How to replace ESS MMV Disk in case of faulty disk ?
Today in this article, we will check how to replace MMV faulty Disk in IBM ESS server.
Lets have a look step by step how to proceed with this.
1. List the available pdisks and get the failed disk name.
ngeaix001--> mmvdisk pdisk list --recovery-group rg_ngeaix022-da declustered recovery group pdisk array paths capacity free space FRU (type) state -------------- ------------ ----------- ----- -------- ---------- --------------- ----- rg_ngeaix022-da e1d1s02 DA1 2 7452 GiB 220 GiB 11XX022 ok rg_ngeaix022-da e1d1s03 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e1d1s04 DA1 2 7452 GiB 220 GiB 11XX022 ok rg_ngeaix022-da e1d1s05 DA1 2 7452 GiB 220 GiB 11XX022 ok rg_ngeaix022-da e2d2s21 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e2d2s29 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e2d2s30 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e2d2s31 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e2d2s32 DA1 0 7452 GiB 7448 GiB 11XX022 failing/replace rg_ngeaix022-da e2d2s33 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e2d2s34 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e2d2s35 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da n001v001 NVR 1 1992 MiB 1816 MiB IPR-10 68C8730 ok rg_ngeaix022-da n002v001 NVR 1 1992 MiB 1816 MiB IPR-10 68C8C10 ok rg_ngeaix022-da e1d1s01ssd SSD 2 745 GiB 744 GiB 00LY451 ok
2. Now we need to replace the failing disk above with name e2d2s32.
Once the engineer is onsite, we need to release this disk.
3. Releasing the failed disk.
ngeaix001--> mmvdisk pdisk replace --prepare --recovery-group rg_ngeaix022-da --pdisk e2d2s32 mmvdisk: Suspending pdisk e2d2s32 of RG rg_ngeaix022-da in location 89R011B-2-32. mmvdisk: Location 89R011B-2-32 is Rack Pyramid Park, E5 U11-15, Enclosure 5147-084-89R011B Drawer 2 Slot 32. mmvdisk: Carrier released. mmvdisk: mmvdisk: - Remove carrier. mmvdisk: - Replace disk in location 89R011B-2-32 with type '11XX022'. mmvdisk: - Reinsert carrier. mmvdisk: - Issue the following command: mmvdisk: mmvdisk: mmvdisk pdisk replace --recovery-group rg_ngeaix022-da --pdisk 'e2d2s32' ngeaix001-->
4. Now the disk is released and you can ask engineer to replace this faulty disk.
Once the disk is replaced, check out the replaced disk by listing the disks.
ngeaix001--> mmvdisk pdisk list --recovery-group rg_ngeaix022-da declustered recovery group pdisk array paths capacity free space FRU (type) state -------------- ------------ ----------- ----- -------- ---------- --------------- ----- rg_ngeaix022-da e1d1s02 DA1 2 7452 GiB 220 GiB 11XX022 ok rg_ngeaix022-da e1d1s03 DA1 2 7452 GiB 216 GiB 11XX022 ok rg_ngeaix022-da e1d1s04 DA1 2 7452 GiB 220 GiB 11XX022 ok rg_ngeaix022-da e1d1s05 DA1 2 7452 GiB 220 GiB 11XX022 ok
5. If it shows fine, then we can format this new disk using below command.
ngeaix001--> mmvdisk pdisk replace --recovery-group rg_ngeaix022-da --pdisk 'e2d2s32' mmvdisk: mmvdisk: mmchcarrier : [I] Preparing a new pdisk for use may take many minutes. mmvdisk: mmvdisk: 2021-08-04_12:34:01.261+0100: [I] Callback: /usr/lpp/mmfs/bin/tspreparenewpdiskforuse /dev/sdhu. mmvdisk: Attempting to update firmware if necessary. Failure will not prevent drive replacement. mmvdisk: Command: mmchfirmware --type drive --serial-number YA1GVVRR0000C020LMQQ --new-pdisk mmvdisk: Command: err 0: mmchfirmware --type drive --serial-number YA1GVVRR0000C020LMQQ --new-pdisk mmvdisk: mmvdisk: The following pdisks will be formatted on node ngeaix022: mmvdisk: //ngeaix022-da/dev/sdbz,//ngeaix022-da/dev/sdhu,//ngeaix023-da/dev/sdy,//ngeaix023-da/dev/sdhu mmvdisk: Pdisk e2d2s32 of RG rg_ngeaix022-da successfully replaced. mmvdisk: Resuming pdisk e2d2s32#0037 of RG rg_ngeaix022-da. mmvdisk: Carrier resumed. ngeaix001-->
Now the disk is formatted and is now resumed to be a part of the recovery disk group and the redundancy is back in place.