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 Software Installation 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).
Checking the Installation Environment

Checking the Installation Environment

Errors That Occur During Installation Environment Check and Solutions to the Errors

When you run an installation environment check script, an error message may be displayed if the parameter settings of the installation environment do not meet the parameter requirements in the FusionInsight Configuration Planning Tool or other errors occur. Table 7-16 and Table 7-17describes the solutions to different errors.

NOTE:

If the installation environment check fails, you can use the precheck/log/precheck_failed.log file to view error messages. For example, run the cat /opt/FusionInsight_SetupTool/precheck/log/precheck_failed.log command.

Table 7-16 Error message and solution

Error Message

Solution

fail to parse configuration file, check the file name or the content of config file.

This error message is displayed because the format of the configuration file checkNodes.Config is incorrect. Export the configuration file checkNodes.Config from the FusionInsight Configuration Planning Tool again.

the ip of host XX is empty.

This error message is displayed because the IP address of XX is not configured in the configuration file checkNodes.Config. Configure the IP address in the configuration file and then export the configuration file again.

cannot transfer the operation scripts to host XX.

This error message is displayed because files cannot be transferred to node XX. Check whether node XX is connected and check whether the firewall configuration and the password are correct.

the XX path value is empty.

This error message is displayed because a blank path exists in the configuration file checkNodes.Config. Export the configuration file checkNodes.Config from the FusionInsight Configuration Planning Tool again.

the partition size format of path XX is not correct, it should be integer or float number.

This error message is displayed because the partition size configured in the configuration file checkNodes.Config is incorrect. Modify the configuration file checkNodes.Config or export it from the FusionInsight Configuration Planning Tool again.

the host name in config file is empty.

This error message is displayed because the host name is not configured in the configuration file checkNodes.Config. Modify the configuration file checkNodes.Config or export it from the FusionInsight Configuration Planning Tool again.

check ntp server failed

This error message is displayed because the NTP time synchronization fails. Perform the following steps to clear the error message:

  1. Check whether the NTP server IP address configured for ntp_server_ip in the Manager installation and configuration file (.ini files generated by the Configuration Planning Tool) is the same as that of the node in the cluster (involving the management IP address, service IP address, and floating IP address).
  2. Run the ping ntp_server_ip command on the current node and check whether the network is connected.
  3. Run the following command on the current node to manually synchronize time:

    sntp -S ${ntp_server_ip}

    NOTE:
    • Commands vary with the OSs and NTP server versions. In this example, an NTP server 4.2.8 or later running the SUSE OS is used.
    • If the Red Hat operating system is used, check whether the NTP service is started. If the NTP service is started, manually stop it before performing the preceding steps.

hostname 'XX' format error, correct format is '^[a-zA-Z0-9]([-a-zA-Z0-9]*[a-zA-Z0-9])*\$'.

This error message is displayed because the format of the host name on the node is incorrect. Change the host name on the node.

the hostname XX in YY does not match ZZ

This error message is displayed because the host name obtained by running the hostname command on the node is inconsistent with the host name in the system configuration file. Change the host name in the system configuration file and make sure that it is consistent with the host name obtained by running the hostname command.

  1. Run the following command to change the host name:

    hostname host name

    For example, to change the host name to Bigdata-OM-01, run the hostname Bigdata-OM-01 command.

  2. Modify the following files.
    • In the Red Hat 6.x environment, run the vi /etc/sysconfig/network command to edit the file, change the value of HOSTNAME to Bigdata-OM-01, save the modification, and exit.
    • In the Red Hat 7.x environment, run the vi /etc/hostname command to edit the file, change the file content to Bigdata-OM-01, save the modification, and exit.
    • In the SUSE environment, run the vi /etc/HOSTNAME command to edit the file, change the file content to Bigdata-OM-01, save the modification, and exit.
  • The '/etc/hosts' file is configured incorrectly.
  • '127.0.0.1 localhost' is not configured in '/etc/hosts'.
  • 'xxip1 hostname1' is mistakenly configured as 'xxip2 hostname2' in '/etc/hosts'.

This error message is displayed because the /etc/hosts file of the current node is not correctly configured. Reconfigure the file by following the requirements:

  • The 127.0.0.1 localhost mapping must be configured.
  • Each host name matches only one IP address (service plane).
  • Each IP address matches only one host name.
  • The file must contain one IP address that matches the local host name.
  • The mapping relationship between the host name and the IP address is consistent with that in the FusionInsight Configuration Planning Tool.

For example:

127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4 
  ::1         localhost localhost.localdomain localhost6 localhost6.localdomain6 

192.168.20.10 Bigdata-OM-01

user 'omm' already exist, please remove user first.

This error message is displayed because user omm exists on the node. Run the userdel command to delete this user manually. Otherwise, the cluster installation fails.

user id '2000' already exist, please remove user first.

This error message is displayed because a user whose ID is 2000 exists on the node. Run the userdel command to delete this user manually. Otherwise, the cluster installation fails.

the cpu cores number 'XX' does not match the config file YY.

This error message is displayed because the number of CPU cores on the node is not the same as that configured in the configuration file. Check whether the number of CPU cores on the node is correct.

the cpu cores number XX format is not correct, it should be integer.

This error message is displayed because the format of the number of CPU cores configured in the configuration file checkNodes.Config is incorrect. Modify the configuration file checkNodes.Config or export it from the FusionInsight Configuration Planning Tool again.

real memory size XX does not match the config file XX.

This error message is displayed because the node memory is not the same as that configured in the configuration file. The following two methods are available:

  • Check the memory size of the node and change the checkNodes.Config file. Alternatively, modify configuration items in the Configuration Planning Tool and export the checkNodes.Config configuration file again.
  • Increase the node memory so that the actual memory is larger than that configured in the configuration file.

the memory size XX format is not correct, it should be integer or float number.

This error message is displayed because the format of the node memory configured in the configuration file checkNodes.Config is incorrect. Modify the configuration file checkNodes.Config or export it from the FusionInsight Configuration Planning Tool again.

timezone check failed, current timezone is 'XX', configured timezone is 'YY'.

This error message is displayed because the time zone is incorrectly configured. Modify the time zone configuration of the node.

  1. Run the following commands to change the time zone. (Use the Beijing time zone as an example.)
    • In the Red Hat environment, run the ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
    • In the SUSE environment, run the ln -sf /usr/share/zoneinfo/Asia/Beijing /etc/localtime
  2. # hwclock -w

real diskSize XX does not match the config file YY.

This error message is displayed because the disk size is not the same as that configured in the configuration file. The following two methods are available:

  • Check the disk size and change the checkNodes.Config file. Alternatively, modify configuration items in the Configuration Planning Tool and export the checkNodes.Config configuration file again.
  • Re-adjust the disk partitions so that the partition capacity of the disk is larger than the disk capacity configured in the configuration file.

the disk size XX format is not correct, it should be integer or float number.

This error message is displayed because the format of the disk size configured in the configuration file checkNodes.Config is incorrect. Modify the configuration file checkNodes.Config or export it from the FusionInsight Configuration Planning Tool again.

partition XX is not mounted by uuid or by-id.

This error message is displayed because the partition is not mounted in UUID or by-id mode. Create the partition again and mount it in UUID or by-id mode.

partition 'XX' size '$partitionSize'(G) does not match the configured size 'YY'(G).

This error message is displayed because the partition size is not the same as that configured in the configuration file. Create the partition again and then check the size of it.

The OS version is not supported. Currently support XX

This error message is displayed because the OS version on the current node is not supported. Change to another node or reinstall a supported OS on the current node.

The Linux kernel version XX is not supported.

This error message is displayed because the Linux kernel version (3.0.101-0.40-default) on the current SUSE node is not supported. Change to another node or reinstall an OS kernel supported on the current node.

Stop other checks on host XX and restore os/kernel first.

This error message is displayed because the OS or kernel check on the current node does not pass and other checks on the node are stopped. Rectify the preceding fault and then use the precheck tool to perform the checks again.

SSH is not installed.

This error message is displayed because the SSH on the current node is unavailable and needs to be reinstalled. Use the preinstall tool to rectify this fault.

Python is not installed.

This error message is displayed because the Python on the current node is unavailable and needs to be reinstalled. Use the preinstall tool to rectify this fault.

The python version is not supported, cur:XX, min:YY, max:ZZ.

This error message is displayed because the Python version of the current node is not supported and needs to be changed. The Python version cannot be earlier than 2.6.5 and be later than 2.7.16.

Failed to obtain the rpm package.

This error message is displayed because the RPM packages lacked on the current node cannot be obtained. This may be caused by an OS fault, for example, the disk space on the node is fully occupied and files cannot be written into.

Need to install: xx

This error message is displayed because some mandatory RPM packages are lacked on the current node and must be supplemented. Use the preinstall tool to rectify this fault.

Failed to find the expect command.

This error message is displayed because the expect tool cannot be used on the current node and the related RPM package needs to be installed. Use the preinstall tool to rectify this fault.

The host name cannot contain only digits.

This error message is displayed because the host name configured on the node is not in correct format. The host name cannot consist of only digits. Change the host name.

'/etc/ntp.conf' is empty / does not exist.

This error message is displayed because the /etc/ntp.conf file does not exist on the current node or the file is empty. Check whether the file is correct.

Insufficient disk space in partition XX.

This error message is displayed because the remaining space of the XX partition on the current node is less than 20% of the total capacity of the partition. Clear the partition space to ensure that the remaining space is sufficient.

The file 'libhadoop.so' exists in '/usr/lib64'.

This error message is displayed because the libhadoop.so file exists in the /usr/lib64 directory, which will affect the startup of the Spark service. Delete this file. For details, see Failed to Start Spark During the Cluster Installation.

The port 20051 is in use. Please disable the port.

This error message is displayed because port 20051 on the current node is used. See DBService Startup Failure During Cluster Installation Due to the Unavailable Port to rectify this fault.

ipmitool command check failed. Please perform the preinstall operation.

This error message is displayed because the ipmitool command does not exist on the current node. Use the preinstall tool to rectify this fault.

iostat command check failed. Please perform the preinstall operation.

This error message is displayed because the iostat command does not exist on the current node. Use the preinstall tool to rectify this fault.

  • Need to install nss-softokn.x86_64 rpm package. Required version is xxx or later.
  • Need to install nss-util.x86_64 rpm package. Required version is xxx or later.
  • Need to upgrade nss-softokn.x86_64 version to xxx or later.
  • Need to upgrade nss-util.x86_64 version to xxx or later.
  • This error message is displayed because the nss-softokn rpm package does not exist on the current node. Install the nss-softokn rpm package of 3.14.3-22 or later.
  • This error message is displayed because the nss-util rpm package does not exist on the current node. Install the nss-softokn rpm package of 3.16.2.3-2 or later.
  • This error message is displayed because the nss-softokn rpm package on the current node needs to be upgraded. The version of nss-softokn must be 3.14.3-22 or later. For details about the upgrade method, contact the OS provider.
  • This error message is displayed because the nss-util rpm package on the current node needs to be upgraded. The version of nss-util must be 3.16.2.3-2 or later. For details about the upgrade method, contact the OS provider.
NOTE:

The two RPM packages depend on other RPM packages of later versions. Upgrade the depended RPM packages to later versions.

For example, upgrade nss-softokn to 3.14.3-22 and nss-util to 3.18.0. Obtain the following RPM packages:

  • nspr-4.10.8-1.el6_6.x86_64.rpm
  • nss-softokn-3.14.3-22.el6_6.x86_64.rpm
  • nss-softokn-freebl-3.14.3-22.el6_6.x86_64.rpm
  • nss-util-3.18.0-1.el6_6.x86_64.rpm

Upgrade the preceding RPM packages to all nodes, for example, the /opt/rpm directory, and run the following commands to install the RPM packages:

cd /opt/rpm

rpm -Uvh *.rpm --nodeps --force

'Defaults requiretty' is configured incorrectly in '/etc/sudoers' the file.

This error message is displayed because Defaults requiretty in the /etc/sudoers file on the current node is incorrectly configured.

Run the vi /etc/sudoers command to edit the /etc/sudoers file. Delete Defaults requiretty.

NOTE:

If the cluster is installed by user omm, change Defaults requiretty to Defaults:omm !requiretty.

'HADOOP_HOME' is configured, and other Hadoop platforms may have been installed.

This error message is displayed because Hadoop programs of other versions have been installed on the current node.

Check whether Hadoop programs of other versions have been installed on the host. If yes, uninstall them.

The '/etc/cron.allow' or '/etc/cron.deny' file is configured incorrectly. User omm does not have the rights to use crontab.

This error message is displayed because the /etc/cron.allow or /etc/cron.deny file on the current node is incorrectly configured. As a result, user omm does not have the running permission on crontab. If both of two files do not exist, you must perform operations as user root. Assign the permission to user omm by performing the following operations:

  • If only the /etc/cron.allow file exists, write user omm into the /etc/cron.allow file.
  • If only the /etc/cron.deny file exists, delete user omm from the /etc/cron.deny file.
  • If both the /etc/cron.allow file and the /etc/cron.deny file exist, write user omm into the /etc/cron.allow file.

The /etc/openldap/ldap.conf configuration file cannot be found. Please restore the file and ensure that the file is correct.

This error message is displayed because the ldap.conf file does not exist.

Copy the file from another OS with the same version to /etc/openldap.

The 'zone_reclaim_mode' is configured incorrectly.

This error message is displayed because zone_reclaim_mode on the current node is incorrectly configured.

  1. Log in to the node as user root, run the following command.

    echo 0 > /proc/sys/vm/zone_reclaim_mode

  2. Run the vi /etc/sysctl.conf command, press Insert to edit the file, add the configuration line vm.zone_reclaim_mode=0 to the configuration file, press Esc to exit the editing, and run :wq to save the configuration and exit.

The NTP version(${ntp_version}) has defects, which may cause an OMS installation failure.

Upgrade the OS NTP version by installing the system patch to meet FusionInsight requirements. For details, see Preparing OS.

The NTP version(${ntp_version}) has defects, which may cause a time synchronization failure between the active OMS and other nodes. Install the system patch to upgrade the system version.

The mount point xxx has noexec attribute.

This error message is displayed because the noexec attribute is configured for the mount point of a key directory, which causes the OMS installation failure. In this case, perform the following operations. Run the following command on the malfunctioning node as user root:

mount -o remount,exec xxx

check DBserviceIP failed

Check on the IP addresses of the active and standby DBservice nodes fails. Locate this problem as follows:

  • Check whether the IP addresses of the active and standby nodes can be pinged.
  • Check whether the active and standby nodes can be connected in SSH mode.
  • Check whether the IP addresses of the active and standby nodes are in the same network segment.

check omsserverIP failed

Check on the IP addresses of the active and standby OMS nodes fails. Locate this problem as follows:

  • Check whether the IP addresses of the active and standby nodes can be pinged.
  • Check whether the active and standby nodes can be connected in SSH mode.
  • Check whether the IP addresses of the active and standby nodes are in the same network segment.
Table 7-17 Warning message and solution

Warnning Message

Solution

The NTP service is not available.

This warning message is displayed because the NTP service on the current node is unavailable. Check whether the NTP service is installed or enabled. If the NTP service is not installed, use the preinstall tool to rectify this fault. If the NTP service is disabled, enable it.

The swap partition is enabled

This warning message is displayed because the swap partition on the node is enabled, which affects the cluster performance. Disable the swap partition.

  1. Run the following command, query the swap partition in the system.

    swapon -s

  2. Run the following command, disable the queried swap partition.

    swapoff Swap partition

    Example:

    swapoff /dev/dm-1

  3. Open the /etc/fstab file, locate the swap-related information, and comment out such information, as shown in the following:
    #/dev/mapper/vg_r160-lv_swap swap swap nosuid 0 0

the partition 'XX' does not exist.

This warning message is displayed because the partition does not exist. Create the partition or modify the checkNodes.Config file.

Failed to restart the sysctl service. Please check whether any invalid configuration item exists in /etc/sysctl.conf.

This warning message is displayed because the sysctl service cannot be restarted. Based on the information printed by the precheck script, check whether invalid configuration items exist in the /etc/sysctl.conf file, and perform related operations to rectify the fault.

You can run the /sbin/sysctl -p command. If error information similar to the following is displayed, you can run the vi /etc/sysctl.conf command to open the /etc/sysctl.conf file, and add a comment tag (#) to the beginning of the corresponding line to comment out the line.

/proc/sys/net12: No such file or directory

The available space of the '/var/log' partition is xxx (GB), which is smaller than the size 130 (GB) of logs to be retained for seven days

This warning message is displayed because the capacity of the /var/log partition on the current node cannot meet the requirements for the storage of logs generated in seven days.

Expand the log disk capacity to ensure that the available capacity is greater than 130 GB.

receive-hashing feature is off. If you install Redis at this node, please run 'ethtool -K eth1 rxhash on' to enable.

Run the following command to enable the receive-hashing function of the NIC (it is assumed that the service IP address is configured on NIC eth1):

ethtool -K eth1 rxhash on

If Cannot set device flag settings: Operation not supported is displayed in the command output, the NIC does not support the receive-hashing function. The warning message can be ignored.

GLIBCXX_3.4.11 was not found, this would cause Impala service unvailable, if you select to install impala components in cluster installation

Run the following command to check whether GLIBC3.4.11 exists in the system:

strings /usr/lib64/libstdc++.so.6 | grep GLIBC | grep GLIBCXX_3.4.11

If no GLIBC3.4.11 is displayed, the GLIBC version does not meet the requirements. Install or upgrade GCC to 4.4.7 or later.

timezone is not configured, skip check.

This warning message is displayed because the time zone is not configured. Run the date command to check the time setting.

  1. Run the following commands to change the time zone. (Use the Beijing time zone as an example.)
    • In the Red Hat environment, run the ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
    • In the SUSE environment, run the ln -sf /usr/share/zoneinfo/Asia/Beijing /etc/localtime
  2. # hwclock -w

partition XX size is not configured, skip check

This warning message is displayed because the size of partition XX is not configured. Modify the checkNodes.Config file.

The available space of the /var/log partition is XX (GB). You are advised to allocate a partition with at least 130(GB) for the log directory

This warning message is displayed because the available space of the partition where the log directory is located is less than 130 GB. Partition the space again and perform the check.

Nic XX is not configured, speed is Unknown

This warning message is displayed because the bandwidth of the XX NIC is not configured. Configure the NIC bandwidth

(eth1 is used as an example).

Run the following command to modify the configuration file of eth1:

vi /etc/sysconfig/network-scripts/ifcfg-eth1

Nic XX invalid config, speed is ${speed}

This warning message is displayed because the bandwidth configuration of NIC XX is incorrect. Modify the bandwidth configuration.

Nic '${nic}' speed configured '${result}Mb/s', speed is not enough

This warning message is displayed because the configured bandwidth of NIC XX is less than the recommended value 100 Mbit/s. Modify the bandwidth configuration.

If nameserver is configured improperly, the network speed is slowed down, which affects the subsequent installation or operations. Check whether the item is correctly configured.

The DNS information is configured in the /etc/resolv.conf file. If the DNS configuration is incorrect, the package download may be slow during the installation. Check whether the DNS configuration in the /etc/resolv.conf file is correct.

The kernel total amount of shared memory pages is too small. The recommended minimum value is ${shmall_suggest}.

This warning message is displayed because the total available size of the shared memory is smaller than the recommended value. Change the value of kernel.shmall. The unit is page.

The kernel.shmall value multiplied by PAGESIZE must be greater than half of the physical memory size.

Run the following command to edit the sysctl.conf file and change the kernel.shmall value:

vi /etc/sysctl.conf

kernel.shmall=1048576 

Run the following command to check the PAGESIZE value:

getconf PAGESIZE

The kernel shared memory is too small. The recommended minimum value is XX.

This warning message is displayed because the kernel shared memory configured on the node is smaller than half of the physical memory (which is the recommended minimum value). Modify the configuration.

The SHMMAX parameter specifies the maximum size (in bytes) of a shared memory segment. You are advised to set this parameter value to half the size of the physical memory. The SHMALL parameter specifies the total available size (in pages) of the shared memory. Set this parameter value to the same as the SHMMAX parameter value (translate the values into the same unit).

Run the following command to open the sysctl.conf file:

vi /etc/sysctl.conf

Press Insert to enter the edit mode. Modify the following parameters. For example, the physical memory of the node is 8 GB.

kernel.shmmax=4294967296 
kernel.shmall=1048576     

Press Esc to exit the editing mode. Then enter :wq to save the configuration and exit.

Network exception to ${targetIP}: ${iLossRate}% packet loss, rtt avg=${iDelay}

This warning message is displayed because the packet loss rate and latency exceed the thresholds (packet loss rate: 0%; latency: 200 ms) due to network connectivity exceptions between XX to YY. Check the network connectivity.

The permission for ${FILE} is ${permission} instead of the expected ${PERMISSION}

This warning message is displayed because the directory or file permissions are abnormal. Change the permissions of the directory or file, for example, directory /tmp.

Command for changing the directory permission to 777: chmod 777 /tmp

Command for add the setgid label to directory /tmp to grant special permissions: chmod o+t /tmp/

Command for changing the directory owner and group: chown root:root /tmp

The owner for ${FILE} is ${owner} instead of the expected ${OWNER}.

The group for ${FILE} is ${group} instead of the expected ${GROUP}

The /usr/sbin/slapd configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This check is applicable only to the nodes running the Red Hat OS. Run the following command to check whether LDAP is installed correctly on the malfunctioning node.

rpm -qa |grep openldap | grep server

  • If no result is displayed, LDAP is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, an installation exception occurs. Uninstall LDAP and install the .rpm package again.

The /etc/rsyslog.conf configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This check is applicable only to the nodes running the Red Hat OS. Run the following command to check whether rsyslog is installed correctly on the malfunctioning node.

rpm -qa |grep rsyslog

  • If no result is displayed, rsyslog is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, the file may be deleted by mistake. Copy the file from another node with the same OS version to /etc.

The /sbin/rsyslogd configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This check is applicable only to the nodes running the Red Hat OS. Run the following command to check whether rsyslog is installed correctly on the malfunctioning node.

rpm -qa |grep rsyslog

  • If no result is displayed, rsyslog is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, an installation exception occurs. Uninstall rsyslog and install the .rpm package again.

The /usr/lib/openldap/slapd configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This check is applicable only to the nodes running the SUSE OS. Run the following command to check whether .rpm package of the LDAP server is installed correctly on the malfunctioning node.

rpm -qa | grep openldap | grep -v client

  • If no result is displayed, .rpm package of the LDAP server is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, an installation exception occurs. Uninstall rsyslog and install the .rpm package again.

The /etc/nscd.conf configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This check is applicable only to the nodes running the SUSE OS. Run the following command to check whether nscd is installed correctly on the malfunctioning node.

rpm -qa | grep nscd

  • If no result is displayed, nscd is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, the file may be deleted by mistake. Copy the file from another node with the same OS version to /etc.

The /usr/sbin/nscd configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This check is applicable only to the nodes running the SUSE OS. Run the following command to check whether LDAP is installed correctly on the malfunctioning node.

rpm -qa | grep nscd

  • If no result is displayed, LDAP is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, an installation exception occurs. Uninstall rsyslog and install the .rpm package again.

The configuration item pam_unix2_key cannot be found in /etc/pam.d/common-session.

Please update /etc/pam.d/common-session and ensure that the configuration item is correct.

This warning occurs only on a node where a SUSE OS is installed. This warning indicates that the pam_unix2_key configuration item is not configured in the PAM configuration file common-session on the node. You need to update the configuration of common-session, and add the correct pam_unix2_key configuration item.

Run the vi /etc/pam.d/common-session command, press Insert to edit the file, add the configuration line sessionrequired pam_unix2.so to the configuration file, press Esc to exit the editing, and run :wq to save the configuration and exit.

The configuration example is as follows:

... 
session  optional   pam_mkhomedir.so 
session required        pam_limits.so 
session required        pam_unix2.so
session optional    pam_ldap.so 
session optional        pam_umask.so

The configuration item pam_limits_key cannot be found in /etc/pam.d/common-session.

Please update /etc/pam.d/common-session and ensure that the configuration item is correct.

This warning occurs only on a node where a SUSE OS is installed. This warning indicates that the pam_limits_key configuration item is not configured in the PAM configuration file common-session on the node. You need to update the configuration of common-session, and add the correct pam_limits_key configuration item.

Run the vi /etc/pam.d/common-session command, press Insert to edit the file, add the configuration line session required pam_limits.so to the configuration file, press Esc to exit the editing, and run :wq to save the configuration and exit.

The configuration example is as follows:

... 
session  optional   pam_mkhomedir.so 
session required        pam_limits.so
session required        pam_unix2.so 
session optional    pam_ldap.so 
session optional        pam_umask.so

The configuration item pam_unix2_key cannot be found in /etc/pam.d/common-password.

Please update /etc/pam.d/common-password and ensure that the configuration item is correct.

This warning occurs only on a node where a SUSE OS is installed. This warning indicates that the pam_unix2_key configuration item is not configured in the PAM configuration file common-password on the node. You need to update the configuration of common-password, and add the correct pam_unix2_key configuration item.

Run the vi /etc/pam.d/common-password command, press Insert to edit the file, add the configuration line passwordsufficient pam_unix2.so use_authtok nullok to the configuration file, press Esc to exit the editing, and run :wq to save the configuration and exit.

The configuration example is as follows:

... 
password    sufficient  pam_unix2.so  use_authtok nullok
password    required    pam_ldap.so try_first_pass use_authtok

password requisite pam_cracklib.so dcredit=0 lcredit=0 ucredit=0 ocredit=0 minlen=8 minclass=3
password required pam_unix.so md5 shadow remember=5 use_authtok 

The configuration item pam_unix2_key cannot be found in /etc/pam.d/common-auth.

Please update /etc/pam.d/common-auth and ensure that the configuration item is correct.

This warning occurs only on a node where a SUSE OS is installed. This warning indicates that the pam_unix2_key configuration item is not configured in the PAM configuration file common-auth on the node. You need to update the configuration of common-auth, and add the correct pam_unix2_key configuration item.

Run the vi /etc/pam.d/common-auth command, press Insert to edit the file, add the configuration line auth sufficient pam_unix2.so to the configuration file, press Esc to exit the editing, and run :wq to save the configuration and exit.

The configuration example is as follows:

... 
auth    required        pam_env.so 
auth    sufficient      pam_unix2.so
auth    required      pam_ldap.so use_first_pass

The configuration item pam_unix2_key cannot be found in /etc/pam.d/common-account.

Please update /etc/pam.d/common-account and ensure that the configuration item is correct.

This warning occurs only on a node where a SUSE OS is installed. This warning indicates that the pam_unix2_key configuration item is not configured in the PAM configuration file common-account on the node. You need to update the configuration of common-account, and add the correct pam_unix2_key configuration item.

Run the vi /etc/pam.d/common-account command, press Insert to edit the file, add the configuration line accountrequired pam_unix2.so to the configuration file, press Esc to exit the editing, and run :wq to save the configuration and exit.

The configuration example is as follows:

... 
account required        pam_unix2.so
account sufficient  pam_localuser.so 
account required    pam_ldap.so use_first_pass

The /etc/sysctl.conf configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This warning message is displayed because the sysctl.conf file does not exist. Copy the file from another node with the same OS version to /etc.

The /sbin/sysctl configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This warning message is displayed because the sysctl command does not exist, which may due to system damages. Install the OS again.

Failed to restart the sysctl service.

Please check whether any invalid configuration item exists in /etc/sysctl.conf.

Check whether sysctl.conf on the malfunctioning node has invalid configuration items. You can log in to the malfunctioning node as the root user and run /sbin/sysctl -p to check the error message and rectify the fault accordingly.

The /usr/bin/openssl configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This warning message is displayed because the openssl command does not exist. Run the following command to check whether openssl is installed correctly on the malfunctioning node.

rpm -qa | grep openssl

  • If no result is displayed, openssl is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, an installation exception occurs. Uninstall openssl and install the .rpm package again.

The openssl version is weakly secure, it cannot support TLSv1.1, TLSv1.2. You are advised to install the recommended OS. Proceeding with the installation indicates that you accepted the security risks.

You are advised to install the recommended OS.

The /etc/openldap/ldap.conf configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This warning message is displayed because the ldap.conf file does not exist. Copy the file from another node with the same OS version to /etc/openldap.

The /usr/bin/ldapsearch configuration file cannot be found.

Please restore the file and ensure that the file is correct.

This warning message is displayed because the ldapsearch command does not exist. Run the following command to check whether .rpm package of the LDAP server is installed correctly on the malfunctioning node.

rpm -qa | grep openldap | grep client

  • If no result is displayed, .rpm package of the LDAP server is not installed. Use the preinstall tool to rectify the fault.
  • If a result is displayed, an installation exception occurs. Uninstall .rpm package of the LDAP server and install the .rpm package again.

diskmgt service has been installed.

Indicates that diskmgt was installed for the node when preinstall was executed. Log in to the node as user root, run the sh /usr/local/diskmgt/script/uninstall.sh -u command to uninstall diskmgt and start preinstall again.

Immediate Preinstallation Failure Due to SSH Permission Disablement of User root

Symptom

A failure message is displayed immediately after you run the ./setuptool.sh preinstall during the Configuring and Checking the Installation Environment.

Possible Causes

The Secure Shell (SSH) permission of user root may have been disabled for the sake of security on the server where FusionInsight HD is to be installed. The Linux OS allows user root to log in over SSH by default.

Procedure

This section uses one server as an example to show how to rectify the fault. The operations are the same for rectifying this fault on other servers.

  1. Use PuTTY to log in to a server where FusionInsight HD is to be installed as user root.
  2. Check whether the sshd_config file contains PermitRootLogin no. If the file contains PermitRootLogin no, the SSH permission of user root is disabled.

    cat /etc/ssh/sshd_config

    • If yes, go to Step 3.
    • If no, the configuration failure is not caused by SSH permission disabling. Rectify this fault using other methods.

  3. Change PermitRootLogin no contained in the sshd_config file to PermitRootLogin yes to enable the SSH permission of user root.

    vi /etc/ssh/sshd_config

  4. Restart the SSH service.

    /etc/init.d/sshd restart

Preinstallation Failure Due to a Formatting Failure

Symptom

A message indicating a formatting failure is displayed immediately after you run the ./setuptool.sh preinstall during configuring and checking the installation environment.

Possible Causes

On the server where FusionInsight HD is to be installed:

  • The number of identifiable hard disks in the OS (except the OS disk) is different from the number of required hard disks in the Preparing OS used for the OS configuration.
  • Partitions have been created or file directories have been mounted on the hard disks except the OS disk.
Procedure

This topic uses one server with five identifiable hard disks (except the OS disk) as an example. The operations are the same for rectifying this fault on other servers.

  1. Use PuTTY to log in to a server where FusionInsight HD is to be installed as user root.
  2. Run the lsscsi command to query the number of identifiable hard disks in the OS. For example, in the partition configuration file shown in following, y is displayed in five rows in the required column, and the disks queried by the lsscsi are sdb, sdc, sdd, sde, and sdf.

    lsscsi

    NOTE:
    • By default, Red Hat 6.4 OS does not have the lsscsi command. Therefore, manually load the lsscsi RPM package.
    • If the number of hard disks in the query result is less than the number required in the partition configuration file, for example, only sdb and sdc are displayed in the result but at least five disks are required in the configuration file(except the OS disk), please add new disks.

  3. Run the df -h  command to check whether the file directories have been mounted to the disks queried in Step 2. If the disks queried in Step 2, such as sdb, sdc, sdd, sde, and sdf, are displayed, go to 3.a. If they are not displayed, go to 3.b.

    1. The configuration does not meet the OS configuration requirements. Run the umount command to unmount all file directories on sdb, sdc, sdd, sde, and sdf disks.
    2. Go to Step 4.

  4. Run the cat /etc/fstab command to check whether the fstab file contains only the OS partitions that the FusionInsight HD cluster requires and the partitions are in UUID mode (Red Hat or CentOS). If other partitions exist, go to 4.a. Otherwise, go to 4.b.

    1. The preinstallation configuration requirements are not met. Delete the UUID information other than that of the OS partitions in the fstab file, save the file, and restart the system. For example, delete the related rows of partitions /srv/BigData/zookeeper, /srv/BigData/namenode, /srv/BigData/dbdata_om, /srv/BigData/journalnode, and /srv/BigData/hadoop/data1 from the fstab file.
      UUID=6322eb41-2aad-4c9d-8784-840b2fc043e9 /                       ext4    defaults,errors=panic,noatime        1 1
      UUID=d8aeaa19-7b3f-4414-b6a7-5e5b7cdb25dd /opt                    ext4    defaults,errors=panic,noatime        1 2
      UUID=1ce196a9-1b65-4f44-ae17-b5d49bf3c48d /srv/BigData            ext4    defaults,noatime        1 2
      UUID=331945ff-33e7-47ec-a8fc-059c8f0326cb /tmp                    ext4    defaults,errors=panic,noatime        1 2
      UUID=7835c51c-0c3f-4520-82c1-74dab4a78a99 /var                    ext4    defaults,errors=panic,noatime        1 2
      UUID=426caf8d-2922-4c86-870f-a13d9fad9727 /var/log                ext4    defaults,errors=panic,noatime        1 2
      tmpfs                   /dev/shm                tmpfs   defaults        0 0
      devpts                  /dev/pts                devpts  gid=5,mode=620  0 0
      sysfs                   /sys                    sysfs   defaults        0 0
      proc                    /proc                   proc    defaults        0 0
      UUID=3d1545cf-71bd-4dd6-a3a1-72e7d0c42785 /srv/BigData/zookeeper ext4 defaults,noatime,nodiratime 1 2
      UUID=8034a232-4fd9-4c2c-8ab5-0c49dde4806e /srv/BigData/namenode ext4 defaults,noatime,nodiratime 1 2
      UUID=dc957048-cace-4093-9cf4-c4bdcdde8e11 /srv/BigData/dbdata_om ext4 defaults,noatime,nodiratime 1 2
      UUID=ccd622ee-4d89-44ff-9cb3-7401ca82a1ab /srv/BigData/journalnode ext4 defaults,noatime,nodiratime 1 2
      UUID=d239c80a-d6ba-499b-ba2f-48cb1e3d2d00 /srv/BigData/hadoop/data1 ext4 defaults,noatime,nodiratime 1 2
      NOTE:

      The fstab file content in a SUSE environment is similar to the following:

      /dev/disk/by-id/scsi-36234567890abcde01c423217443897fa-part1 /                    ext3       acl,user_xattr,errors=panic,noatime        1 1
      /dev/disk/by-id/scsi-36234567890abcde01c423217443897fa-part7 /opt                 ext3       acl,user_xattr,errors=panic,noatime        1 2
      /dev/disk/by-id/scsi-36234567890abcde01c423217443897fa-part6 /srv/BigData         ext3       acl,user_xattr,noatime        1 2
      /dev/disk/by-id/scsi-36234567890abcde01c423217443897fa-part2 /tmp                 ext3       acl,user_xattr,errors=panic,noatime        1 2
      /dev/disk/by-id/scsi-36234567890abcde01c423217443897fa-part3 /var                 ext3       acl,user_xattr,errors=panic,noatime        1 2
      /dev/disk/by-id/scsi-36234567890abcde01c423217443897fa-part5 /var/log             ext3       acl,user_xattr,errors=panic,noatime        1 2
      proc                 /proc                proc       defaults              0 0
      sysfs                /sys                 sysfs      noauto                0 0
      debugfs              /sys/kernel/debug    debugfs    noauto                0 0
      usbfs                /proc/bus/usb        usbfs      noauto                0 0
      devpts               /dev/pts             devpts     mode=0620,gid=5       0 0
      UUID=3d1545cf-71bd-4dd6-a3a1-72e7d0c42785 /srv/BigData/zookeeper ext4 defaults,noatime,nodiratime 1 2
      UUID=8034a232-4fd9-4c2c-8ab5-0c49dde4806e /srv/BigData/namenode ext4 defaults,noatime,nodiratime 1 2
      UUID=dc957048-cace-4093-9cf4-c4bdcdde8e11 /srv/BigData/dbdata_om ext4 defaults,noatime,nodiratime 1 2
      UUID=ccd622ee-4d89-44ff-9cb3-7401ca82a1ab /srv/BigData/journalnode ext4 defaults,noatime,nodiratime 1 2
      UUID=d239c80a-d6ba-499b-ba2f-48cb1e3d2d00 /srv/BigData/hadoop/data1 ext4 defaults,noatime,nodiratime 1 2
    2. The preinstallation configuration requirements are met. Rectify this fault in other ways.

How Do I Format a Disk After Successfully Running the preinstall Script?

Scenario

This section describes how to format a disk after the preinstall script is successfully executed and the disk plan is adjusted.

Procedure
  1. Use PuTTY to log in to any node that needs to be partitioned as user root.
  2. Run the following command to uninstall diskmgt:

    /usr/local/diskmgt/script/uninstall.sh -u

    NOTE:

    In this command, the -u parameter is used to automatically uninstall the partitions mounted during the preinstall script execution and delete the partition records from /etc/fstab. If this parameter is not specified, only the diskmgt is uninstalled. In this case, you need to manually unmount the partitions and delete records from /etc/fstab.

  3. Run the df -h command to check whether all partitions mounted during the preinstall script execution are unmounted. If any partition is not unmounted, manually unmount it.
  4. Run the vi /etc/fstab command to check whether records of all partitions automatically mounted during the preinstall script execution are deleted. If they are not, delete them manually.
  5. Log in to other nodes that need to be partitioned in sequence and perform Step 2 to Step 4.
  6. Use the configuration planning tool to generate a new configuration file, and run the preinstall script again.
Download
Updated: 2019-05-17

Document ID: EDOC1100074555

Views: 6118

Downloads: 6

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