It comes from reallocate not being fully effective previous to 8.1, but not because it doesn't work.  

I never used measure to know when or why to run.   I look at cp-read to write io ratios in statit.    Wafl free space health is the issue not data locality IMHO.  

Sent from my iPhone

On Jan 18, 2013, at 7:16 AM, Fletcher Cocquyt <fcocquyt@stanford.edu> wrote:

Back in Oct 2011 we identified several volumes recommended for reallocation  (by reallocate measure), and opened a case to learn the official risks.
They ended up citing TR doc recommending no reallocate for deduped volumes

Now I read 8.1 changes this recommendation?


I am running some more reallocate measures to see if it could help our current latency spike case where we are seeing outlier busy disks.

thanks

On Jan 17, 2013, at 9:35 PM, Jeff Mohler <speedtoys.racing@gmail.com> wrote:

It's not as bad as you think.  

You can't reallocate deduped blocks in a volume.  Not 'the whole volume'.  

If you have a 50% deduped volume, then you can still reorder half the blocks.  

Which may not be so bad

Sent from my iPhone

On Jan 18, 2013, at 3:50 AM, "Matt Kilham (Stratton Finance)" <matt.kilham@strattonfinance.com.au> wrote:

Hi Brad,


We faced a similar situation a while back, wanting to add just a couple of extra disks to an existing RAID group on a FAS2050 running 7.3.6.

After emails to this list, several conversations with NetApp support, etc, the answer we received was that you simply can't reallocate deduplicated data in v7.x - there's no way to do it short of manually moving data out of and back I to the aggregate.

Rather unfortunate - in our case we simply had to abandon the plan to add these extra disks to our RG as we would've ended up with a major hotspot.



--

Matt

On 18/01/2013, at 6:04 AM, Brad Smith <bsmith@iup.edu> wrote:


Looking for help/direction on reallocating deduplicated volumes. Have a FAS2040 running 7.3.6 and adding an additional DS4243 shelf with SAS drives and want to extend current aggregate by adding a new raid group created with the new disk to the aggregate.  On this aggregate I have volumes that are deduplicated and have snapshots. I have read several NetApp documents that seem to conflict on running reallocate on the volumes after adding the new disks to the existing aggregate.  I have read you should reallocate but not on a volume that is deduplicated.  I have also read that if you use the -p switch this will be OK,  reallocate start -f -p /vol/VOLUMENAME.

Will I have problems if I run reallocate on my deduplicated volumes?

Thanks.

Brad
_______________________________________________
Toasters mailing list
Toasters@teaparty.net
http://www.teaparty.net/mailman/listinfo/toasters


_______________________________________________
Toasters mailing list
Toasters@teaparty.net
http://www.teaparty.net/mailman/listinfo/toasters

_______________________________________________
Toasters mailing list
Toasters@teaparty.net
http://www.teaparty.net/mailman/listinfo/toasters