Replacing an NVMe disk

Follow the instructions to replace an NVMe disk.

About this task

The procedure steps are applicable only for the Global Data Platform storage. For Data Foundation storage, see Replacing NVMe disk for Data Foundation storage.

Procedure

  1. Open the IBM Fusion HCI System user interface.
  2. Click the App Switcher icon in title bar and click Storage outbound arrow from the list.
    As soon as you click the Storage outbound arrow, the IBM Storage Scale user interface is displayed.
  3. Click Sign In to login to the IBM Storage Scale user interface.
  4. Click the Events icon and check whether there is any error event showing the bad disk.
  5. Click Run Fix Procedure and follow the instructions to replace the bad disk.
    As soon as you click the Run Fix Procedure, the Fix Procedure: Replace Disks page is displayed.
  6. Click Next.
  7. Replace the disk physically and select the The disk has been replaced check box.
  8. Click Finish.
    The message is displayed in the Fix Procedure: Replace Disks page:
    Successfully replaced the disks
  9. Click Close.

What to do next

  1. Ensure that the error event of a bad disk disappears from the Events page.
  2. Ensure that the new disk shows up with the different serial number in the Physical Disks page.
  3. Check the file system mount and physical disks in a core scale pod.
    • Run the OC command to check Scale core pod.
      oc get pod

NAME READY STATUS RESTARTS AGE

compute-0 2/2 Running 0 17d

compute-1 2/2 Running 0 17d

compute-2 2/2 Running 0 17d

control-0 2/2 Running 0 17d

control-1 2/2 Running 0 17d

control-2 2/2 Running 0 17d
    • Run the OC command to see all the containers in this pod. 
      oc describe pod/compute-2 -n ibm-spectrum-scale
    • Run the following command to check the file system mount.
      mmlsmount all
    • Run the following commands to check the current physical disks health. 
      mmvdisk pdisk list --rg all --not-ok

mmvdisk: All pdisks are ok.
      mmlsrecoverygroup rg1 -L --pdisk

declustered current allowable

recovery group arrays vdisks pdisks format version format version

----------------- ----------- ------ ------ -------------- --------------

rg1 1 25 12 5.1.5.0 5.1.5.0

declustered needs replace scrub background activity

array service vdisks pdisks spares threshold BER trim free space duration task progress priority

----------- ------- ------ ------ ------ --------- ------- ---- ---------- -------- -------------------------

DA1 no 25 12 1,3 1 enable no 326 GiB 14 days rebalance 2% low

n. active, declustered state,

pdisk total paths array free space remarks

----------------- ----------- ----------- ---------- -------

n001p001 1, 1 DA1 212 GiB ok

n001p002 1, 1 DA1 208 GiB ok

n002p001 1, 1 DA1 300 GiB ok

n002p002 1, 1 DA1 252 GiB ok

n003p001 1, 1 DA1 236 GiB ok

n003p002 1, 1 DA1 234 GiB ok

n004p001 1, 1 DA1 4552 GiB ok

n004p002 1, 1 DA1 512 GiB ok

n005p001 1, 1 DA1 232 GiB ok

n005p002 1, 1 DA1 266 GiB ok

n006p001 1, 1 DA1 250 GiB ok

n006p002 1, 1 DA1 222 GiB ok