rpool, to have a single
>disk replicating an entire system.
>We use to prepare the destination disk with format, creating correct slices to 
>contain all the pools into
>s0,s1 ecc.
>We would like to be able to automatically look into a destination disk, 
>discover Solaris format and usable slice sizes, decide if they fit the source 
>data and then if to use them.
>Or, decide if to destroy partition table and change slices to fit what we need.
>Problem is, prtvtoc, format and so on prdocues quite hard output to parse and 
>decide.
>Is there any other simpler tool that already output usable slices and sizes?
>And any other simpler tool to create the needed slices?
>Last but not least: am I forced to use Solaris format and slices in this case? 
>Or is there any other
>partitioning scheme I could use to create more pools into a single disk?

This simplest tool is ZFS itself. There is no need to make a target pool
per source pool. Just create one backup pool on full disk(s) and create
use that as a target to receive all datasets of all source pools.


Example (this is what I do):

Source pools:

rpool, with datasets:
        rpool/ROOT
        rpool/ROOT/BE1
        rpool/ROOT/BE1/var
        rpool/dump
        rpool/export
        rpool/export/home
        rpool/swap

datapool1
        datapool1/dataset1
        datapool1/dataset2
datapool2
        datapool2/dataset3
        datapool2/dataset4

Target pool:

zback, with datasets 

        zback/zsr/rpool/ROOT
        zback/zsr/rpool/ROOT/BE1
        zback/zsr/rpool/ROOT/BE1/var
        zback/zsr/rpool/export
        zback/zsr/rpool/export/home
        zback/zsr/datapool1/dataset1
        zback/zsr/datapool1/dataset2
        zback/zsr/datapool2/dataset3
        zback/zsr/datapool2/dataset4
        zback/other

zback/zsr indicates all datasets below it receive their contents from
zfs Send/Receive, hence zsr.

The last dataset, zback/other is used to put random stuff in, not
related to datasets in the source pools.

>Thanx for any clarification.
>Gabriele.

-- 
Groet, Cordialement, Pozdrawiam, Regards,

Kees Nuyt



-------------------------------------------
illumos-discuss
Archives: https://www.listbox.com/member/archive/182180/=now
RSS Feed: https://www.listbox.com/member/archive/rss/182180/21175430-2e6923be
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21175430&id_secret=21175430-6a77cda4
Powered by Listbox: http://www.listbox.com

Reply via email to