1Z0-884 - Oracle Solaris Cluster 3.2 System Administrator Certified Professional Exam
Go back to Oracle
The Sun Cluster 3.2 framework software has already been installed. In order to support a clusterunaware application, which additional software must be installed?
data service agent
When configuring IPMP groups for use in the Sun Cluster 3.2 environment, which two statements are true? (Choose two.)
Adapters within the same IPMP group must be of the same transport type.
Adapters on different subnets must be in different IPMP groups.
The cluster administrator wants to disable the BREAK signal on the serial port. Which file needs to be edited?
While setting up logical network interfaces within IPMP groups, what is the maximum number of virtual network interfaces for each physical adapter?
You can have a maximum of 8192 virtual interfaces for each physical adapter.
The cluster administrator is running SunCluster 3.2 on x86 machines running Solaris 10 update 3, and needs to boot one of the x86 machines into non-cluster mode. To which menu item does the cluster administrator add -x' using the GRUB editor?
The cluster framework contains a kernel heartbeat module. Which two functions are provided by kernel heartbeats? (Choose two.)
detection of node faults
monitoring cluster transport interfaces
Your Oracle database is configured with Sun Cluster HA-Oracle using a resource called oracle-rs, and you need to perform a backup of the database files using the tar utility. Which operation should be performed?
#clrs disable oracle-rs perform backup #clrs enable oracle-rs
Your customer has just created a new Veritas Volume Manager volume named oravol in the already registered disk group oradg. Now they are trying to create a UFS filesystem on top of the volume and keep getting the following error message after executing the command: newfs /dev/vx/rdsk/oradg/oravol /dev/vx/rdsk/oradg/oravol: No such device or address Which two statements describe the cause of the error? (Choose two.)
You did not run cldg sync oradg.
The global namespace does NOT know about this new volume on the other nodes.
Sun Cluster 3.2 requires an IPMP configuration for a logical host resource. Which action does NOT lead to a persistent IPMP configuration?
Configure the IPMP group with ifconfig at the command prompt.
What path should a system administrator add to the MANPATH variable to find the manual pages for the commands in the Sun Cluster 3.2 CLI?
Your customer has asked you to create a Veritas Volume Manager volume. They want the volume to be RAID level 1+0. The name of the group is oradg and they want the name of the volume to be oravol, and 3 Gbytes in size. Which three statements are correct? (Choose three.)
RAID 1+0 requires at least 4 disks
The volume oravol will be a layered volume
vxassist -g oradg make oravol 3g layout=stripe-mirror
What is the name of the cluster console package that the cluster administrator will load onto the administrative workstation?
A customer has a three-node cluster configured as a pair-plus-one topology. The customer asks about whether they need to use any quorum devices, and if so how many. Which answer best describes how the administrator will respond?
You do NOT need any quorum devices, although it would make your cluster more highly available. You should configure TWO quorum devices for highest availability.
Which two are requirements for Solaris Volume Manager shared storage disksets in Sun Cluster 3.2? (Choose two.)
They must contain disks only in the shared storage arrays.
If they contain disks that are in more than one array, then those arrays must be attached to exactly the same set of nodes.
What functionality does the global device implementation provide?
Storage connected to any cluster node is available to all cluster nodes using standard device semantics.
Cluster framework daemons are started with boot scripts on Solaris 9 and with SMF on Solaris 10. Which statement is true?
The daemons start automatically at cluster boot time regardless of whether you are running Solaris 9 or Solaris 10.
You have been running your four-node cluster with three private networks for a while and now you want to change the IP address space used for the entire cluster transport from 172.16.x.x to 192.168.5.x. How can you accomplish this change?
Boot all nodes to non-cluster mode, enable root ssh or rsh between them, and run clsetup from one node only to change the private network information.
When setting up a three node, shared all topology using Sun Cluster 3.2, the cluster administrator wants to use a shared disk as the quorum device. How many quorum disk devices should be installed and how many votes will each quorum device provide?
One quorum disk device providing two votes.
Which is important when installing Sun Cluster with Solaris Volume Manager to manage the boot disks?
The /global/.devices/[email protected], /global/.devices/[email protected] should have unique volume names for each node in the cluster.
Your customer has chosen to use Solaris Volume Manager for the shared disks. A diskset named ds1 has already been created with node1 and node2 being able to master the diskset. The customer would like to build a mirrored volume to contain their application data. The diskset ds1 contains two 36 Gbyte disk drives using did devices d4 and d8. The customer would like to use as volume names d101 and d102 as the submirrors and d100 as the mirror. The customer would like to create two soft partitions named d10 and d11, with d10 5gb and d11 10gb in size. Which sequence of commands will accomplish this task?
metainit -s ds1 d101 1 1 /dev/did/rdsk/d4s0 metainit -s ds1 d102 1 1 /dev/did/rdsk/d8s0 metainit -s ds1 d100 -m d101 metattach -s ds1 d100 d102 metainit -s ds1 d10 -p d100 5g metainit -s ds1 d11 -p d100 10g
Suppose you have a Veritas Volume Manager disk group named appdg with default settings registered using: #cldg create -t vxvm -n node1,node2 appdg. You are now asked to configure the following: appdg should be online on node1 by default, and node1 should take appdg automatically back when it joins the cluster. Which identifies the correct action to accomplish this task?
cldg set -p preferenced=true -p failback=true appdg
In the Sun Cluster 3.2 environment, where can you store the actual data for RAC databases?
on raw devices using the Solaris Volume Manager multi-owner diskset feature
When you want to fail over your application between zones, you have to understand the principles of integrating non-global zones in a cluster. Which statement below is incorrect for the zones integration?
All the zones have to be booted with the HA Container agent.
Suppose you have a resource apache-rs and you have been asked to make sure that this resource can be restarted on the same node, but must NOT trigger a failover to another node. Which command will accomplish this task?.
clrs set -p Failover_mode=RESTART_ONLY apache-rs
Which two network adaptors can be used simultaneously as both public and private interfaces? (Choose two.)
You have a two-node cluster configured. Node one is out of the cluster. You make changes to node two (add resource groups and disk device groups). You now shut down node two and try to boot node one, which fails due to amnesia prevention. Which two keep node one from starting the cluster? (Choose two.)
Persistent reservations on the quorum device.
Node one's key has been previously preempted from the quorum device.
Which IPMP restriction for multiple adapters in an IPMP group is no longer a restriction in Solaris 10?
Every adapter must have a dedicated test IP address.
Sun Cluster 3.2 supports using Veritas Volume Manager. Which two versions are currently supported on the SPARC platform? (Choose two.)
As cluster administrator, you need to run an application that fails over between non-global zones, and you will have a LogicalHostname resource to control IP address failover between the zones. The application needs to be able to resolve the IP address in order to run properly. Where do you need to include the /etc/hosts entry for the IP address, assuming you are using no other name service?
In both the global zones and non-global zones.
Which command will shut down all nodes of the cluster?
cluster shutdown -y -g0
As a precautionary measure the cluster administrator wants to have a backup of the cluster configuration. Which command can be used to provide a usable backup of the entire cluster configuration?
usr/cluster/bin/cluster export -o /var/tmp/configfile.xml
Which tasks are performed by the Sun Cluster 3.2 framework components?
The RGM is used to start and stop the application on the cluster nodes.
Which two steps are required to be able to boot your servers into the cluster? (Choose two.)
install the framework packages
configure the Sun Cluster 3.2 software with scinstal
What is the minimum number of private interconnects recommended to build a two-node cluster using Sun Cluster 3.2?
How is a disk repartitioned when it is added to a diskset?
A small portion of the drive is mapped to slice 7 or 6 and the remainder to slice 0.
A system administrator has a two-node cluster to which it becomes necessary to add a third node. Once the hardware has been connected, which two steps are required to add the new node to the cluster? (Choose two.)
Run scinstall on the third node to add it to the existing cluster.
On one existing cluster node use the clsetup or claccess command to add the new hostname to the authorized node list.
Which two name service lookup entries are modified with an entry of cluster in /etc/nsswitch.conf for Sun Cluster 3.2? (Choose two.)
A three-node cluster contains three resource groups which have these Nodelists: one-rg has a Nodelist: node1 node2 node3 two-rg has a Nodelist: node2 node3 node1 three-rg has a Nodelist: node3 node1 node2 The Fallback property on each of the resource groups is set to FALSE. Following a maintenance outage in which node1 and node2 have been shut down and rebooted into the cluster with node1 joining the cluster before node2: Where will the resource groups be located, and which command should be run to move the resource groups to their preferred nodes?
one-rg will be on node3 two-rg will be on node3 three-rg will be on node3 clrg remaster +
Which is an advantage of Sun Cluster 3.2 integration with SMF?
An application managed by SMF need not be rewritten to be used as a cluster service.
A new cluster has been installed and the customer needs to create a HA-NFS service. Which steps have to be executed ?
Register SUNW.nfs dataservice, create a resource group, create a SUNW.HAStoragePlus resource, and create NFS resource.
If the cluster administrator wants to restart a process, they have to be sure it is NOT one of the cluster daemons. Which two daemons listed below can you send a kill -HUP signal to without affecting Sun Cluster 3.2? (Choose two.)
The Sun Cluster 3.2 framework provides Global Storage Services. Which is applicable?
Shared disks that may have different logical names on different nodes are given a unique cluster-wide DID number.
You are setting up a cluster. You want to mirror your storage devices using Veritas Volume Manager from one array to another. Which software RAID volume management application would you use if your database application was Oracle RAC?
Veritas Cluster Volume Manager
Which two hardware platforms are supported by Sun Cluster 3.2? (Choose two.)
Which component of Sun Cluster 3.2 provides global naming consistency for shared storage?
Which two commands are recommended ways to shut down a node? (Choose two.)
shutdown -y -i0 -g0
A customer has a four-node cluster configured as a scalable topology. No quorum devices have been configured. How many nodes must be present in order for the cluster to be able to offer any of its application services?
Which is a reason for using link-state testing rather than "ping testing" for public network adapters in IPMP groups in Solaris 10?
You do NOT have to consume public network IP addresses for test addresses if you use linkstate testing.
Identify the required steps in order to enable automatic reboot on a two-node cluster (nodes mars and earth), if all the monitored disk paths from that node are broken and there is another node that still has working paths to at least one of the shared devices.
1) Identify all DID numbers that represent disks only connected to one node. 2) For each device, perform cldev unmonitor d 3) clnode set -p reboot_on_path_failure=enabled mars 4) clnode set -p reboot_on_path_failure=enabled earth
Before you can configure your cluster, what must you do on each node to prepare for the global devices device files?
You must have an unused partition on your boot device.