Leal,

> These are the steps that i'm following to switch back a resource  
> managed by the MRSL.NONsharedDevice agent:
>
> Scenario: Did occur a failover (the system is in logging mode), the  
> resource is running on the secondary node.
>
> Steps (commands):
> 1 - Bring the resourcegroup offline (clresourcegroup offline pool-rg)
> 2 - Do a reverse/update sync (sndradm -C local -g POOL -n -r -u)
> 3 - Here we have two options using the NONsharedDevice agent:
>       a) Put the AVS software in logging mode again (sndradm -C  
> local -g POOL -n -l) or
>       b) Set the Last_Started extension to the "primary" node  
> hostname (clresource set -p Last_started=primarynodename pool-rs)
> 4 - Bring the resourcegroup online on the primary node  
> (clresourcegroup online -n primarynodename -M pool-rg)
>
> The question is:
> While the systems are resync the services are unavailable, the good  
> point is that is a switch back scenario, so we can schedule it. But  
> would be nice if i could make a reverse/sync while still using the  
> resourcegroup. I mean, if there is a lot of data to resync, can be a  
> long, long maintenance time.

Once the "reverse/update sync" has been invoked, SNDR will start the  
background process of resynchronizing the secondary volume to the  
primary volume. While this activity is underway, access to the SNDR  
primary volume can be resumed,  and thus services can be made  
available through a replication feature called on-demand pull.

What this means is that if a block is accessed on the SNDR primary,  
but it has yet to be resychronized, an on-demand request will be made  
to fetch the block. Be forewarned, that when running in this mode,  
performance will be less, maybe much less than running SNDR in  
synchronous replication mode, since the background resynchronizing and  
the on-demand pull will be competing for the same resources.

This mode of operation is based on the assumption that only a small  
percentage of a volumes recently changed data is needed to continue  
needed I/O services.

Jim



>
>
> Thanks very much for your time!
>
> Leal.
>
>
> This message posted from opensolaris.org
> _______________________________________________
> storage-discuss mailing list
> [email protected]
> http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Jim Dunham
Storage Platform Software Group
Sun Microsystems, Inc.


http://blogs.sun.com/avs
http://www.opensolaris.org/os/project/avs/
http://www.opensolaris.org/os/project/iscsitgt/
http://www.opensolaris.org/os/community/storage/

_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to