Error: No resources are available to complete the request. Failed to create the volume

[vjmartinez]

They could help us with the expansion of the Storage MSA that we have.
We are having trouble creating new volumes in Pool A.
The size of PoolA was expanded, adding new disks and later a new DiskGroup was created in the Pool of type RAid5.
Apparently everything is ok, but when creating the Volume it indicates the following error:
Error: No resources are available to complete the request. (TestVol) - Failed to create the volume. (2023-01-05 10:19:29)

Looking at the properties of the Pool, it can be seen that it still has 15Tb of space available.

There are no other errors registered in the MSA.

command output
# create volume TestVol pool A size 100GB access rw lun 333
Error: No resources are available to complete the request. (TestVol) - Failed to create the volume. (2023-01-05 10:19:29)

# show pools
Name Serial Number Class Total Size Avail Snap Size OverCommit Disk Groups Volumes Low Thresh
Mid Thresh High Thresh Sec Fmt Health Reason Action
-------------------------------------------------- -------------------------------------------------- ----------------------
A 00c0ff26a44c0000fcd3a85601000000 Virtual 37.1TB 15.6TB 0B Enabled 3 4 25.00%
50.00% 99.42% Mixed OK
B 00c0ff26a2d800003eb8796001000000 Virtual 17.9TB 2848.6GB 0B Enabled 2 3 25.00 %
50.00% 98.80% 512n OK
-------------------------------------------------- -------------------------------------------------- ----------------------
Success: Command completed successfully. (2023-01-05 10:42:42)

[JonPaul]

@vjmartinez 
What MSA model are you using and what FW version?
I see 1 issue with your command and 1 possible issue
The create volume command if used with LUN # would require an initiator portion as well
My 2060 system with the latest firmware and it shows an error that the initiator needs to be specified:
# create volume TestVol-99 pool A size 100GB access rw lun 333
Error: If either lun or initiator are specified, then both lun and initiators must be specified together. (2023-01-05 10:26:31)

The other issue is that your LUN number is higher than a standard OS can utilize.  So unless you are using HP-UX you will need a LUN number of 255 or less.

[vjmartinez]

When I try to create the volume from the Web-UI console,

I define the following values:

Vol Name, Size, Vol Number, Preference, PoolA ,

Capture Attachment

After applying these values I have the same error but displayed in the gui. I only have this problem with PoolA, which was extended. The operation works correctly with PoolB.

Viendo la configuracion del PoolA esta correcto.

[vjmartinez]

The last firmware Version is GL225P002-02

MSA2040

[JonPaul]

@vjmartinez 
There's a lot to unpack with the information you provided.  
Pool A has 3 disk groups  all of which are different:
   dgA01 - 15x  1.2TB drives in RAID 5
   DG-Virtualization - 6x  1.2TB drives in RAID 5
   dgA02 -  7x 2.4TB drives in RAID 5
This configurations is outside the Best Practices in a number of ways:
  1. Disk Groups of differing configurations - to maintain consitent performance disk groups should be of same/similar configuration
  2. Use of RAID5 with spinning media - current recommendation is RAID6
  3. Use of non-Power of 2 disk groups - can impact performance of certain workloads
Some of those Best Practices have evolved from the initial shipment of MSA 2040 so I just put them down here for reference. Here is the MSA 2040 Best Practices:  https://www.hpe.com/psnow/doc/4aa4-6892enw

You can also look at the MSA 2050 Best Practices:  https://www.hpe.com/psnow/doc/a00015961enw?jumpid=in_lit-psnow-red
And the 2050 Virtual Storage Technical Reference: https://www.hpe.com/psnow/doc/a00087404enw

The Disk Group, DG-Virtualization, is completely full. which makes me wonder if that is limiting the system to create a new volume.

[vjmartinez]

@JonPaul 

since the Disk Group, DG-Virtualization, is completely full. Is it possible to redistribute the data that was in that Diskgroup to the others that are available, in order to delete it?

[JonPaul]

@vjmartinez 
Sorry for the delay. I was attempting to duplicate something similar to your configuration.  I was unable to duplicate.
Q:  Is your Pool configured to allow OverCommit?
To answer your question: Can you delete the full disk-group?
Yes you can remove the full disk-group as long as there is enough space in the remaining disk-groups to retain all the data. In your Pool A you have ~3TB uncommitted which means you should be able to create volumes with or without overcommit enabled so the initial error has an unknown cause.  The disk-group you want to remove is ~6TB so if you do remove the disk-group you would be over-committed on the Pool.
The caution would be that disk-group DRAIN is a long process as it attempts to not interfere with standard I/O.  So a DRAIN would be on the order of multiple days or longer.