Data Backup and Recovery

Lun Reclaim from unix end and yet at storage end it does not show released size

backupstorage
3,445 Views

Hi team, 

 

We are in process of reclaiming space from linux server and ran the command of fstrim also but at the storage end it does not show same space reclaimed. 

Anything from storage to check or scan to have relase space from storage end 

 

Regards, 

 

1 ACCEPTED SOLUTION

pedro_rocha
3,074 Views

Hi,

 

rescan is required if the LUN has been changed space-allocation from disabled to enabled

 

it should not pose any downtime to rescan

 

you can do any kind of rescan from the storage side

 

you should run fstrim to reclaim the space after all

 

 

 

View solution in original post

7 REPLIES 7

pedro_rocha
3,420 Views

Hi,

 

Is it a LUN or NFS mount?

 

fstrim did any reclamation or none at all?

 

fstrim -v gives the potential reclaimable data?

 

is the volume/lun THIN?

 

Regards,

Pedro.

Ontapforrum
3,410 Views

Two things:

 

1st - You didn't mention which *nix, KB here.

https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/What_are_the_recommendations_for_the_host_stacks_supporting_SCSI_UNMAP...


2nd - In any case, If the Host OS TRIM is manually done, then ensure this option is enabled.

-space-allocation enabled (This option enables ONTAP to reclaim space automatically when your host deletes data. It is set to disabled by default.)


Steps here: (You need to follow the exact steps)

https://docs.netapp.com/us-en/ontap/san-admin/enable-space-allocation-scsi-thin-provisioned-luns-task.html

backupstorage
3,350 Views

@pedro_rocha @Ontapforrum  Thank you for reply 

 

 volume/lun THIN? Yes it is Thin provisioned 

 

reclaim space automatically is enabled already 

 

fstrim  -->it didn't help much there is difference of 200 GB 

 

Please help to know if anything has to be done on it. 

 

 

 

 

 

 

 

pedro_rocha
3,333 Views

is it a LUN (iSCSI / FC) or volume (NFS)?

 

as @Ontapforrum said, if LUN, -space-allocation was enabled after it was presented to the host? If yes you should rescan

 

which file system are you using if LUN?

 

did you review this? https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/Storage_auto-reclamation_does_not_occur

 

backupstorage
3,327 Views

@pedro_rocha  Will check and get back to you accordingly 

backupstorage
3,106 Views

@pedro_rocha 

 

which file system are you using if LUN? Lun is ISCI

 

yes we did review the above KB and space-allocation enabled  is already set 

 

Let me know if below rescanning is required to be done 

 

PS Link Refer

 

 

https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Software/ONTAP_OS/How_to_rescan_and_recover_LUN_paths_in_a_host_after_modifying_SLM_repo...

 

 

Linux hosts:

  • Rescan after add-reporting-nodes.
    1. Starting RHEL 6.5 & RHEL 7.0 onwards, run the following command to update active/optimized paths after add-reporting-nodes:
      # /usr/bin/rescan-scsi-bus.sh –a
    2. For RHEL 5 and RHEL 6.4 (including previous updates), run the following command to update active/optimized paths after add-reporting-nodes:
      # /usr/bin/rescan-scsi-bus.sh
      Note: Nothing additional has to be done in the multipath layer.
  • Rescan after remove-reporting-nodes
    1. Separate rescan steps are required for SCSI layer and Multipathing layer in Linux storage stack to clean up stale disk paths after remove-reporting-nodes in SLM.
    2. Run the following command to remove stale LUN paths in SCSI layer
      # /usr/bin/rescan-scsi-bus.sh –r
    3. Next run the following command to remove stale LUN paths in multipath layer:
      # multipath -r
      Note: The /usr/bin/rescan-scsi-bus.sh script is available as part of the native sg3_utils package.

Anything from the storage end scan can be done ?

Scanning on storage end will it require reboot ?

If Scanning is possible on storage end , is it disruptive operation?

 

Thanks in advance 

pedro_rocha
3,075 Views

Hi,

 

rescan is required if the LUN has been changed space-allocation from disabled to enabled

 

it should not pose any downtime to rescan

 

you can do any kind of rescan from the storage side

 

you should run fstrim to reclaim the space after all

 

 

 

Public