No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

FusionInsight HD 6.5.0 Administrator Guide 02

Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Changing IP Addresses

Changing IP Addresses

Scenario

After the FusionInsight cluster is deployed, new IP addresses are required for network segment migration or node maintenance.

Administrators can change IP addresses in the following scenarios. In addition, after the IP addresses are changed, you do not need to reinstall the cluster or restore data. The switching between dual-plane isolation networking and single-plane networking is not supported.

  • Change the IP addresses of OMSServer.
  • Change the mediator IP address of the active and standby OMSServer.
  • Change the floating IP address, network port, mask, and gateway of the OMSServer.
  • Change the IP address, network port, mask, and gateway of the Web service.
  • Change the IP address of the NTP server.
  • Change the IP address of the service plane on the node.
  • Change the IP address of the management plane on the node.
  • Change the component configuration IP address.
  • Change the SSO IP address.

This topic provides guidance for system administrators to change IP addresses. Change the IP address by following the required steps; otherwise, the proper running of FusionInsight HD may be affected and data exceptions may occur. Figure 11-2 shows the process for changing the IP address.

Figure 11-2 Process for changing the IP address

Impact on the System

  • To change the IP address, you need to stop the FusionInsight HD cluster first. Services are unavailable before the cluster is restarted.
  • If the Manager on two management nodes is deployed, IP address changing script cannot be performed on the standby management node and can be performed only on the active management node.
  • In the dual-plane networking, after the node service plane IP addresses or component (OMSServer and Web service are not included) floating IP addresses are changed, you need to download the client again.
  • In the single-plane networking, after the node IP addresses or component (OMSServer and Web service are not included) floating IP addresses are changed, you need to download the client again.
  • After the SSO IP address, node management plane IP addresses, node service plane IP addresses, or component (including OMSServer and Web service) floating IP addresses are changed, the backup files of Manager and LdapServer cannot be used to restore the system, but the backup files of DBService, HBase, HDFS, and Hive are not affected.
  • After changing the web service IP address, use the new IP address to log in to FusionInsight Manager.
  • If the web service IP address is changed, the old web page certificate cannot match the new IP address. In this case, FusionInsight Manager can be accessed but the certificate may not be trusted. You are advised to change the CA certificate. For details, see Configuring the CA Certificate.
  • After changing a node IP address, use the new IP address to log in to the node.
  • After changing the service plane IP address, modify the IP address configuration on components.
  • After changing a node IP address, modify related parameters.
    • After configuring the cross-cluster mutual trust, change the service IP address of KrbServer. Modify peer_realms of KrbServer on the peer cluster. Check whether the IP address of Kerberos on the peer cluster and the IP address of the standby HBase cluster in the DR scenario are correctly configured. If the IP addresses are configured incorrectly, change them to valid ones.
    • If the service IP address of NameNode in the standby cluster is changed after cross-cluster replication is configured, modify dfs.namenode.rpc-address.haclusterX.remotenn1 and dfs.namenode.rpc-address.haclusterX.remotenn2 of Yarn in the active cluster. If a RemoteHDFS(DistCp) backup task is created, modify Target NameNode IP address.
    • In the scenario where the cluster has been configured with avro or a source is bound to an IP address, if the IP address of Flume is changed, change the node IP address in ${BIGDATA_HOME}/FusionInsight_Porter_6.5.0/*_*_Flume/etc/properties.properties to the new IP address. If the client sends data to the node with the previous IP address, change the IP address in the configuration file of the client to the new IP address.
    • If the IP address of FTP-Server is changed, change the IP address in the configuration file of the client to the new IP address.
    • After you change node IP addresses of HBase, check whether new IP addresses have been configured for the client. If there is no new IP address, update the IP addresses of the client.
    • If IP addresses have been set for configuration items hbase.replication.allowedIPs and hbase.regionserver.dns.nameserver, check whether they are valid.

Prerequisites

  • The changed IP address needs to comply with the planning rules. For the planning data of the preparations for installation, see the Software Installation.
  • If an OMSServer on two management nodes is deployed, the active and standby management nodes must work properly and can be accessed during IP address changing.
  • Before changing the IP address for a node, stop all roles installed on the node. After changing the IP address, start the roles again. See Starting and Stopping All Instances on a Host.
  • The passwords of users root and omm for logging in to the cluster node operating system have been obtained.

Procedure

Check the environment.

  1. Use PuTTY to log in to the active management node as user root using the IP address of the active management node.
  2. Run the following command to disable logout on timeout:

    TMOUT=0

  3. Run the following command to check whether the OMS status is normal.

    sh ${BIGDATA_HOME}/om-server/om/sbin/status-oms.sh

    In the command output, check whether the values of all ResHAStatus parameters are Normal.

    • If yes, go to Step 4.
    • If no, the task is complete. See Fault Management.

  4. Check whether the node whose HAActive parameter value is active is the current node.

    • If yes, go to Step 5.
    • If no, the current node is the standby management node. Log in to the active management node. Go to Step 5.

  5. On FusionInsight Manager, choose Cluster > Service to check whether Running Status of all components is Normal.

    • If yes, go to Step 6.
    • If no, the task is complete. See the Fault Management.

Export the IP address configuration file

  1. Run the following command on PuTTY of the active management node to switch to user omm:

    su - omm

  2. Run the following command to go to the related directory:

    cd ${BIGDATA_HOME}/om-server/om/tools/modifyip/

  3. Run the following command to export the IP configuration file ClusterInfo.xml in the current directory:

    ./export.sh

    If the following information is displayed, the ClusterInfo.xml configuration file is successfully exported:

    begin to get the cluster info and export...
    get oms info success.
    get node info success.
    get configuration success.
    export xml success.

Stop the cluster.

  1. Check whether the Manager is installed on two management nodes.

    • If yes, run the following command in the PuTTY session on the active management node to disable the active/standby switchover, and then go to Step 10.

      ${BIGDATA_HOME}/om-server/OMS/workspace0/ha/module/hacom/tools/ha_client_tool --ip=127.0.0.1 --port=20013 --forbidswitch --name=product --time=60

    • If no, go to Step 10.

  2. On FusionInsight Manager, choose Homepage > Stop and enter the password of the current login administrator user.

    In the displayed page, click OK to stop the cluster.

  3. Log in to the active management node as user omm, and run the following command to stop the active OMS:

    sh ${BIGDATA_HOME}/om-server/om/sbin/stop-oms.sh

    If the following information is displayed, the OMS is successfully stopped:

    stop HA successfully.

  4. Log in to the standby management node as user omm, and run the following command to stop the standby OMS:

    sh ${BIGDATA_HOME}/om-server/om/sbin/stop-oms.sh

    If the following information is displayed, the OMS is successfully stopped:

    stop HA successfully.

Change the node IP address

  1. Determine whether to change the management IP address or service IP address of the node.

    • If yes, log in to the node as user root. On Linux of the node, change the node IP address, and run the service network restart command to restart the network service. Then perform Step 14.
      Table 11-4 Examples of changing the IP address in different operating systems

      Operating System

      Procedure

      SUSE

      1. Run the ifconfig command to check the name of the NIC whose IP address needs to be changed.
      2. Run the vi/etc/sysconfig/network/ifcfg-name of the NIC interface to be modified command.
      3. Press I to enable editing.
      4. After you change the IP address, press Esc and enter :wq! to save the settings and exit.

      Red Hat/CentOS/Euler OS

      1. Run the ifconfig command to check the name of the NIC whose IP address needs to be changed.
      2. Run the vi/etc/sysconfig/network-scripts/ifcfg-name of the NIC interface to be modified command.
      3. Press I to enable editing.
      4. After you change the IP address, press Esc and enter :wq! to save the settings and exit.
      NOTE:
      • If the IP address of the active management node is changed, use the new IP address to log in to the active management server.
      • If the system is running slowly after the service IP addresses of LdapServer service are modified and network services are restarted, see What Do I Do If Linux Is Running Slowly When Changing an IP Address? for more information.
    • If no, you do not need to change the IP address in the Linux operating system. Go to Step 14.

Modify the IP address configuration file

  1. In the PuTTY session on the active management node, change each IP address in the configuration file.

    Run the following command as user omm:

    vi ${BIGDATA_HOME}/om-server/om/tools/modifyip/ClusterInfo.xml

    NOTE:

    If the management IP address of the Linux operating system on the active management node is changed, use the new IP address to log in to the active management node as user omm.

    Use a number sign (#) to separate the old and new IP addresses, for example, oldIPValue#newIPValue. Table 11-5 describes the parameters in the configuration file. After modifying the parameters, save the configuration file and exit.
    Table 11-5 Parameters in the ClusterInfo.xml configuration file

    Operation

    Description

    Change the IP addresses of OMSServer.

    Change the IP addresses of the local and peer nodes. local1-local4 and peer1-peer4 are supported.

    When you reset these parameters, you also need to change the management plane IP address on the management node.

    Effect after IP address changing:

    <oms>
    ...
    <local1>192.168.0.30#10.168.0.30</local1>
    <peer1>192.168.0.24#10.168.0.24</peer1>
    ...
    </oms>

    Change the mediator IP address of the active and standby OMSServer.

    Set mediatorip under oms.

    Effect after IP address changing:

    <oms>
    ...
    <mediatorip>192.168.0.254#10.168.0.254</mediatorip>
    ...
    </oms>
    NOTE:

    Specifies the third-party arbitration IP address, which must be an IP address that can be connected and is permanently valid. You are advised to use the gateway IP address (IP address of the management plane).

    Change the floating IP address, network port, mask, and gateway of the OMSServer.

    If the OMSServer floating IP address and the Web service IP address are the same, change both of them.

    The network port is configured after the OMSServer is started and can be set to any value as required. The gateway and mask must be changed to the gateway and mask that correspond to the IP address. Set ipaddr, intf, mask, and gw under omip.

    Effect after IP address changing:

    <oms>
    ...
    <omip>
    <ipaddr>192.168.0.111#10.168.0.111</ipaddr>
    <intf>eth0:oms#eth0:oms1</intf>
    <mask>255.255.255.0#255.255.0.0</mask>
    <gw>192.168.0.254#10.168.0.254</gw>
    </omip>
    ...
    </oms>

    Change the IP address, network port, mask, and gateway of the Web service.

    If the Web service IP address and the OMSServer floating IP address are the same, change both of them.

    If the Web service IP address and the SSO IP address are the same, change both of them.

    The network port is configured after the OMSServer is started and can be set to any value as required. The gateway and mask must be changed to the gateway and mask that correspond to the IP address. Set ipaddr, intf, mask, and gw under wsip.

    Effect after IP address changing:

    <oms>
    ...
    <wsip>
    <ipaddr>192.168.0.111#10.168.0.111</ipaddr>
    <intf>eth0:ws#eth0:ws1</intf>
    <mask>255.255.255.0#255.255.0.0</mask>
    <gw>192.168.0.254#10.168.0.254</gw>
    </wsip>
    ...
    </oms>

    Change the IP address of the NTP server.

    Set ntpip under oms. The value of ntpip must be an existing IP address that provides NTP services.

    <oms>
    ...
    <ntpip>192.168.1.0#10.168.1.0</ntpip>
    ...
    </oms>

    Change the IP address of the service plane on the node.

    Set businessip under node.

    Effect after IP address changing:

    <node>
    <businessip>192.168.0.30#10.168.0.30</businessip>
    </node>
    NOTE:

    After changing the service plane IP address, modify the IP address configuration on components.

    Change the IP address of the management plane on the node.

    Set nodeip under node.

    Effect after IP address changing:

    <node>
    <nodeip>192.168.0.30#10.168.0.30</nodeip>
    </node>

    Change the component configuration IP address.

    Distinguish the levels of services, roles, and instances.

    Effect after IP address changing:

    <service>
    ...
    <role>
    <instance>
    <name>dbservice.floatip</name>
    <desc>DBserver foloat ip</desc>
    <value>192.168.0.30#10.168.0.30</value>
    </instance>
    </role>
    ...
    </service>

    Change the SSO IP address.

    Set ssoip under oms.

    If the SSO IP address and the Web service floating IP address are the same, change both of them.

    Effect after IP address changing:

    <oms>
    ...
    <ssoip>192.168.0.111#10.168.0.111</ssoip>
    </oms>

Run the script for changing the IP address.

  1. Run the following command for the new IP address to take effect in the database and configuration file on the active management node. Enter Y or y or press Enter to continue for the IP address to be automatically changed.

    sh ${BIGDATA_HOME}/om-server/om/tools/modifyip/modify.sh

    If the following information is displayed, the IP address is changed successfully and the system will start the active and standby OMSs:

    Modify ip address successfully.
    NOTE:

    Run the command sh ${BIGDATA_HOME}/om-server/om/tools/modifyip/modify.sh -q to change the IP address without confirmation.

Replace the HA certificate.

  1. Check whether the management IP address of the active or the standby management node is changed.

    • If the management IP address is changed and Manager on two management nodes is installed and the FusionInsight HD cluster is networked based on dual planes, replace the HA certificate and perform Step 17. If a Single-node Manager is installed or the single-plane networking is adopted, perform Step 20.
    • If the management IP address is not changed, perform Step 20.

  2. Run the following command to generate an HA certificate and perform automatic replacement:

    sh ${BIGDATA_HOME}/om-server/om/sbin/replacehaSSLCert.sh

    Enter password as prompted and press Enter.

    Please input ha ssl cert password:
    NOTE:

    If the administrator replaces the HA certificate by referring to Replacing HA Certificates, the password must be same as that used for applying a root certificate. If the administrator has never replaced the HA certificate, the default password is Changeme_123.

    The HA certificate is replaced successfully if the following information is displayed:

    [INFO] Succeed to replace ha ssl cert.

  3. Run the following command to restart the OMS.

    sh ${BIGDATA_HOME}/om-server/om/sbin/restart-oms.sh

    The following information is displayed:

    start HA successfully.

  4. Use PuTTY to log in to the standby management node as user omm with the IP address of the standby management node. Repeat Step 17 to Step 18.

    Run sh ${BIGDATA_HOME}/om-server/om/sbin/status-oms.sh to check whether HAAllResOK of the management node is Normal and whether FusionInsight Manager can be logged in to again. If yes, the operation is successful. Perform Step 20. If the node status cannot be restored to Normal, see Fault Management.

Replace the HA certificate.

  1. Log in to FusionInsight Manager, choose Homepage > More > Synchronize configurations to synchronize the cluster configurations.

    In the displayed window, click OK. Then the system starts to synchronize the configuration. After the system displays Operation succeeded.

  2. Choose Homepage > Start to restart the cluster for the modified IP address to take effect.

    NOTE:

    After the cluster is started, alarms may not be cleared at once. Wait for a period. If the alarms are not automatically cleared, see Fault Management to resolve the problem.

  3. Check whether the Manager is installed on two management nodes.

    • If yes, run the following command in the PuTTY session on the active management node to enable the active/standby switchover:

      ${BIGDATA_HOME}/om-server/OMS/workspace0/ha/module/hacom/tools/ha_client_tool --ip=127.0.0.1 --port=20013 --cancelforbidswitch --name=product

    • If no, the task is complete.

Download
Updated: 2019-05-17

Document ID: EDOC1100074522

Views: 7383

Downloads: 12

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next