[DRBD-user] Migration from drbdmanage to linstor failed

Denis drbdsys at made.net
Mon Jan 14 10:02:19 CET 2019


Hello,

Yesterday i did try the migration on a 4 node cluster with proxmox 5.1, 
latest version of drbd and drbdmanage.

I installed linstor controller and satellite on all nodes and activate 
controller on the first.

I logged to the controller, created and reviewed dmmigrate.sh and 
executed it.


### Nodes ###
linstor node create --node-type Combined pve212 10.12.150.2
linstor node create --node-type Combined pve213 10.12.150.3
linstor node create --node-type Combined pve211 10.12.150.1
linstor node create --node-type Combined zern1 10.12.150.4

### Storage ###
linstor storage-pool-definition create drbdpool
linstor storage-pool create lvm pve212 drbdpool drbdpool
linstor storage-pool create lvm pve213 drbdpool drbdpool
linstor storage-pool create lvm pve211 drbdpool drbdpool
linstor storage-pool create lvm zern1 drbdpool drbdpool

### Resource: vm-101-disk-1 ###
linstor resource-definition create --port 7000 vm-101-disk-1
linstor resource-definition drbd-options --allow-two-primaries yes 
vm-101-disk-1
linstor volume-definition create --vlmnr 0 --minor 108 vm-101-disk-1 
52428800K
linstor volume-definition set-property vm-101-disk-1 0 OverrideVlmId 
vm-101-disk-1_00
linstor volume-definition set-property vm-101-disk-1 0 
AllowLargerVolumeSize true
linstor resource create --node-id 1 --storage-pool drbdpool pve212 
vm-101-disk-1
linstor resource create --node-id 2 --storage-pool drbdpool pve213 
vm-101-disk-1
linstor resource create --node-id 0 --storage-pool drbdpool zern1 
vm-101-disk-1


The last command timed out after 5 minutes with no errors.

Satellite is running on all 4 nodes.

Then I tried to continue:

### Resource: vm-108-disk-1 ###
linstor resource-definition create --port 7002 vm-108-disk-1
linstor resource-definition drbd-options --allow-two-primaries yes 
vm-108-disk-1
linstor volume-definition create --vlmnr 0 --minor 110 vm-108-disk-1 
52428800K
linstor volume-definition set-property vm-108-disk-1 0 OverrideVlmId 
vm-108-disk-1_00
linstor volume-definition set-property vm-108-disk-1 0 
AllowLargerVolumeSize true
linstor resource create --node-id 1 --storage-pool drbdpool pve211 
vm-108-disk-1

...timed out after 5 minutes...

I made other steps with same result and then rolled back to drbdmanage.

BTW: I did see resize and UUID modification during migration process, 
can it be dangerous restoring old drbdmanage .res files?


If you need some other information, please tell me.

I have drbdmange and linstor running now, but linstor resourced are not 
included by drbd and proxmox still uses drbdmanage plugin


Thank you,

Denis




More information about the drbd-user mailing list