So, I have a customer who I needs to move 3 “boot from iscsi san” luns from a Netapp OnTap 8.2 Cluster Mode vServer shared with regular iscsi luns, to their own dedicated vserver. How do you easily move between vservers? NDMPCOPY!
Why do they need to do this? Well their boot from iscsi is 1gb, and they have multiple 10gb, and Windows is being a stupid idiot and the iscsi initiator isn’t routing it right.
It’s a big ordeal with a managed service provider/datacenter .
I digress, simple fix, move the boot luns to their own vServer. k thx bye.
Not so fast. How will you move them? This customer cant have a complete outage of their 3 node SQL Cluster. If we could take an extended full outage, we could snapmirror and just bring down the old vol, break, and bring up the new volume, and remap the luns.
We could snapmirror to 3 destinations if SM is licensed. That leaves alot of cruft around though.
Here’s what I’m thinking.
Assume we have an old vserver (s0-fas3220win-iscsi1) with an old volume (boot_s0_pwdb_vol) with 3 luns.
|
1 2 3 4 5 |
vserver path ostype -------------------- ---------------------------------------- ------------ s0-fas3220win-iscsi1 /vol/boot_s0_pwdb_vol/boot-s0-db03.lun windows_2008 s0-fas3220win-iscsi1 /vol/boot_s0_pwdb_vol/boot-s0-db04.lun windows_2008 s0-fas3220win-iscsi1 /vol/boot_s0_pwdb_vol/boot-s0-db05.lun windows_2008 |
What would be possible, is:
No flex cloning or lun cloning needed in this method.
NDMP/ndmpcopy doesn’t work the same in 8.2 cDot mode. It actually has 2 different modes it can work:
node-scope-mode makes the following bad mojo happen:
All NDMP operations are restricted to resources on the node Vserver-aware NDMP commands are disabled NDMP authentication falls back to DATA ONTAP 8.1 NDMP authentication scheme
So, I have lun /vol/boot_s0_pwdb_vol/boot-s0-db03.lun on s0-fas3220win-iscsi1, and I want it to be /vol/ndmptarget1/boot-s0-db03.lun on so-fas3220-win-boot-iscsi1
Verify you have intercluster lifs. Why? You always should… this is how a node in the cluster talks and sends crap to another node in the cluster (SnapMirror, SnapVault, etc need this)
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
network interface create -vserver ha-cluster-01 -lif intercluster_lif-1-01 -role intercluster -home-node ha-cluster-01 -home-port e1b-1 -address 192.168.1.34 -netmask 255.255.255.0 -status-admin up network interface create -vserver ha-cluster-02 -lif intercluster_lif-1-01 -role intercluster -home-node ha-cluster-02 -home-port e1b-1 -address 192.168.1.35 -netmask 255.255.255.0 -status-admin up ha-cluster::system services ndmp> network interface show -role intercluster<br /> Logical Status Network Current Current Is<br /> Vserver Interface Admin/Oper Address/Mask Node Port Home<br /> ----------- ---------- ---------- ------------------ ------------- ------- ----<br /> ha-cluster-01<br /> intercluster_lif-1-01<br /> up/up 192.168.1.34/24 ha-cluster-01 e1b-1 true<br /> intercluster_lif-401-01<br /> up/up 10.40.1.231/24 ha-cluster-01 e1b-401 true<br /> ha-cluster-02<br /> intercluster_lif-1-01<br /> up/up 192.168.1.35/24 ha-cluster-02 e1b-1 true<br /> intercluster_lif-401-02<br /> up/up 10.40.1.232/24 ha-cluster-02 e1b-401 true<br /> 4 entries were displayed. |
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 |
# Enable NDMP to be enabled "cluster wide" with a special root user (this is not a normal root user)</p> system services ndmp modify -node * -enable true -user-id root # Enable ndmpcopy to be ran on all nodes node run -node ha-cluster-02 -command options nodescope.reenabledcmds ndmpcopy node run -node ha-cluster-01 -command options nodescope.reenabledcmds ndmpcopy # Enable the ndmp system service on all nodes system services ndmp on -node * # Ensure "node-scope-mode" is disabled, so we can use clusterip:/vserver/volume pathing system services ndmp node-scope-mode off # Find the name of the ADMIN (cluster) vserver ha-cluster::> vserver show -type admin Admin Root Name Name Vserver Type State Volume Aggregate Service Mapping ----------- ------- --------- ---------- ---------- ------- ------- ha-cluster admin - - - - - # Find the CLUSTER admin IP (same one you ssh into) ha-cluster::> network interface show -role cluster-mgmt Logical Status Network Current Current Is Vserver Interface Admin/Oper Address/Mask Node Port Home ----------- ---------- ---------- ------------------ ------------- ------- ---- ha-cluster cluster_mgmt up/up 192.168.1.232/24 ha-cluster-01 e0a true ha-cluster::vserver services ndmp> security login create -vserver ha-cluster -username backupuser -application ssh -authmethod password -role backup Please enter a password for user 'backupuser': Please enter it again: Password must contain both letters and numbers. Please enter a password for user 'backupuser': Please enter it again: # Create a "backupuser" account on the ADMIN (cluster) vserver with the proper role to be able to use ndmpcopy vserver services ndmp generate-password -vserver ha-cluster -user backupuser Vserver: ha-cluster User: backupuser Password: a1VFgKiXxTIpoU1f <--- this is randomly generated. # Initiate your ndmpcopy from the SOURCE node. (I am going from a vserver on ha-cluster-01 to a vserver on ha-cluster-02) node run local ha-cluster::vserver services ndmp> node run local Type 'exit' or 'Ctrl-D' to return to the CLI ha-cluster-01> <--- notice im now at the local node! ha-cluster-01> ndmpcopy -sa backupuser:a1VFgKiXxTIpoU1f -da backupuser:a1VFgKiXxTIpoU1f 192.168.1.232:/s0-fas3220win-iscsi1/boot_s0_pwdb_vol 192.168.1.232:/so-fas3220-win-boot-iscsi1/ndmptarget Ndmpcopy: Starting copy [ 9 ] ... Ndmpcopy: 192.168.1.232: Notify: Connection established Ndmpcopy: 192.168.1.232: Notify: Connection established Ndmpcopy: 192.168.1.232: Connect: Authentication successful Ndmpcopy: 192.168.1.232: Connect: Authentication successful Ndmpcopy: 192.168.1.232: Log: DUMP: creating "/s0-fas3220win-iscsi1/boot_s0_pwdb_vol/../snapshot_for_backup.3" snapshot. Ndmpcopy: 192.168.1.232: Log: DUMP: Using Full Volume Dump Ndmpcopy: 192.168.1.232: Log: DUMP: Using snapshot_for_backup.3 snapshot Ndmpcopy: 192.168.1.232: Log: DUMP: Date of this level 0 dump: Thu Jun 19 17:11:59 2014. Ndmpcopy: 192.168.1.232: Log: DUMP: Date of last level 0 dump: the epoch. Ndmpcopy: 192.168.1.232: Log: DUMP: Dumping /s0-fas3220win-iscsi1/boot_s0_pwdb_vol to NDMP connection Ndmpcopy: 192.168.1.232: Log: DUMP: mapping (Pass I)[regular files] Ndmpcopy: 192.168.1.232: Log: DUMP: mapping (Pass II)[directories] Ndmpcopy: 192.168.1.232: Log: DUMP: estimated 109220781 KB. Ndmpcopy: 192.168.1.232: Log: DUMP: dumping (Pass III) [directories] Ndmpcopy: 192.168.1.232: Log: DUMP: dumping (Pass IV) [regular files] Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:12:01 2014: Begin level 0 restore Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:12:01 2014: Reading directories from the backup Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:12:01 2014: Creating files and directories. Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:12:01 2014: Writing data to files. Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:17:00 2014 : We have read 38437160 KB from the backup. Ndmpcopy: 192.168.1.232: Log: DUMP: Thu Jun 19 17:17:00 2014 : We have written 38438214 KB. Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:22:00 2014 : We have read 83323241 KB from the backup. Ndmpcopy: 192.168.1.232: Log: DUMP: Thu Jun 19 17:22:00 2014 : We have written 83324500 KB. Ndmpcopy: 192.168.1.232: Log: ACL_START is '111403507712' Ndmpcopy: 192.168.1.232: Log: DUMP: dumping (Pass V) [ACLs] Ndmpcopy: 192.168.1.232: Log: RESTORE: Thu Jun 19 17:24:52 2014: Restoring NT ACLs. Ndmpcopy: 192.168.1.232: Log: DUMP: 108792495 KB Ndmpcopy: 192.168.1.232: Log: DUMP: DUMP IS DONE Ndmpcopy: 192.168.1.232: Log: DUMP: Deleting "/s0-fas3220win-iscsi1/boot_s0_pwdb_vol/../snapshot_for_backup.3" snapshot. Ndmpcopy: 192.168.1.232: Log: RESTORE: RESTORE IS DONE Ndmpcopy: 192.168.1.232: Notify: restore successful Ndmpcopy: 192.168.1.232: Log: DUMP_DATE is '5698165215' Ndmpcopy: 192.168.1.232: Notify: dump successful Ndmpcopy: Transfer successful [ 0 hours, 13 minutes, 0 seconds ] Ndmpcopy: Done ha-cluster-01> # Check for the new luns in that vserver on the new path! ha-cluster::> lun show -vserver so-fas3220-win-boot-iscsi1 -path /vol/ndmptarget/* Vserver Path State Mapped Type Size --------- ------------------------------- ------- -------- -------- -------- so-fas3220-win-boot-iscsi1<br /> /vol/ndmptarget/boot-s0-pwdb03.lun<br /> offline unmapped windows_2008 50.01GB so-fas3220-win-boot-iscsi1 /vol/ndmptarget/boot-s0-pwdb04.lun offline unmapped windows_2008 50.01GB so-fas3220-win-boot-iscsi1 /vol/ndmptarget/boot-s0-pwdb05.lun offline unmapped windows_2008 50.01GB 3 entries were displayed. |
There it is!
Comments are closed.