VERITAS CLUSTER-1 (BASICS)

VERITAS CLUSTER-1

VCS COMPONENTS:

RESOURCES:

As per VCS, A resource is an entity that may be brought online, offline, or monitored on a particular system.

Resource Types:

There is a defined set of resources for relevant match. Like “NIC Resource” type to be selected for Ethernets, “IP Resource” Type to be selected for IP addresses.

A resource type can be considered as a “TEMPLATE” that defines the characteristics or attributes required to define a particular resource of that type.

PHYSICAL RESOURCES:

·         Nodes

·         Shared Storage

·         Networking Components

A: Between cluster nodes and application clients and external

   systems.

           B: Between cluster nodes inter communication, called Heartbeat.

LOGICAL RESOURCES:

These can be Hardware/Software, like Disk Group, File system, NIC, IP, applications.

Resource Categories:

PERSISTENT:

Never turned off (NIC), if we want to down the cluster then also persistent resources should up. Some of OS daemons are also treated as “persistent”, they are must have for OS operation and never taken down.

These resources cannot be brought online or taken offline. For example, a network interface card cannot be started or stopped, but it is required to configure an IP address. Failure of a Persistent resource triggers a service group failover.

ON-ONLY:

VCS starts On-Only resources, but does not stop them. For example, VCS requires NFS daemons to be running to export a file system. VCS starts the daemons if required, but does not stop them if the associated service group is taken offline.

ON-OFF:

VCS starts and stops On-Off resources as required. For example

File system- mount it / unmounts it

Disk Groups- import it / deport it

RESOURCE-DEPENDENCIES:

Resource dependencies determine online-offline order of resources. We can understand it in very simple manner,

Let’s I have to go Mumbai…. How I will reach there…?

By Train… Flight… Car… Bus…

Right…!!!

Let’s opt a simple one which suits our budget…JJ

I.e. Train,

Well, mode of transportation is selected. Is it enough to reach Mumbai…??

No… I need to reach station from home also by any mean to catch train.

Let’s book a cab for station. These are my Resources.

Now all set,

CAB àà STATION àà TRAIN àà MUMBAI

From my flow chart at least one thing is clear that any how I cannot alter the order of my resources.

Without cab I can’t reach station, and train only available at station so I must reach station on time then I can catch the train to reach Mumbai.

Super…

So all are dependent upon each other to fulfill one goal, i.e. to reach Mumbai.

Now, if I talk in terms of VCS then there is a “Parent-Child” relationship between resources. Let’s see how…

Cab is child for Station, Station is child for train, and Train is child for Mumbai.

Back in reverse order, Mumbai is parent for all, Train is parent for Station and Station is parent for cab.

So, what is significance of this “Parent-Child” relationship?

Child resources must be online before parent get online.

Parent must be offline before child get offline.

ATTRIBUTES:

VCS components are configured using attributes. Attributes contain data about the cluster, systems, service groups, resources, resource types, agent, and heartbeats if using global clusters. For example, the value of a service group’s System List attribute specifies on which systems the group is configured and the priority of each system within the group.

Everything in VCS has attributes, and attributes are basically variables which hold some information.

RESOURCE ATTRIBUTE:

Individual resource properties are called Resource Attributes.

Each resource has a set of required attributes that must be defined in order to enable VCS to manage resources.

Ex: “mount” is command for UNIX, and it is resource for VCS. This resource is defined in 4 attributes to operate.

1.  Mount Point (dir of mount)

2.  Block device (device to mount)

3.  FSTYPE (type of file system to mount)

4.  FSCK Opt (option for fsck command)

Here, let’s say “resource” is “test_mount”.

Attribute:      Value:

MountPoint      /test_dir

BlockDevice     /dev/vx/dsk/testdg/testdatavol

FSType          vxfs

MountOpt        -ro

FsckOpt         -y

*default MountOpt is “-rw”, we can specify if any other is required. Other wise no need to mention.

Now we understand the “Resources”, but there are two things to consider.

What the VCS will do with resources…??    

Answer is SERVICE GROUP

Who will manage these resources…??  

Answer is AGENTS

Let’s see them one by one.

SERVICE GROUP:

Service group is virtual container that enables VCS to manage an application service as a single unit.

SG is made up of all the HW/SW resources related to particular application.

So all the resources required to run a single service are gathered under one roof and treated as single unit, which is further referred as Service Group.

If any resource failed, the entire SG will collapse. A single node can contain many SG’s depending upon configuration. All SG’s on one node operate independently. If one SG failed, it does not affect the operation of other SG.

SERVICE GROUP TYPES:

FAILOVER (ACTIVE-PASSIVE):

SG runs on one node at a time. Most common configuration.

PARALLEL (ACTIVE-ACTIVE):

SG runs on more than one node simultaneously. Complex application required with capability to run on more than one node at a time without data corruption.

Ex- oracle real application cluster (RAC).

HYBRID:

This SG is a combination of Failover & Parallel SG’s. From VCS 4.0 and above.

This SG behaves as a Failover group within a defined set of Nodes and a Parallel SG within a different set of nodes.

Replicated data clusters (RDC’s), which are based on Veritas Volume Replicator.

AGENTS:

Agents are processes that provides logic to manage resources, VCS has one agent per resource type. Ex- a single IP agent manages all IP resources.

When the agent is started, it obtains the necessary configuration information from VCS. It then periodically monitors the resources, and updates VCS with the resource status. VCS agents are multithreaded, meaning a single VCS agent monitors multiple resources of the same resource type on one host.

Agents control resources using a defined set of actions, called “ENTRY POINTS”.

Following are the entry points/ Agent Operation for Agents,

Agent Operation:

Online:    Brings a specific resource ONLINE from an OFFLINE state.

Offline:   Takes a resource from an ONLINE state to an OFFLINE state.

Monitor:   Tests the status of a resource to determine if the resource

is online or offline.

Clean:     Cleans up after a resource fails to come online, fails to

go offline, or fails while in an ONLINE state. Forced

termination.

Action:    Performs actions that can be completed in a short time

(Typically, a few seconds), and which are outside the scope of traditional activities such as online and offline.

Info:      Retrieves specific information for an online resource.