单选题An application running on a production cluster failed over to the backup node， however， the DLPARintegration scripts failed. On inspection， it was discovered that the scripts failed as a result of /var being 100% full. The administratoradded more spac

第1题：

Company.com has non-IBM fibre attached storage. The storage is connected to a node in the  cluster that has multiple adapters and virtual paths. There is a concern about detecting the loss of one of the fibre adapters since there is no indication to the administrator that one has failed.  What feature of HACMP can be sued to assist with this issue？（）  

• A、 Disk heartbeat
• B、 Error notification
• C、 Event notification
• D、 Application monitor

第2题：

Company.com has a four-node cluster （nodes ’A’， ’B’， ’C’， and ’D’）. The nodes are set up in  Resource group pairs. Node ’A’ falls over to Node ’B’， and Node ’C’ falls over to Node ’C’. During a maintenance windows， it was discovered that the six resource groups on Node ’A’ are too much to be adequately handled by Node ’B’.  What could be done to better utilize the available resources during fallover？（）  

• A、 Implement Work Load manager so that the application monitors for additional resource      requirements.
• B、 Add a fifth node to function as a load leveler in order to control the dispatching of the resource jobs.
• C、 Implement Dynamic Node Priorities in the resource groups in order to find a node with suitable system resources.
• D、 Reduce the number of resource groups on Node ’A’ to 1 in order to ensure faster and more robust fallover to a single system.

第3题：

A customer who has a two node PowerHA cluster， with each node being at a different site and using cross siteLVM， plans to add a third heartbeat to further reduce the chance of data divergence or partitioned cluster. Theheartbeat will make use of spare adapters in the nodes. However， the heartbeats must go via an externalnetwork requiring static routes to be added for cluster functionality.How can this be achieved？（）

• A、Configure the static route in the /etc/rc.net file
• B、Using application Start/Stop scripts to manage the static route
• C、Configure Pre and Post event commands to manage the static route
• D、Using smitty route command to create the static route and update the ODM

第4题：

What does the Fast Connection Failover feature of Active GridLink for RAC provide?（）

• A、instant notification of a RAC node failure so applications never have to retry a transaction that was sent to a node that failed during the transaction
• B、near-instant notification of the failure of a RAC node failure that minimizes the possibility connection to a failed node being provided to an application
• C、application level notification of a failed RAC node such that an application can retry a transaction if required
• D、faster failover for Multi Datasources
• E、guaranteed transaction high availability when interacting with an Oracle RAC Database

第5题：

An administrator was notified that a resource group has failed over from the home node， Node1 to a standbynode， Node2 and that Node1 had been powered off. Upon restarting Node1 the administrator checked the AIXerror log and the cluster.log Which of the following is most likely the cause of the failover？（）

• A、AIX was shut down
• B、A fatal corruption occurred in the gsclvmd daemon
• C、The topsvcs daemon was stopped by the system resource controller
• D、The grpsvcs process was stopped by the system resource controller

第6题：

第7题：

第8题：

第9题：

第10题：

第11题：

第12题：

第13题：

An enterprise has three nodes： -Node ’A’ is a production database， -Node ’B’ is a production  Application server， and -Node ’C’ is a test and development node. Node ’C’ is used by six programmers to develop and the production application as business critical and would like to make it highly available with HACMP. Application development and testing is not business critical.  Which of the following describes the best solution？（）  

• A、 There is not an adequate solution because HACMP should not be configured in this environment without a dedicated standby node.
• B、 Nodes ’A’ and ’B’ should be configured in a cluster with mutual fallover of two cascading resource groups to provide both nodes will deliver acceptable response time with both resource groups.
• C、 Nodes ’A’ and ’C’ should be configured in one cluster， and Nodes ’B’ and ’C’ in another cluster to prevent both the database and application server from falling over to Node ’C’ at the same time.
• D、 Nodes ’A’， ’B’， and ’C’ should be configured in a cluster with Node ’C’ the hot standby for both the database and application server resource groups. Workload Manager （WLM） must be configured on database and application developers from over utilizing the CPU resources.

第14题：

During initial cluster planning session， a requirement is mandated that application monitoring will  Be used to monitor the production database and three application servers. Since all the application servers rely on the database being active， only one resource group has been planned to manage the starting/stopping of the cluster resources.  How ill the initial design impact the use of application monitoring？（）  

• A、 Only one application server per resource group can be monitored.
• B、 Process monitoring is part of RSCT and requires no special scripts.
• C、 Any application can be monitored as long as it has an application server.
• D、 Having all application servers in one resource group improves recovery time.

第15题：

A cluster administrator has decided to add a cluster verification to the weekly maintenance scripts  That are run from corn.  When a cluster verification is performed，which of the following will require a custom verification method？（）  

• A、 HACMP ODM object classes
• B、 HACMP Error Notification
• C、 HACMP Topology & Resources
• D、 HACMP Application Server Start/Stop scripts

第16题：

You are a network administrator for your company. You install Windows Server 2003 on two servers named Server1 and Server2. You configure Server1 and Server2 as a two-node cluster.   You configure a custom application on the cluster by using the Generic Application resource， and you put all resources in the Application group. You test the cluster and verify that it fails over properly and that you can move the Applications group from one node to the other and back again.   The application and the cluster run successfully for several weeks. Users then report that they cannot access the application. You investigate and discover that Server1 and Server2 are running but the Application group is in a failed state.   You restart the Cluster service and attempt to bring the Application group online on Server1. The Application group fails. You discover that Server1 fails， restarts automatically， and fails again soon after restarting. Server1 continues to fail and restart until the Application group reports that it is in a failed state and stops attempting to bring itself back online.   You need to configure the Application group to remain on Server2 while you research the problem on Server1.   What should you do？  （）

• A、 On Server2， configure the failover threshold to 0.
• B、 On Server2， configure the failover period to 0.
• C、 Remove Server1 from the Possible owners list.
• D、 Remove Server1 from the Preferred owners list.

第17题：

第18题：

第19题：

第20题：

第21题：

第22题：

第23题：