HPE Storage Users Group
https://3parug.com/

UNMAP - > RCG .r VLUN disk not shrinking? / Total reserved
https://3parug.com/viewtopic.php?f=18&t=3260
Page 1 of 2

Author:  MBILC [ Mon Aug 05, 2019 11:00 am ]
Post subject:  UNMAP - > RCG .r VLUN disk not shrinking? / Total reserved

Hello,

Having to do some long over due maintenance on virtual volumes. I need to do this to free up some space to create new VMFS6 datastores for our front end.

I have been running UNMAP on the existing VMFS5 datastores to free up space. I have had great luck with some volumes that were at 80% plus, down to 30-40% Total used.

My issue is, it appears that the remote copy group volumes on our other 3PAR are not "updating" or shrinking now that the master volume has dropped. I presumed that the shrinking would replicated over to the .r volume?

I have run compactcpg -f -pat * several times on both systems, but so far no luck? Does only a full Tunesys do this, or do i need to do something else with the RCG's to get the *.r virtual volumes to shrink down in usage? (stop the RCG, delete and let it resync?)

Also, after running UNAMP, it seems the Total Reserved is also not dropping after a compactcpg -f -pat * is run.

Any help would be appreciated, some tier's are getting tight on storage making it difficult for me to build out a new storage setup.

Author:  MammaGutt [ Mon Aug 05, 2019 3:05 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

Lots of questions here...

Firstly, tunesys will never change the stats for a single VV, so that will not do anything.

Secondly, compactcpg will never change the stats for a single VV so that will not do anything (expect release free space in the CPG to the system wide pool).

What are you looking at when talking about volumes reducing from 80% to 30%? Usr Used? Usr Rsvd?

Next question that will set the stage for furter questions.

What 3PAR OS version, what types of volumes and what type of replication?

Author:  MBILC [ Mon Aug 05, 2019 4:24 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

Thank you again for responding.

Lets see if i can give the info required.

Under Virtual Volume / Capacity, example volume:

vm_001 (3PAR01)

Capacity Utilization
48% Total used
Total used
1,945.84 GiB
Virtual size
4,096.00 GiB
Total reserved
3,430.13 GiB
Host written
0.00 GiB


vm_001 is in a remote copy group which Source System is 3PAR01. The 2nd system is 3PAR02.

Name 3PAR01_3PAR02
Configuration 3PAR01, 3PAR01
Mode Sync
Source system 3PAR01
Source role Primary
Source user AO_FC_Silver_r5
Source copy AO_FC_Silver_r5
Writable LUNs 3PAR01
Target system 3PAR02
Target role Secondary
Target user AO_FC_Silver_r5
Target copy AO_FC_Silver_r5

The remote copy target VLUN info:

vm_001.r (3PAR02)

Capacity Utilization
98% Total used
Total used
4,034.36 GiB
Virtual size
4,096.00 GiB
Total reserved
5,485.75 GiB
Host written
0.00 GiB

So my question would be, how do i get the vm_001.r down to 48% utilized like its primary partner on 3PAR01 (or closer to that number)

Info on the systems, both identical:

Model HP_3PAR 7400

HPE 3PAR OS version
Current 3.2.2.709 (MU6)+P99,P107,P119,P131,P135,P137,P138

These are standard Virtual Volumes, Thin provisioned with access to 3 tiers (SSD/FC/NL)
vm_001 (3PAR01)
Virtual size 4,096.00 GiB
Compaction 2.1:1
Type Base
Provisioning Thin
Dedup No
Compression No
Mode Read/Write
RAID RAID 5

Author:  MBILC [ Tue Aug 06, 2019 1:29 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

So, mentioned this to HPE while on another support ticket and they said basically you can not speed up the reclamation of the reserved space, it is a background process and will work in 128MB chunks and will run when it decides.

Problem is i have several virtual volumes that i ran UNMAP on a week ago, and nothing has changed on them and our arrays are not super loaded or busy.

Author:  MammaGutt [ Tue Aug 06, 2019 11:15 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

Could you give the output of showvv -s for the volumes? Your output isn't detailed enough.

Author:  MBILC [ Wed Aug 07, 2019 11:14 am ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

MammaGutt wrote:
Could you give the output of showvv -s for the volumes? Your output isn't detailed enough.



Certainly, the ones circled in blue are the ones I have been working with doing UNAMP on

Image

Author:  MammaGutt [ Wed Aug 07, 2019 11:55 am ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

Is that from the primary or secondary array? Looking at Usr Used compared to Usr Rsvd it is easy to see that unmap has been done on the volume and that the 3PAR has data to be reclaimed.

Author:  MBILC [ Wed Aug 07, 2019 1:02 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

That was from our Primary 3PAR, this is from our 2nd

Image

Author:  MammaGutt [ Wed Aug 07, 2019 1:15 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

Same there, unmap is done but reclaim is pending.

You do however have pretty big snapshots there on all volumes. That is the primary difference.

Author:  MammaGutt [ Wed Aug 07, 2019 3:16 pm ]
Post subject:  Re: UNMAP - > RCG .r VLUN disk not shrinking? / Total reserv

And I just remembered. 3.2.2 MU6 P144 has some reclaim bug fixes.

You have some TB pending reclaim so it might make some difference, but deleting snapshot on just the last volumes and running freespace on those and compactcpg afterwards will give you 18 TB instant.

Page 1 of 2 All times are UTC - 5 hours
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/