Skip to content

Latest commit

 

History

History
221 lines (168 loc) · 11 KB

README.md

File metadata and controls

221 lines (168 loc) · 11 KB
Raw

Linstor Storage Driver (unofficial)

Description

Community driven full-feature Linstor storage driver for OpenNebula

Read the blog article for more details...

Comparsion to addon-linstor

Why not simply use the official Linstor driver? I was trying an official Linstor driver for OpenNebula, before I made a decidion to write my own implementation. And I didn't like it because of reasons, that are mostly described below

  • Bash-written. Unlike official Linstor driver which is written in python, this driver is written in bash. Any driver action is just a conventional bash script, which is calling standard shell-commands, meaning these scripts can be easily updated or extended.
  • Uses OpenNebula native library This driver uses standard OpenNebula library which provides simplicity developing and debugging driver actions, e.g. you will always see what exact command was unsuccessful from the VM log, if something went wrong.
  • Does not requires external dependings. Official driver needs configured linstor-client and extra python-bindings on every compute node. This driver has a central managment model from the OpenNebula frontend node, so it requires only jq and linstor-client installed on the frontend node, and no external components on compute nodes. It's also worth noting, that you still need linstor-satellite and drbd9 module to build linstor-cluster.
  • Usually supports newer versions of OpenNebula.
  • Can work with any backend, even without snapshots support.

Compatibility

This add-on is compatible with:

  • OpenNebula 5.8+
  • Linstor server 1.7.0+
  • Linstor client 1.1.1+

OpenNebula Installation

Requirements

  • Installed jq and linstor on the OpenNebula frontend.
  • Configured Linstor cluster and access to it from the OpenNebula frontend server.
  • All OpenNebula hosts should have DRBD9 module installed and be registred in Linstor cluster as satellite nodes.

Installation steps

Only for OpenNebula 5.10 and below:

  • Uncomment LEGACY_MONITORING=1 option in linstor_un.conf

Update oned.conf:

Modify VM_MAD section for the kvm driver:

  • add save=save_linstor_un and restore=restore_linstor_un overrides to local actions.
 VM_MAD = [
     NAME           = "kvm",
-    ARGUMENTS      = "-t 15 -r 0 kvm",
+    ARGUMENTS      = "-t 15 -r 0 kvm -l save=save_linstor_un,restore=restore_linstor_un",
 ]

Modify TM_MAD section for the one_tm driver:

  • add linstor_un transfer drivers list.
 TM_MAD = [
     EXECUTABLE = "one_tm",
-    ARGUMENTS = "-t 15 -d dummy,lvm,shared,fs_lvm,qcow2,ssh,ceph,dev,vcenter,iscsi_libvirt"
+    ARGUMENTS = "-t 15 -d dummy,lvm,shared,fs_lvm,qcow2,ssh,ceph,dev,vcenter,iscsi_libvirt,linstor_un"
 ]

Modify DATASTORE_MAD section for the one_datastore driver:

  • add linstor_un to datastore mads and system datastore tm drivers list.
 DATASTORE_MAD = [
     EXECUTABLE = "one_datastore",
-    ARGUMENTS  = "-t 15 -d dummy,fs,lvm,ceph,dev,iscsi_libvirt,vcenter -s shared,ssh,ceph,fs_lvm,qcow2,vcenter"
+    ARGUMENTS  = "-t 15 -d dummy,fs,lvm,ceph,dev,iscsi_libvirt,vcenter,linstor_un -s shared,ssh,ceph,fs_lvm,qcow2,vcenter,linstor_un"
 ]

Add new TM_MAD_CONF section:

TM_MAD_CONF = [
    NAME = "linstor_un", LN_TARGET = "NONE", CLONE_TARGET = "SYSTEM", SHARED = "yes",
    DS_MIGRATE = "YES", DRIVER = "raw", ALLOW_ORPHANS="yes"
]

Add new DS_MAD_CONF section:

DS_MAD_CONF = [
    NAME = "linstor_un", PERSISTENT_ONLY = "NO",
    MARKETPLACE_ACTIONS = "export"
]

Update vmm_execrc:

  • add kvm-linstor_un to the LIVE_DISK_SNAPSHOTS list
-LIVE_DISK_SNAPSHOTS="kvm-qcow2 kvm-ceph"
+LIVE_DISK_SNAPSHOTS="kvm-qcow2 kvm-ceph kvm-linstor_un"

LINSTOR Configuration

  • Install Linstor-satellite and DRBD9 kernel module on all your compute nodes.
  • Nodes in linstor must have same name like for OpenNebula hosts.
  • Linstor-client should be installed and working on control-plane nodes.
  • Create Resource group: 
    linstor resource-group create opennebula --place-count 2 --storage-pool thinlvm
linstor volume-group create opennebula
    Resource group must have storage-pool assigned and single volume-group created.

OpenNebula Configuration

To use your Linstor cluster with the OpenNebula, you need to define a System and Image datastores. Each Image/System Datastore pair will share same following Linstor configuration attributes:

Attribute Description Mandatory
NAME The name of the datastore YES
CLONE_MODE snapshot - will create snapshot for instantiate VMs. copy - create full copy of image, this is default mode. NO
BRIDGE_LIST Space separated hosts list used for transfer operations. Copy data between images and etc. Default: all hosts. NO
LS_CONTROLLERS Comma separated linstor controllers list for establish connection. NO
LS_CERTFILE SSL certificate file. NO
LS_KEYFILE SSL key file. NO
LS_CAFILE SSL CA certificate file. NO
RESOURCE_GROUP Resource group to spawn the resources. YES 1
NODE_LIST Space separated hosts list to place replicas. Replicas will always be created on all these hosts. YES 1
LAYER_LIST Comma separated layer list to place replicas. NO
PROVIDERS Comma separated providers list to place replicas. NO
REPLICAS_ON_SAME Space separated aux-properties list to always place replicas on hosts with same aux-property. NO
REPLICAS_ON_DIFFERENT Space separated aux-properties list to always place replicas on hosts with different aux-property. NO
REPLICA_COUNT Number of replicas for creating new volumes. YES 1
CHECKPOINT_REPLICA_COUNT Number of replicas for save checkpoint file for suspend and offline migration process. NO
DO_NOT_PLACE_WITH Space separated resources list to avid placing replicas on same place with them. NO
DO_NOT_PLACE_WITH_REGEX Regular expression to avoid placing replicas on same place with targeted resources. NO
STORAGE_POOL Space separated storage pool names to place replicas. YES 2
DISKLESS_POOL Diskless pool to place diskless replicas. Default: DfltDisklessStorPool. NO
PREFER_NODE yes - try to place and copy the data on the node that will afterwards be used by the VM NO
ENCRYPTION yes - will enable encryption during volume creation. NO

1 - only one attribute required
2 - required if no RESOURCE_GROUP specified

Note: You may add another Image and System Datastores pointing to other pools with different allocation/replication policies in Linstor.

Create an Image Datastore

Apart from the previous attributes, that need to be the same as the associated System Datastore, the following can be set for an Image Datastore:

Attribute Description Mandatory
NAME The name of the datastore YES
DS_MAD linstor_un YES
TM_MAD linstor_un YES
DISK_TYPE BLOCK YES
STAGING_DIR Default path for image operations in the bridges NO

An example of datastore:

cat > images-ds.conf <<EOT
NAME="linstor-images"
TYPE="IMAGE_DS"
RESOURCE_GROUP="opennebula"
DISK_TYPE="BLOCK"
DS_MAD="linstor_un"
TM_MAD="linstor_un"
RESTRICTED_DIRS=/
SAFE_DIRS=/var/tmp
EOT

onedatastore create images-ds.conf

Create a System Datastore

System Datastore also requires these attributes:

Attribute Description Mandatory
TYPE SYSTEM_DS YES
TM_MAD linstor_un YES

Create a System Datastore in Sunstone or through the CLI, for example:

cat > system-ds.conf <<EOT
NAME="linstor-system"
TYPE="SYSTEM_DS"
STORAGE_POOL="data"
RESOURCE_GROUP="opennebula"
PREFER_NODE="yes"
CHECKPOINT_REPLICA_COUNT="1"
TM_MAD="linstor_un"
EOT

onedatastore create system-ds.conf

Development

To contribute bug patches or new features, you can use the github Pull Request model. It is assumed that code and documentation are contributed under the Apache License 2.0.

More info:

Author