Oracle Database Appliance now includes a powerful virtualisation layer based on KVM and fully integrated with odacli. For sure, an ODA is mostly dedicated to Oracle databases, but you can also benefit from unused cores to run application VMs in the same box as your databases. And it works like a charm.<\/p>\n\n\n\n
One of my customer recently asked if it would be possible to restore a part of the VM data without interrupting the service, the goal being to browse VM data at a point in time and compare to what’s in the current filesystem. This is not for a full VM restore but for getting back files from an older version.<\/p>\n\n\n\n
On my ODA X8-2M running 19.19 version, let’s create a CPU pool for the VMs, create the VM storage on DATA diskgroup and create a data vdisk. This vdisk will be the one I’ll use for data recovery:<\/p>\n\n\n\n
odacli list-vms\nNo data found for VM\n\nodacli create-cpupool -n CpuPool4VMs -c 4 -vm\nsleep 30; odacli list-cpupools\nName Type Configured on Cores Associated resources Created Updated \n-------------------- ------------------ ------------------------- ----- ------------------------------ ------------------------ ------------------------\nCpuPool4VMs VM dbioda01 4 NONE 2023-10-05 11:24:07 CEST 2023-10-05 11:24:07 CEST\n\nodacli create-vmstorage -n VMsDATA -dg DATA -s 500G\nsleep 30; odacli list-vmstorages\nName Disk group Volume name Volume device Size Used Used % Available Mount Point Created Updated\n-------------------- --------------- --------------- ------------------------------ ---------- ---------- ---------- ---------- ----------------------------------- ------------------------ ------------------------\nVMsDATA DATA VMSDATA \/dev\/asm\/vmsdata-214 500.00 GB 1.36 GB 0.27% 498.64 GB \/u05\/app\/sharedrepo\/vmsdata 2023-10-05 11:27:58 CEST 2023-10-05 11:27:58 CEST\n\n \nodacli create-vdisk -n dbivm1-data -vms VMsDATA -s 100G\n\nsleep 120 ; odacli describe-job -i 1f438468-0835-4c8f-8a0d-beb99416524f\n\nJob details\n----------------------------------------------------------------\n ID: 1f438468-0835-4c8f-8a0d-beb99416524f\n Description: VDisk dbivm1-data creation\n Status: Success\n Created: October 5, 2023 11:31:14 AM CEST\n Message:\n\nTask Name Start Time End Time Status\n---------------------------------------- ----------------------------------- ----------------------------------- ----------\nValidate Virtual Disk doesn't exist October 5, 2023 11:31:14 AM CEST October 5, 2023 11:31:14 AM CEST Success\nValidate Vm Storage exists October 5, 2023 11:31:14 AM CEST October 5, 2023 11:31:14 AM CEST Success\nValidate Vm Storage space October 5, 2023 11:31:14 AM CEST October 5, 2023 11:31:14 AM CEST Success\nCreate Virtual Disk snapshot October 5, 2023 11:31:14 AM CEST October 5, 2023 11:31:14 AM CEST Success\nCreate Virtual Disk directories October 5, 2023 11:31:14 AM CEST October 5, 2023 11:31:14 AM CEST Success\nCreate Virtual Disk October 5, 2023 11:31:14 AM CEST October 5, 2023 11:33:22 AM CEST Success\nCreate metadata October 5, 2023 11:33:22 AM CEST October 5, 2023 11:33:22 AM CEST Success\nPersist metadata October 5, 2023 11:33:22 AM CEST October 5, 2023 11:33:22 AM CEST Success\n\n\nodacli list-vdisks\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST<\/code><\/pre>\n\n\n\nNow it’s time to create the dbivm1 virtual machine. It will have 32GB of RAM and an Oracle Linux 7.9 distribution for the first boot. I will also connect my data vdisk I created before and this VM will use 2 CPU cores from the CPU pool. This VM will be connected to the pubnet network, it will have a 50GB boot disk and it will listen on my ODA’s IP for GUI setup.<\/p>\n\n\n\n
odacli create-vm -n dbivm1 -m 32G -src \/opt\/dbi\/V1009690-01.iso -vd dbivm1-data -vc 2 -cp CpuPool4VMs -vn pubnet -vms VMsDATA -s 50G -g \"vnc,listen=10.86.20.241\"\n\nodacli describe-job -i bcc6662a-7a0b-4568-b853-9da86384ca13\n\nJob details\n----------------------------------------------------------------\n ID: bcc6662a-7a0b-4568-b853-9da86384ca13\n Description: VM dbivm1 creation\n Status: Success\n Created: October 5, 2023 1:41:22 PM CEST\n Message:\n\nTask Name Start Time End Time Status\n---------------------------------------- ----------------------------------- ----------------------------------- ----------\nValidate dependency resources October 5, 2023 1:41:22 PM CEST October 5, 2023 1:41:22 PM CEST Success\nValidate resource allocations October 5, 2023 1:41:22 PM CEST October 5, 2023 1:41:23 PM CEST Success\nAllocate resources October 5, 2023 1:41:23 PM CEST October 5, 2023 1:41:23 PM CEST Success\nProvision new VM October 5, 2023 1:41:23 PM CEST October 5, 2023 1:41:25 PM CEST Success\nAdd VM to Clusterware October 5, 2023 1:41:25 PM CEST October 5, 2023 1:41:26 PM CEST Success\nSave domain in ACFS October 5, 2023 1:41:26 PM CEST October 5, 2023 1:41:26 PM CEST Success\nCreate VM metadata October 5, 2023 1:41:26 PM CEST October 5, 2023 1:41:26 PM CEST Success\nPersist metadata October 5, 2023 1:41:26 PM CEST October 5, 2023 1:41:26 PM CEST Success\n\n\nodacli list-vms\nName VM Storage Current State Target State Created Updated\n-------------------- -------------------- --------------- --------------- ------------------------ ------------------------\ndbivm1 VMsDATA ONLINE ONLINE 2023-10-05 13:41:26 CEST 2023-10-05 13:41:26 CEST\n\n\nodacli describe-vm -n dbivm1\nVM details\n--------------------------------------------------------------------------------\n ID: 07a741e0-d356-4b5c-adb5-d5f3c8ae5d03\n Name: dbivm1\n Created: 2023-10-05 13:41:26 CEST\n Updated: 2023-10-05 13:41:26 CEST\n VM Storage: VMsDATA\n Description: NONE\n VM image path: \/u05\/app\/sharedrepo\/vmsdata\/.ACFS\/snaps\/vm_dbivm1\/dbivm1\n VM size: 50.00 GB\n Source: V1009690-01.iso\n Cloned from: N\/A\n OS Type: NONE\n OS Variant: NONE\n Graphics settings: vnc,listen=10.86.20.241\n Display Port: 10.86.20.241:0\n\n Status\n--------------------------\n Current node: dbioda01\n Current state: ONLINE\n Target state: ONLINE\n\n Parameters\n--------------------------\n Preferred node: NONE\n Boot option: NONE\n Auto start: YES\n Fail over: NO\n NUMA enabled: NO\n\n Config Live\n ------------------------- -------------------------\n Memory: 32.00 GB 32.00 GB\n Max Memory: 32.00 GB 32.00 GB\n vCPU count: 2 2\n Max vCPU count: 2 2\n CPU Pool: CpuPool4VMs CpuPool4VMs\n Effective CPU set: 1-2,5-6,9-10,13-14 1-2,5-6,9-10,13-14\n vCPUs: 0:1-2,5-6,9-10,13-14 0:1-2,5-6,9-10,13-14\n 1:1-2,5-6,9-10,13-14 1:1-2,5-6,9-10,13-14\n vDisks: dbivm1-data:vdb dbivm1-data:vdb\n vNetworks: pubnet:52:54:00:1f:f4:84 pubnet:52:54:00:1f:f4:84<\/code><\/pre>\n\n\n\nDisplay port is 10.86.20.241:0: I will then connect to this new VM using a VNC Viewer on 10.86.20.241:5900. Default port for VNC is 5900, just add the defined port here (5900+0). I will then do the setup of the operating system: everything with default values appart from network settings (10.86.20.248\/24). I will only use the 50GB boot disk for Linux setup, I will configure the data disk later.<\/p>\n\n\n\n
<\/p>\n\n\n\n
Once the OS is deployed, let’s configure the 100GB data disk:<\/p>\n\n\n\n
ssh root@10.86.20.248\nfdisk -l \/dev\/vdb | grep GB\nDisk \/dev\/vdb: 107.4 GB, 107374182400 bytes, 209715200 sectors\n\npvcreate \/dev\/vdb\nvgcreate vg_data \/dev\/vdb\nlvcreate -L 80G -n lv_data vg_data\nmkfs.ext4 \/dev\/mapper\/vg_data-lv_data\nmkdir \/data01\necho \"\/dev\/mapper\/vg_data-lv_data \/data01 ext4 defaults 1 2\" >> \/etc\/fstab\nmount -a\n\ndf -h\nFilesystem Size Used Avail Use% Mounted on\ndevtmpfs 16G 0 16G 0% \/dev\ntmpfs 16G 0 16G 0% \/dev\/shm\ntmpfs 16G 8.6M 16G 1% \/run\ntmpfs 16G 0 16G 0% \/sys\/fs\/cgroup\n\/dev\/mapper\/ol-root 44G 1.4G 43G 4% \/\n\/dev\/vda1 1014M 184M 831M 19% \/boot\ntmpfs 3.2G 0 3.2G 0% \/run\/user\/0\n\/dev\/mapper\/vg_data-lv_data 79G 57M 75G 1% \/data01<\/code><\/pre>\n\n\n\nMy data disk is ready, let’s put dummy files inside:<\/p>\n\n\n\n
rm -rf dummyfile_* ; for a in {0001..5000}; do head -c $RANDOM \/dev\/urandom > \/data01\/dummyfile_$a.dmf; done\n\ndf -h\nFilesystem Size Used Avail Use% Mounted on\ndevtmpfs 16G 0 16G 0% \/dev\ntmpfs 16G 0 16G 0% \/dev\/shm\ntmpfs 16G 8.6M 16G 1% \/run\ntmpfs 16G 0 16G 0% \/sys\/fs\/cgroup\n\/dev\/mapper\/ol-root 44G 1.4G 43G 4% \/\n\/dev\/vda1 1014M 184M 831M 19% \/boot\ntmpfs 3.2G 0 3.2G 0% \/run\/user\/0\n\/dev\/mapper\/vg_data-lv_data 79G 144M 75G 1% \/data01\n\nls -lrt | head\ntotal 89860\ndrwx------. 2 root root 16384 Oct 5 09:30 lost+found\n-rw-r--r--. 1 root root 15781 Oct 5 10:03 dummyfile_0001.dmf\n-rw-r--r--. 1 root root 14976 Oct 5 10:03 dummyfile_0002.dmf\n-rw-r--r--. 1 root root 22209 Oct 5 10:03 dummyfile_0003.dmf\n-rw-r--r--. 1 root root 20250 Oct 5 10:03 dummyfile_0005.dmf\n-rw-r--r--. 1 root root 21634 Oct 5 10:03 dummyfile_0004.dmf\n-rw-r--r--. 1 root root 2424 Oct 5 10:03 dummyfile_0006.dmf\n-rw-r--r--. 1 root root 23033 Oct 5 10:03 dummyfile_0007.dmf\n-rw-r--r--. 1 root root 11018 Oct 5 10:03 dummyfile_0008.dmf<\/code><\/pre>\n\n\n\nCreating vdisk backups<\/h2>\n\n\n\n
Vdisks can be cloned. Cloning a vdisk is a kind of backup you may use for another VM for example. You will need to stop the source VM to make a clone of its vdisks:<\/p>\n\n\n\n
odacli list-vdisks\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST\n\nodacli stop-vm -n dbivm1; sleep 30 ; odacli clone-vdisk -n dbivm1-data -cn dbivm1-data-`date +\"%Y%m%d-%H%M\"` ; sleep 30 ; odacli start-vm -n dbivm1\n<\/code><\/pre>\n\n\n\nCloning a vdisk is quite fast thanks to NVMe SSDs:<\/p>\n\n\n\n
odacli describe-job -i f05139b2-92cc-4b65-a7df-987807bbb36f\n\nJob details\n----------------------------------------------------------------\n ID: f05139b2-92cc-4b65-a7df-987807bbb36f\n Description: VDisk dbivm1-data cloning\n Status: Success\n Created: October 5, 2023 4:11:56 PM CEST\n Message:\n\nTask Name Start Time End Time Status\n---------------------------------------- ----------------------------------- ----------------------------------- ----------\nValidate Virtual Disk exists October 5, 2023 4:11:56 PM CEST October 5, 2023 4:11:56 PM CEST Success\nValidate Virtual Disk is not attached October 5, 2023 4:11:56 PM CEST October 5, 2023 4:11:56 PM CEST Success\nto any running vms\nClone Virtual Disk October 5, 2023 4:11:56 PM CEST October 5, 2023 4:11:56 PM CEST Success\nPersist metadata October 5, 2023 4:11:56 PM CEST October 5, 2023 4:11:56 PM CEST Success\n<\/code><\/pre>\n\n\n\nI now have another vdisk not attached to any VM:<\/p>\n\n\n\n
odacli list-vdisks\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 16:11:56 CEST 2023-10-05 16:11:56 CEST\n-20231005-1611<\/code><\/pre>\n\n\n\nI can create as many clones I want, for example once a day I will keep a week:<\/p>\n\n\n\n
odacli stop-vm -n dbivm1; sleep 30 ; odacli clone-vdisk -n dbivm1-data -cn dbivm1-data-`date +\"%Y%m%d-%H%M\"` ; sleep 30 ; odacli start-vm -n dbivm1\n\nodacli list-vdisks\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 16:11:56 CEST 2023-10-05 16:11:56 CEST\n-20231005-1611\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 16:26:24 CEST 2023-10-05 16:26:24 CEST\n-20231005-1626<\/code><\/pre>\n\n\n\nThese vdisks are visible from the ODA, under a hidden folder:<\/p>\n\n\n\n
du -hs \/u05\/app\/sharedrepo\/vmsdata\/.ACFS\/snaps\/*\n101G \/u05\/app\/sharedrepo\/vmsdata\/.ACFS\/snaps\/vdisk_dbivm1-data\n101G \/u05\/app\/sharedrepo\/vmsdata\/.ACFS\/snaps\/vdisk_dbivm1-data-20231005-1611\n101G \/u05\/app\/sharedrepo\/vmsdata\/.ACFS\/snaps\/vdisk_dbivm1-data-20231005-1626\n2.1G \/u05\/app\/sharedrepo\/vmsdata\/.ACFS\/snaps\/vm_dbivm1<\/code><\/pre>\n\n\n\nAltering data on the filesystem<\/h2>\n\n\n\n
Now let’s simulate a loss of data on the data filesystem of my VM:<\/p>\n\n\n\n
ssh root@10.86.20.248\ncd \/data01\nrm -rf dummyfile_0*.dmf\nls -lrt | head\ntotal 71596\ndrwx------. 2 root root 16384 Oct 5 09:30 lost+found\n-rw-r--r--. 1 root root 10930 Oct 5 10:03 dummyfile_1000.dmf\n-rw-r--r--. 1 root root 11321 Oct 5 10:03 dummyfile_1001.dmf\n-rw-r--r--. 1 root root 32045 Oct 5 10:03 dummyfile_1002.dmf\n-rw-r--r--. 1 root root 27619 Oct 5 10:03 dummyfile_1003.dmf\n-rw-r--r--. 1 root root 9086 Oct 5 10:03 dummyfile_1004.dmf\n-rw-r--r--. 1 root root 31690 Oct 5 10:03 dummyfile_1005.dmf\n-rw-r--r--. 1 root root 23343 Oct 5 10:03 dummyfile_1006.dmf\n-rw-r--r--. 1 root root 3594 Oct 5 10:03 dummyfile_1007.dmf<\/code><\/pre>\n\n\n\nOK, I lost data and I will need to compare and pick-up lost data from an old copy of my vdisk.<\/p>\n\n\n\n
Attaching a vdisk to an existing VM<\/h2>\n\n\n\n
I will attach an old version of the vdisk to my VM. I need to specify the “live” option because I can’t reboot my VM now:<\/p>\n\n\n\n
odacli modify-vm -n dbivm1 -avd dbivm1-data-20231005-1611 --live\n\nodacli describe-job -i d1f15f1a-5499-486d-83fb-18b0225ede57\n\nJob details\n----------------------------------------------------------------\n ID: d1f15f1a-5499-486d-83fb-18b0225ede57\n Description: VM dbivm1 modification\n Status: Success\n Created: October 5, 2023 5:11:09 PM CEST\n Message:\n\nTask Name Start Time End Time Status\n---------------------------------------- ----------------------------------- ----------------------------------- ----------\nValidate dependency resources October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nDefine VM locally October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nValidate vDisk attachment pre-reqs October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nAttach vDisks October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nEdit VM CRS Configuration October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nSave domain in ACFS October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nModify VM metadata October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success\nPersist metadata October 5, 2023 5:11:09 PM CEST October 5, 2023 5:11:09 PM CEST Success<\/code><\/pre>\n\n\n\nConnecting this vdisk to the VM was quite easy.<\/p>\n\n\n\n
Making data from vdisk available<\/h2>\n\n\n\n
Data from this vdisk is not yet available because it has the same UUID as source data vdisk currently in use. But a vgimportclone will solve this problem by changing the UUID:<\/p>\n\n\n\n
mkdir \/rescue01\nvgimportclone -i -n vg_rescue \/dev\/vdc\n\nvgdisplay\n --- Volume group ---\n VG Name vg_data\n System ID\n Format lvm2\n Metadata Areas 1\n Metadata Sequence No 4\n VG Access read\/write\n VG Status resizable\n MAX LV 0\n Cur LV 1\n Open LV 1\n Max PV 0\n Cur PV 1\n Act PV 1\n VG Size <100.00 GiB\n PE Size 4.00 MiB\n Total PE 25599\n Alloc PE \/ Size 20480 \/ 80.00 GiB\n Free PE \/ Size 5119 \/ <20.00 GiB\n VG UUID L0BHcA-0h03-d4kz-0eY3-xx1a-N6ZF-7DUIQP\n\n --- Volume group ---\n VG Name ol\n System ID\n Format lvm2\n Metadata Areas 1\n Metadata Sequence No 3\n VG Access read\/write\n VG Status resizable\n MAX LV 0\n Cur LV 2\n Open LV 2\n Max PV 0\n Cur PV 1\n Act PV 1\n VG Size <49.00 GiB\n PE Size 4.00 MiB\n Total PE 12543\n Alloc PE \/ Size 12543 \/ <49.00 GiB\n Free PE \/ Size 0 \/ 0\n VG UUID QPR9Tk-CqNq-GsZ1-1qVO-tcDH-bcfh-EuBRQl\n\n --- Volume group ---\n VG Name vg_rescue\n System ID\n Format lvm2\n Metadata Areas 1\n Metadata Sequence No 6\n VG Access read\/write\n VG Status resizable\n MAX LV 0\n Cur LV 1\n Open LV 0\n Max PV 0\n Cur PV 1\n Act PV 1\n VG Size <100.00 GiB\n PE Size 4.00 MiB\n Total PE 25599\n Alloc PE \/ Size 20480 \/ 80.00 GiB\n Free PE \/ Size 5119 \/ <20.00 GiB\n VG UUID NiWkOI-5dvL-f2Pf-z7Wa-bk9c-4TvL-JK09E8\n\n vgscan\n Reading volume groups from cache.\n Found volume group \"vg_data\" using metadata type lvm2\n Found volume group \"ol\" using metadata type lvm2\n Found volume group \"vg_rescue\" using metadata type lvm2\n\nlvdisplay\n --- Logical volume ---\n LV Path \/dev\/vg_data\/lv_data\n LV Name lv_data\n VG Name vg_data\n LV UUID GcfE3o-X5YB-CJF3-emBr-socY-yJIZ-Wiq5N0\n LV Write Access read\/write\n LV Creation host, time localhost.localdomain, 2023-10-05 09:30:45 -0400\n LV Status available\n # open 1\n LV Size 80.00 GiB\n Current LE 20480\n Segments 1\n Allocation inherit\n Read ahead sectors auto\n - currently set to 8192\n Block device 252:2\n\n --- Logical volume ---\n LV Path \/dev\/ol\/swap\n LV Name swap\n VG Name ol\n LV UUID Iu1Xk6-2aAC-0ZDd-amQl-k2LD-f3kv-RLSFFW\n LV Write Access read\/write\n LV Creation host, time dhcp-10-36-0-229, 2023-10-05 08:01:24 -0400\n LV Status available\n # open 2\n LV Size 5.00 GiB\n Current LE 1280\n Segments 1\n Allocation inherit\n Read ahead sectors auto\n - currently set to 8192\n Block device 252:1\n\n --- Logical volume ---\n LV Path \/dev\/ol\/root\n LV Name root\n VG Name ol\n LV UUID wWG6wE-cF9r-Li2x-BYU6-OhAJ-cZZa-dj0pcn\n LV Write Access read\/write\n LV Creation host, time dhcp-10-36-0-229, 2023-10-05 08:01:24 -0400\n LV Status available\n # open 1\n LV Size <44.00 GiB\n Current LE 11263\n Segments 1\n Allocation inherit\n Read ahead sectors auto\n - currently set to 8192\n Block device 252:0\n\n --- Logical volume ---\n LV Path \/dev\/vg_rescue\/lv_data\n LV Name lv_data\n VG Name vg_rescue\n LV UUID GcfE3o-X5YB-CJF3-emBr-socY-yJIZ-Wiq5N0\n LV Write Access read\/write\n LV Creation host, time localhost.localdomain, 2023-10-05 09:30:45 -0400\n LV Status NOT available\n LV Size 80.00 GiB\n Current LE 20480\n Segments 1\n Allocation inherit\n Read ahead sectors auto<\/code><\/pre>\n\n\n\nLet’s change the logical volume status to make it available, then mount the filesystem on \/rescue01:<\/p>\n\n\n\n
lvchange -a \/dev\/vg_rescue\/lv_data\n\nmount \/dev\/mapper\/vg_rescue-lv_data \/rescue01\/\n\ndf -h\n\nFilesystem Size Used Avail Use% Mounted on\ndevtmpfs 16G 0 16G 0% \/dev\ntmpfs 16G 0 16G 0% \/dev\/shm\ntmpfs 16G 8.6M 16G 1% \/run\ntmpfs 16G 0 16G 0% \/sys\/fs\/cgroup\n\/dev\/mapper\/ol-root 44G 1.4G 43G 4% \/\n\/dev\/vda1 1014M 184M 831M 19% \/boot\n\/dev\/mapper\/vg_data-lv_data 79G 127M 75G 1% \/data01\ntmpfs 3.2G 0 3.2G 0% \/run\/user\/0\n\/dev\/mapper\/vg_rescue-lv_data 79G 144M 75G 1% \/rescue01<\/code><\/pre>\n\n\n\n“Restored” data is now available and I can now compare with my live data vdisk.<\/p>\n\n\n\n
Putting back files from old vdisk version<\/h2>\n\n\n\n
Let’s copy back the files missing on my \/data01 filesystem:<\/p>\n\n\n\n
cd \/data01\nls -lrt | head\ntotal 71596\ndrwx------. 2 root root 16384 Oct 5 09:30 lost+found\n-rw-r--r--. 1 root root 10930 Oct 5 10:03 dummyfile_1000.dmf\n-rw-r--r--. 1 root root 11321 Oct 5 10:03 dummyfile_1001.dmf\n-rw-r--r--. 1 root root 32045 Oct 5 10:03 dummyfile_1002.dmf\n-rw-r--r--. 1 root root 27619 Oct 5 10:03 dummyfile_1003.dmf\n-rw-r--r--. 1 root root 9086 Oct 5 10:03 dummyfile_1004.dmf\n-rw-r--r--. 1 root root 31690 Oct 5 10:03 dummyfile_1005.dmf\n-rw-r--r--. 1 root root 23343 Oct 5 10:03 dummyfile_1006.dmf\n-rw-r--r--. 1 root root 3594 Oct 5 10:03 dummyfile_1007.dmf\n\ncd \/rescue01\nls -lrt | head\ntotal 89860\ndrwx------. 2 root root 16384 Oct 5 09:30 lost+found\n-rw-r--r--. 1 root root 15781 Oct 5 10:03 dummyfile_0001.dmf\n-rw-r--r--. 1 root root 14976 Oct 5 10:03 dummyfile_0002.dmf\n-rw-r--r--. 1 root root 22209 Oct 5 10:03 dummyfile_0003.dmf\n-rw-r--r--. 1 root root 20250 Oct 5 10:03 dummyfile_0005.dmf\n-rw-r--r--. 1 root root 21634 Oct 5 10:03 dummyfile_0004.dmf\n-rw-r--r--. 1 root root 2424 Oct 5 10:03 dummyfile_0006.dmf\n-rw-r--r--. 1 root root 23033 Oct 5 10:03 dummyfile_0007.dmf\n-rw-r--r--. 1 root root 11018 Oct 5 10:03 dummyfile_0008.dmf\n\ncp \/rescue01\/dummyfile_0*.dmf \/data01\/\n\ncd \/data01\nls -lrt dummyfile_0*.dmf | head\n-rw-r--r--. 1 root root 15781 Oct 5 11:36 dummyfile_0001.dmf\n-rw-r--r--. 1 root root 14976 Oct 5 11:36 dummyfile_0002.dmf\n-rw-r--r--. 1 root root 22209 Oct 5 11:36 dummyfile_0003.dmf\n-rw-r--r--. 1 root root 21634 Oct 5 11:36 dummyfile_0004.dmf\n-rw-r--r--. 1 root root 20250 Oct 5 11:36 dummyfile_0005.dmf\n-rw-r--r--. 1 root root 2424 Oct 5 11:36 dummyfile_0006.dmf\n-rw-r--r--. 1 root root 23033 Oct 5 11:36 dummyfile_0007.dmf\n-rw-r--r--. 1 root root 11018 Oct 5 11:36 dummyfile_0008.dmf\n-rw-r--r--. 1 root root 27026 Oct 5 11:36 dummyfile_0009.dmf\n-rw-r--r--. 1 root root 17601 Oct 5 11:36 dummyfile_0010.dmf<\/code><\/pre>\n\n\n\nRemove the restored vdisk<\/h2>\n\n\n\n
I don’t need the old data disk anymore on the VM:<\/p>\n\n\n\n
umount \/rescue01\nrmdir \/rescue01\/\n\nvgchange -a n vg_rescue\n<\/code><\/pre>\n\n\n\nBack to the ODA, let’s remove the vdisk from the VM (with “live” option):<\/p>\n\n\n\n
odacli modify-vm -n dbivm1 -dvd dbivm1-data-20231005-1611 --live\n\nodacli describe-job -i 38d8a632-0a4d-46c4-88dd-f2abcd26835a\n\nJob details\n----------------------------------------------------------------\n ID: 38d8a632-0a4d-46c4-88dd-f2abcd26835a\n Description: VM dbivm1 modification\n Status: Success\n Created: October 5, 2023 5:37:08 PM CEST\n Message:\n\nTask Name Start Time End Time Status\n---------------------------------------- ----------------------------------- ----------------------------------- ----------\nValidate dependency resources October 5, 2023 5:37:08 PM CEST October 5, 2023 5:37:08 PM CEST Success\nDefine VM locally October 5, 2023 5:37:08 PM CEST October 5, 2023 5:37:08 PM CEST Success\nValidate vDisk detachment pre-reqs October 5, 2023 5:37:08 PM CEST October 5, 2023 5:37:08 PM CEST Success\nDetach vDisks October 5, 2023 5:37:08 PM CEST October 5, 2023 5:37:09 PM CEST Success\nEdit VM CRS Configuration October 5, 2023 5:37:09 PM CEST October 5, 2023 5:37:09 PM CEST Success\nSave domain in ACFS October 5, 2023 5:37:09 PM CEST October 5, 2023 5:37:09 PM CEST Success\nModify VM metadata October 5, 2023 5:37:09 PM CEST October 5, 2023 5:37:09 PM CEST Success\nPersist metadata October 5, 2023 5:37:09 PM CEST October 5, 2023 5:37:09 PM CEST Success<\/code><\/pre>\n\n\n\nThat’s it, my VM is now back to its original configuration with only one data vdisk.<\/p>\n\n\n\n
Clean up the old vdisk clones<\/h2>\n\n\n\n
Removing a vdisk clone is quite easy:<\/p>\n\n\n\n
odacli list-vdisks\n\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 16:11:56 CEST 2023-10-05 16:11:56 CEST\n-20231005-1611\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 16:26:24 CEST 2023-10-05 16:26:24 CEST\n-20231005-1626\n\n\nodacli delete-vdisk -n dbivm1-data-20231005-1611\n\nsleep 30 ; odacli list-vdisks\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 16:26:24 CEST 2023-10-05 16:26:24 CEST\n-20231005-1626\n\n\nodacli delete-vdisk -n dbivm1-data-20231005-1626\n\nsleep 30 ; odacli list-vdisks\n\nName VM storage Size Shared Sparse Created Updated\n-------------------- -------------------- ---------- ---------- ---------- ------------------------ ------------------------\ndbivm1-data VMsDATA 100.00 GB NO NO 2023-10-05 11:33:22 CEST 2023-10-05 11:33:22 CEST\n\n<\/code><\/pre>\n\n\n\nConclusion<\/h2>\n\n\n\n
This is a smart solution for this kind of need. The only cost is the ODA storage, and everyone know that it’s quite expensive. But cloning the vdisk is lightning fast on ODA, so VM downtime is limited to the minimum.<\/p>\n","protected":false},"excerpt":{"rendered":"
Introduction Oracle Database Appliance now includes a powerful virtualisation layer based on KVM and fully integrated with odacli. For sure, an ODA is mostly dedicated to Oracle databases, but you can also benefit from unused cores to run application VMs in the same box as your databases. And it works like a charm. One of […]<\/p>\n","protected":false},"author":45,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[42,59],"tags":[],"type_dbi":[],"class_list":["post-28467","post","type-post","status-publish","format-standard","hentry","category-operating-systems","category-oracle"],"acf":[],"yoast_head":"\n
Vdisk backup on ODA - dbi Blog<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Vdisk backup on ODA\" \/>\n<meta property=\"og:description\" content=\"Introduction Oracle Database Appliance now includes a powerful virtualisation layer based on KVM and fully integrated with odacli. For sure, an ODA is mostly dedicated to Oracle databases, but you can also benefit from unused cores to run application VMs in the same box as your databases. And it works like a charm. One of […]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\" \/>\n<meta property=\"og:site_name\" content=\"dbi Blog\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-06T09:30:27+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-06T09:40:34+00:00\" \/>\n<meta name=\"author\" content=\"J\u00e9r\u00f4me Dubar\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"J\u00e9r\u00f4me Dubar\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\"},\"author\":{\"name\":\"J\u00e9r\u00f4me Dubar\",\"@id\":\"https:\/\/www.dbi-services.com\/blog\/#\/schema\/person\/0fb4bbf128b4cda2f96d662dec2baedd\"},\"headline\":\"Vdisk backup on ODA\",\"datePublished\":\"2023-10-06T09:30:27+00:00\",\"dateModified\":\"2023-10-06T09:40:34+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\"},\"wordCount\":685,\"commentCount\":0,\"articleSection\":[\"Operating systems\",\"Oracle\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\",\"url\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\",\"name\":\"Vdisk backup on ODA - dbi Blog\",\"isPartOf\":{\"@id\":\"https:\/\/www.dbi-services.com\/blog\/#website\"},\"datePublished\":\"2023-10-06T09:30:27+00:00\",\"dateModified\":\"2023-10-06T09:40:34+00:00\",\"author\":{\"@id\":\"https:\/\/www.dbi-services.com\/blog\/#\/schema\/person\/0fb4bbf128b4cda2f96d662dec2baedd\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Accueil\",\"item\":\"https:\/\/www.dbi-services.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Vdisk backup on ODA\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.dbi-services.com\/blog\/#website\",\"url\":\"https:\/\/www.dbi-services.com\/blog\/\",\"name\":\"dbi Blog\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.dbi-services.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.dbi-services.com\/blog\/#\/schema\/person\/0fb4bbf128b4cda2f96d662dec2baedd\",\"name\":\"J\u00e9r\u00f4me Dubar\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/secure.gravatar.com\/avatar\/efaa5a7def0aa4cdaf49a470fb4a7641a3ea6e378ae1455096a0933f99f46d6b?s=96&d=mm&r=g\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/efaa5a7def0aa4cdaf49a470fb4a7641a3ea6e378ae1455096a0933f99f46d6b?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/efaa5a7def0aa4cdaf49a470fb4a7641a3ea6e378ae1455096a0933f99f46d6b?s=96&d=mm&r=g\",\"caption\":\"J\u00e9r\u00f4me Dubar\"},\"description\":\"J\u00e9r\u00f4me Dubar has more than 15 years of experience in the field of Information Technology. Ten years ago, he specialized in the Oracle Database technology. His expertise is focused on database architectures, high availability (RAC), disaster recovery (DataGuard), backups (RMAN), performance analysis and tuning (AWR\/statspack), migration, consolidation and appliances, especially ODA (his main projects during the last years). Prior to joining dbi services, J\u00e9r\u00f4me Dubar worked in a Franco-Belgian IT service company as Database team manager and main consultant for 7 years. He also worked for 5 years in a software editor company as technical consultant across France. He was also teaching Oracle Database lessons for 9 years. J\u00e9r\u00f4me Dubar holds a Computer Engineering degree from the Lille Sciences and Technologies university in northern France. His branch-related experience covers the public sector, retail, industry, banking, health, e-commerce and IT sectors.\",\"url\":\"https:\/\/www.dbi-services.com\/blog\/author\/jerome-dubar\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Vdisk backup on ODA - dbi Blog","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/","og_locale":"en_US","og_type":"article","og_title":"Vdisk backup on ODA","og_description":"Introduction Oracle Database Appliance now includes a powerful virtualisation layer based on KVM and fully integrated with odacli. For sure, an ODA is mostly dedicated to Oracle databases, but you can also benefit from unused cores to run application VMs in the same box as your databases. And it works like a charm. One of […]","og_url":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/","og_site_name":"dbi Blog","article_published_time":"2023-10-06T09:30:27+00:00","article_modified_time":"2023-10-06T09:40:34+00:00","author":"J\u00e9r\u00f4me Dubar","twitter_card":"summary_large_image","twitter_misc":{"Written by":"J\u00e9r\u00f4me Dubar","Est. reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#article","isPartOf":{"@id":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/"},"author":{"name":"J\u00e9r\u00f4me Dubar","@id":"https:\/\/www.dbi-services.com\/blog\/#\/schema\/person\/0fb4bbf128b4cda2f96d662dec2baedd"},"headline":"Vdisk backup on ODA","datePublished":"2023-10-06T09:30:27+00:00","dateModified":"2023-10-06T09:40:34+00:00","mainEntityOfPage":{"@id":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/"},"wordCount":685,"commentCount":0,"articleSection":["Operating systems","Oracle"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/","url":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/","name":"Vdisk backup on ODA - dbi Blog","isPartOf":{"@id":"https:\/\/www.dbi-services.com\/blog\/#website"},"datePublished":"2023-10-06T09:30:27+00:00","dateModified":"2023-10-06T09:40:34+00:00","author":{"@id":"https:\/\/www.dbi-services.com\/blog\/#\/schema\/person\/0fb4bbf128b4cda2f96d662dec2baedd"},"breadcrumb":{"@id":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.dbi-services.com\/blog\/vdisk-backup-on-oda\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Accueil","item":"https:\/\/www.dbi-services.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Vdisk backup on ODA"}]},{"@type":"WebSite","@id":"https:\/\/www.dbi-services.com\/blog\/#website","url":"https:\/\/www.dbi-services.com\/blog\/","name":"dbi Blog","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.dbi-services.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/www.dbi-services.com\/blog\/#\/schema\/person\/0fb4bbf128b4cda2f96d662dec2baedd","name":"J\u00e9r\u00f4me Dubar","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/efaa5a7def0aa4cdaf49a470fb4a7641a3ea6e378ae1455096a0933f99f46d6b?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/efaa5a7def0aa4cdaf49a470fb4a7641a3ea6e378ae1455096a0933f99f46d6b?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/efaa5a7def0aa4cdaf49a470fb4a7641a3ea6e378ae1455096a0933f99f46d6b?s=96&d=mm&r=g","caption":"J\u00e9r\u00f4me Dubar"},"description":"J\u00e9r\u00f4me Dubar has more than 15 years of experience in the field of Information Technology. Ten years ago, he specialized in the Oracle Database technology. His expertise is focused on database architectures, high availability (RAC), disaster recovery (DataGuard), backups (RMAN), performance analysis and tuning (AWR\/statspack), migration, consolidation and appliances, especially ODA (his main projects during the last years). Prior to joining dbi services, J\u00e9r\u00f4me Dubar worked in a Franco-Belgian IT service company as Database team manager and main consultant for 7 years. He also worked for 5 years in a software editor company as technical consultant across France. He was also teaching Oracle Database lessons for 9 years. J\u00e9r\u00f4me Dubar holds a Computer Engineering degree from the Lille Sciences and Technologies university in northern France. His branch-related experience covers the public sector, retail, industry, banking, health, e-commerce and IT sectors.","url":"https:\/\/www.dbi-services.com\/blog\/author\/jerome-dubar\/"}]}},"_links":{"self":[{"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/posts\/28467","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/users\/45"}],"replies":[{"embeddable":true,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/comments?post=28467"}],"version-history":[{"count":5,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/posts\/28467\/revisions"}],"predecessor-version":[{"id":28472,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/posts\/28467\/revisions\/28472"}],"wp:attachment":[{"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/media?parent=28467"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/categories?post=28467"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/tags?post=28467"},{"taxonomy":"type","embeddable":true,"href":"https:\/\/www.dbi-services.com\/blog\/wp-json\/wp\/v2\/type_dbi?post=28467"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}