000-331 - Power Systems Sales for AIX and Linux
Go back to IBM
An administrator wants to shut down a node which has a resource group from the HMC. Which shutdown option results in a node shutdown without failover?
Operating System Immediate
An administrator created a 2-node PowerHA cluster. Each node is in a separate location, and each location has its own network. After starting the cluster, the network interfaces did not come up. Which action will correct this problem?
Allow multicast network packets to flow between the two locations.
When implementing PowerHA in a cross site LVM configuration, which logical volume option is required?
What cluster service settings can be defined in SMIT on one node of a cluster and automatically updated on all other nodes?
Verify Cluster Prior to Startup
A 2-node cluster has three multi-node disk heartbeat (MNDHB) networks and data volume groups on separate disks. Application resource groups are online on Node1. What happens if Node1 loses access to two of the three MNDHB disks?
Application resource groups remain on Node1
A resource group is configured with a startup policy of "Online on First Available Node" and a service IP address which uses IPAT via aliasing. An administrator needs a custom event to run only during a failover. Which pre-defined event would be best to specify the custom event?
A PowerHA 6 cluster uses IPAT with aliasing and disk heartbeat. The administrator noticed the standby node crashes during a failover test. There are entries related to the Dead Man Switch (DMS) timeout in the AIX error log. Which action is recommended to reduce the risk of DMS timeouts?
Set Failure detection rate of diskhb network module to slow.
Which operation can be done by DARE?
Change application server name
Which are names of default file collections in PowerHA 7?
After migrating to PowerHA 7 and restarting the cluster for the first time, the following error is reported: Starting Cluster Services on node: Jessica This may take a few minutes. Please wait... Jessica: cl_rsh had exit code = 13, see cspoc.log and/or clcomd.log for more information What is the most common cause of this error?
etc/cluster/rhosts is incorrectly configured.
What must be verified before configuring GLVM?
TCP/IP communication between sites is possible.
18.104.22.168 22.214.171.124 !REQD!ALL 126.96.36.199 !REQD!ALL 188.8.131.52 Given the netmon.cf file above, a single-adapter configuration must recognize a network_down event if a node loses access to_______________:
100.127.9OR 100.12 7.10
An administrator performed a DARE operation in their cluster configuration. During this operation one of the cluster nodes halted. The node halt was an isolated problem, however, as a result the DARE operation did not complete. When the failed node was re-integrated to the cluster, the following error was received when trying 20 to perform the DARE operation again: cldare: A lock for a Dynamic Reconfiguration event has been detected. Which action will resolve the problem?
Use the PowerHA problem determination function to "Release Locks Set By Dynamic Reconfiguration" and re-try the DARE operation
An administrator has configured a 2-node PowerHA 6 cluster, Nodel and Node2, with no non-IP networks. Sometime after configuring the cluster, the administrator noticed that Node2 appeared to have been shut down. After restarting Node2 the administrator analyzed the cluster log files in an attempt to determine the cause of the problem. The clstrmgr.debug log file contained the following information: Based on the above information, which of the following could explain why Node2 was shut down?
Node isolation occurred causing a domain merge.
When using PowerHA 7, a shared volume group (VG) must be configured as___________.
Enhanced concurrent VG
In which default file can an administrator specify special options for nfs exported filesystems in PowerHA?
In which configuration would an AIO Cache LV be required?
When using mirror pools
An administrator wants to change the nodename of one node while a PowerHA 6 cluster is running. Which actions will achieve this?
Stop cluster services on the node. Change the name of the node. Verify and synchronize cluster. Start cluster services.
Given the following cluster topology information, why is this PowerHA 6 cluster configuration unsuitable for migration to PowerHA 7?
IPAT via replacement is not supported.
Failure to plan for which element will have a negative impact on PowerHA cluster availability, when using virtual I/O resources on the cluster nodes?
Planning for update of Virtual I/O Servers
PowerHA automatic error notification for shared data volume groups traps on which specific error?
A customer has completed installing PowerHA 7 filesets. Which task, or tasks, should be completed next?
Verify and synchronize the cluster
A cluster administrator wants to be sure that specific resource group will failover to the node with the most available processor resources. Which configuration supports the requirement?
Use dynamic node priority based on PctTotalTimeIdle.
A PowerHA 6 cluster has missed too many heartbeats during a large disk write operation, and has triggered the deadman switch. Which tunable can help ensure that HA Cluster Manager continues to run?
I/O Pacing and syncd frequency
In order to utilize PowerHA 7 SAN based communications in a VIO Server environment, what task must be performed on the AIX client nodes?
Add a virtual Ethernet adapter and specify VLAN 3358.
An administrator has a 2-node cluster and has chosen the C-SPOC option to bring a resource group offline. The resource group was stable on the primary node prior taking resource offline, however it has gone into an error state. The administrator corrects the error and issues the clruncmd to continue cluster processing. What will happen to the resource group?
The resource group goes offline.
An administrator notices that PowerHA has moved a resource group from one node to another. He looks on the vacated node, but doesn't see any problems reported in the application controller or application monitor logs. Why are the logs missing the pertinent information?
PowerHA has been restarted.
Consider the following PowerHA DLPAR configuration for a cluster, and LPAR profile on the HMC for a standby node in that cluster: How should the standby node profile be reconfigured so that an application runs with the minimum 13 hardware resources?
Minimum processing units = 0.6 and both Desired and Maximum processing units = 1.5
An administrator added a filesystem on a shared volume group using C-SPOC. The cluster is active and the resource group is on line. What is the effect of the administrator action?
The filesystem is available immediately.
Which application dependency can cause problems if the application is installed on a shared disk?
An application which depends on CPU ID
Company policy requires logical volumes to be mirrored across separate disks and controllers for resilience. An administrator needs to create a 2-node cluster with the existing partitions, LPAR1 and LPAR2.The cluster will host an application with data on a shared volume group. LPAR1 and LPAR2 are currently hosted on a single Power 770. Each partition is assigned two integrated SAS controllers, each of which has internal hard disk drives (HDDs) that are currently used for rootvg. Which group of actions will create a shared volume group that complies with company storage policy?
Assign LPAR1 &LPAR2 two fibre adapters each. Zone disk from two independent storage servers to each LPAR fibre adapter. Create shared volume group on new HDDs, ensuring the logical volume copies are across separate HDDs & storage servers.
administrator is using PowerHA 7 to define a new cluster using SMIT option "Setup a Cluster, Nodes and Networks" and encountered the following error message: What is the root cause of the problem?
The /etc/cluster/rhosts are not populated correctly
What does the gsclvmd daemon do?
Enhanced concurrent mode logical volume sizes are kept synchronized across nodes
In PowerHA 7 cluster, how many multicast IP addresses are required?
One for the entire cluster
Given the information from hacmp.out below, why is the cluster in config_too_long status?
The application did not shut down correctly, which caused the rg_move operation to exceed the time limit to complete.
The database is running on node1 and application is running on node2. The application cannot run unless database is active and the application must be stopped before the database. What actions will ensure the application and database run correctly?
Configure Parent/Child Dependency and set Parent/Child Dependency between Resource Groups
After installing PowerHA packages, what is the next step when configuring the cluster?
Add cluster boot addresses, service addresses and hostnames to /etc/hosts.
In a PowerHA 6 environment, what is recommended to ensure I/O disk-write buffers are flushed and to reduce chance of deadman switch timeouts?
Change "syncd" frequency to 10 seconds
To provide a secure cluster environment, an administrator will use VPN tunnels for communication between the nodes in a cluster. What must be defined on the nodes for communication across the VPN tunnel?
Persistent IP Labels
An active, active (mutual takeover) 2-node PowerHA cluster is currently up and stable. Both nodes are located in separate buildings, one kilometer apart. Node2 in SiteB suddenly powers off. On reboot of the failed node, the cluster administrator checks the error log and discovers the node had been halted. The network team confirms there was a temporary loss of the IP network between sites. Why did Node2 halt?
The cluster became partitioned. On recovery of the IP network, Node1 sent a halt signal to Node2.
A PowerHA 7 cluster is running on two fully virtualized LPARs. An administrator simulates a network failure on the cluster by running the "ifconfig enX down" command on all network interfaces on one node. The resource group moved, and clstat shows the cluster as stable on both nodes. What prevented the resource group from being active on both nodes?
The nodes are able to connect to each other via dpcom interface.
An administrator is configuring a PowerHA 6 cluster to use disk heartbeat as a non IP network. A shared volume group, dhbvg, has been configured using a single disk, hdisklO, on NodeA and NodeB. The administrator believes that the prerequisites have been met, but would like to test connectivity of the disk heartbeat network before adding the network into the cluster. 4 How can the connectivity of this disk heartbeat network be tested?
On NodeA run the command/usr/sbin/rsct/bin/dhb_read-p hdisk10-r On NodeB run the command /usr/sbin/rsct/bin/dhb_read-p hdisk10-t
An administrator wants to assign a tape drive to a resource group for ease of management and control. The tape drive is already attached to two cluster nodes. Node1 and Node2. On Node1, the tape drive has a logical device name of rmt1. Node1 also has a tape drive, rmt0, that is not controlled by the cluster. On Node2 the tape drive has a logical device name of rmt0. Which of the following must the administrator do before adding the tape drives to the cluster resource group?
Redefine the logical device name rmt0 as rmt1 on Node2.
The addition of a service IP address to a resource group fails to pass verification and synchronization. What is the most likely reason?
The service IP address is incorrectly defined or is missing in/etc/hosts on at least one node in the resource group.
A 2-node cluster configuration has an application server named appl. The appl server has custom application monitors named appmon_a and appmon_b. The setting of appmon_a and appmon_b follows: appmon_a: monitoring process=A, monitor interval=20, stabilization interval=60, restart count=1 appmon_b: monitoring process=B, monitor interval=10, stabilization interval=60, restart count=0 If process B fails 30 seconds after process A fails and successfully restarts, what is the most amount of time it will take for a failover event to begin?
about 10 seconds
When PowerHA SystemMirror 7.1 is installed on AIX 7.1, what RSCT component does Cluster Aware AIX (CAA) replace?
What must be done before defining a pre-event command for a predefined cluster event?
Define a custom event command using the SMIT "Add a Custom Event" dialog
An administrator has a 4-node PowerHA cluster which was successfully verified and synchronized on Sunday. On Wednesday the cluster was changed by a different administrator. 28 The cluster must be returned to the Sunday configuration and resynchronized. How would an administrator restore the cluster to the original configuration when the changes made on Wednesday are not known?
Restore from the cluster snapshot automatically generated at the last synchronize.
An administrator unintentionally changed the name of an online resource group on one node of a 2-node cluster. The cluster has not been synchronized since the change was made. How can the cluster configuration revert to the previous configuration without affecting the online resource group?
Restore HACMP configuration database from active configuration.
What are the two types of application monitors that can be created?
Process Monitors and Custom Monitors