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

FusionCloud 6.3.1.1 Troubleshooting 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).
Common Operations

Common Operations

Rebuilding VMs on Other Hosts

Scenarios

If VMs on a host cannot run due to a host fault and these VMs do not have HA enabled, rebuild the VMs on other hosts.

This method applies only to KVM VMs, such as management VMs if FusionCompute is connected in the private cloud scenarios.

Prerequisites

You have obtained the IDs of the VMs to be rebuilt.

Procedure
  1. Use PuTTY to log in to FusionSphere OpenStack through the IP address of the External OM plane.

    The username is fsp and the default password is Huawei@CLOUD8.
    NOTE:
    • The system supports login authentication using a password or private-public key pair. If a private-public key pair is used for login authentication, seeUsing PuTTY to Log In to a Node in Key Pair Authentication Mode.
    • For details about the IP address of the External OM plane, see the LLD generated by FCD sheet of the xxx_export_all.xlsm file exported from FusionCloud Deploy during software installation, and search for the IP addresses corresponding to VMs and nodes.The parameter names in different scenarios are as follows:
      • Cascading layer in the Region Type I scenario : Cascading-ExternalOM-Reverse-Proxy, Cascaded layer : Cascaded-ExternalOM-Reverse-Proxy.
      • Region Type II and Type III scenarios : ExternalOM-Reverse-Proxy.

  2. Run the following command to switch to user root, and enter the password of user root as prompted:

    su - root

    The default password of user root is Huawei@CLOUD8!.

  3. Run the TMOUT=0 command to disable user logout upon system timeout.
  4. Import environment variables. For details, see Importing Environment Variables.
  5. Run the following command to query whether the VM is booted from an image:

    nova show vm_id

    vm_id specifies the ID of the VM to be rebuilt.

    Take a note of the OS-EXT-SRV-ATTR:hypervisor_hostname value in the command output. This value is the ID of the host accommodating the faulty VM.

    Check the image field in the command output.

    • The VM is started from the secondary volume if information similar to the following is displayed. In this case, go to step 7.
      | image | Attempt to boot from volume - no image supplied|
    • The VM is started from the image if information similar to the following is displayed. Information in the brackets is the image ID. In this case, contact technical support for assistance.
      | image | cirros (d0bd0551-07f2-45f6-8516-f481e0152715)|
      NOTE:

      While you rebuild a VM on another host, if the VM is started from an image, the original data in the system and temporary disks of the VM will be lost. Contact technical support before performing this high-risk operation, and install services again after the rebuilding.

  6. Check whether the OS is faulty based on the following description:

    NOTE:
    • Linux VMs fail to be logged in to and the file system is damaged, which cannot be rectified.
    • Windows VMs fail to be logged in to, blue screen of death (BSOD) occurs, and the file system damage cannot be rectified.
    • The VM OS cannot be started.
    • If yes, go to 7.
    • If no, go to 8.

  7. Run the following commands to rebuild a VM.

    To rebuild a VM in cascading scenarios,

    Run the following command at the cascading layer:

    /opt/cloud/services/cascading-nova/venv/bin/python2.7 /etc/nova/nova-util/reschedule_vm.py <vm_id>

    Run the following command at the cascaded layer:

    /opt/cloud/services/nova/venv/bin/python2.7 /etc/nova/nova-util/reschedule_vm.py <vm_id>

    To rebuild a VM in non-cascading scenarios, run the following command:

    /opt/cloud/services/nova/venv/bin/python2.7 /etc/nova/nova-util/reschedule_vm.py <vm_id>

    vm_id specifies the ID of the VM to be rebuilt.

    If the information "####Reschedule Command has been accepted!####" is displayed in the command output, the rebuilding instruction is successfully issued. In this case, go to 12.

    If the information "Instance XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX sys volume is not found" is displayed in the command output, the system volume has been detached from the VM. In this case, follow the instructions provided in A Fault Occurs If the System Disk of a VM Is Detached.

    If other command output displayed, contact technical support for assistance.

  8. Run the following command to query the VM information:

    nova show uuid

    uuid: specifies the value obtained from the audit result processing. For details, see the methods of obtaining the value in the corresponding section.

    NOTE:
    • The host name is the value of OS-EXT-SRV-ATTR:host in the command output.
    • In the command output, the value of id in os-extended-volumes:volumes_attached is the ID of the volume attached to the VM.

  9. Run the following command to query the image ID of the VM system volume:

    cinder show uuid

    uuid: specifies the ID of the attached volume information obtained in 8.

    NOTE:
    • The drive letter of the attached disk is the value of device in attachments in the command output.
    • The volume image ID is the value of image_id in volume_image_metadata in the command output.
    • If multiple volume IDs are obtained in 8, run the command repeatedly to obtain the image IDs of the volumes whose drive letters are /dev/vda, /dev/xvda, or/dev/sda.

  10. Run the following command to check whether the image is in the ACTIVE status: In the command output, the value of status is the image status.

    nova image-show image_id

    image_id: specifies the image ID.

    • If yes, go to 11.
    • If no, contact technical support for assistance.

  11. Run the following command to rebuild the VM using the image ID:

    nova rebuild <vm_id> image_id
    • <vm_id>: indicates the VM ID obtained in 8.
    • image_id: specifies the image ID.

  12. Run the following command to check whether the VM is successfully rebuilt:

    nova show <vm_id>

    vm_id specifies the ID of the VM to be rebuilt.

    Check the status and task_state columns in the command output:

    • If the value of status is ACTIVE and the value of task_state is -, the VM is successfully rebuilt. Check whether you can log in to the VM.
    • The value of status is ERROR and the value of task_state is -. the VM rebuilding fails probably due to resource insufficiency. In this case, perform 7 again. If the VM rebuilding still fails, contact technical support for assistance.
    • If the value of task_state is not -, wait for 5 minutes, and then repeat this step to check whether the rebuilding is successful. If the value of task_state is still not -, perform the rebuilding again. If the rebuilding still fails, contact technical support for assistance.

NOTE:

If the rebuilt VM is the management VM of the controller node, perform steps 13 to 15.

  1. After the VM is successfully rebuilt, run the following command to obtain the ID of the host where the VM is rebuilt:

    nova show <vm_id>

    vm_id specifies the ID of the VM to be rebuilt.

    Take note of the OS-EXT-SRV-ATTR:hypervisor_hostname value in the command output. This value is the ID of the host accommodating the faulty VM.

  2. Check whether the host ID recorded in 5 is the same as that recorded in 13.

    • If yes, go to 15.
    • If no, migrate the rebuilt VM to the new host by following the instructions provided in Migrating a VM. After the migration succeeded, go to 15.

  3. Check whether the VM is started from an image by following the instructions provided in 5. If yes, check what services are deployed on the original VM based on the VM name and contact technical support for assistance.

Abnormal VM Status

Symptom

The tenant VM status is abnormal or the VM status is inconsistent with the power supply status.

Possible Causes
  • Invalid parameters are specified during the VM operation process.
  • Only some OSs support the VM operations.
  • Two or more operations are performed on the VM at the same time, for example, attaching a volume, detaching a volume, installing a network interface card (NIC), uninstalling a network interface card (NIC), and rebuilding, migrating, or modifying specifications.
  • An internal system exception or network exception occurs during the VM operation process.
Procedure

Obtain the system audit document from the FusionSphere version release path and rectify the abnormal audit items.

For details about the Region Type I scenario, see "System Audit (Region Type I)" in the FusionCloud 6.3.1.1 Alarm and Event Reference.

For details about the Region Type II and Region Type III scenarios, see "System Audit (Region Type II&Region Type III)" in the FusionCloud 6.3.1.1 Alarm and Event Reference.

Table 7-2 Abnormal VM statuses and troubleshooting methods

Abnormal VM status

Troubleshooting Method

VM CPU Usage Exceeds the Threshold

ALM-6036 VM CPU Usage Exceeds the Threshold in FusionCloud 6.3.1.1 Alarm and Event Reference

VM State Error

ALM-70102 VM State Error in FusionCloud 6.3.1.1 Alarm and Event Reference

VM HA Stuck in the Intermediate State

ALM-70106 VM HA Stuck in the Intermediate State in FusionCloud 6.3.1.1 Alarm and Event Reference

VM HA Triggered

ALM-70107 VM HA Triggered in FusionCloud 6.3.1.1 Alarm and Event Reference

Faulty VM Directory File

ALM-70108 Faulty VM Directory File in FusionCloud 6.3.1.1 Alarm and Event Reference

VM HA Malfunctions

ALM-70130 VM HA Malfunctions in FusionCloud 6.3.1.1 Alarm and Event Reference

VM Restarts Repeatedly

ALM-73107 VM Restarts Repeatedly in FusionCloud 6.3.1.1 Alarm and Event Reference

Replacing the SSL Certificate Used by FusionSphere OpenStack to Access Service OM Services

Scenarios

After the root Secure Sockets Layer (SSL) authentication certificate of Service OM is replaced, manually replace the SSL authentication certificate used by FusionSphere OpenStack to access Service OM services.

Prerequisites
  • A tool that can be used for remote access on various platforms, such as PuTTY, is available.
  • You have obtained a valid root certificate and named it ca.crt.
  • You have obtained the reverse proxy IP address of the first host as well as the passwords of the fsp and root users for logging in to the host.
Procedure
  1. Use WinSCP to upload the valid root certificate ca.crt to the first FusionSphere OpenStack host as the fsp user. The /home/fsp directory is used as an example in the following steps.

    Ensure that the FusionSphere OpenStack reverse proxy IP address is used to establish the connection. Set the protocol to SFTP and use the default port 22.

    The default password of the fsp user is Huawei@CLOUD8.

  2. Use PuTTY to log in to the first host in the AZ through the IP address of the External OM plane.

    The username is fsp and the default password is Huawei@CLOUD8.
    NOTE:
    • The system supports login authentication using a password or private-public key pair. If a private-public key pair is used for login authentication, seeUsing PuTTY to Log In to a Node in Key Pair Authentication Mode.
    • For details about the IP address of the External OM plane, see the LLD generated by FCD sheet of the xxx_export_all.xlsm file exported from FusionCloud Deploy during software installation, and search for the IP addresses corresponding to VMs and nodes.The parameter names in different scenarios are as follows:
      • Cascading layer in the Region Type I scenario : Cascading-ExternalOM-Reverse-Proxy, Cascaded layer : Cascaded-ExternalOM-Reverse-Proxy.
      • Region Type II and Type III scenarios : ExternalOM-Reverse-Proxy.

  3. Run the following command to switch to the root user, and enter the root password as prompted:

    su - root

    The default password of the root user is Huawei@CLOUD8!.

  4. Run the TMOUT=0 command to disable user logout upon system timeout.
  5. Import environment variables. For details, see Importing Environment Variables.

    Enter 1 to import Keystone V3 environment variables.

  6. Run the following command to switch to the directory containing the uploaded certificate:

    cd /home/fsp

  7. Run the following commands to import a certificate to the server:

    ZAPPER_PATH=$(cat /etc/init.cfg|grep repo |awk -F '=' '{print $2}')

    echo $ZAPPER_PATH|grep "/$" || ZAPPER_PATH="${ZAPPER_PATH}/"

    INTERNAL_CPS_PWD=`python -c 'from FSSecurity import crypt;import ConfigParser;sys_file_parser = ConfigParser.RawConfigParser();sys_file_parser.read("/etc/huawei/fusionsphere/cfg/sys.ini");print crypt.decrypt(dict(sys_file_parser.items("system_account"))["internal_cps_password"])'`

    curl -k -i -H "X-Auth-User:internal_cps_admin" -H "X-Auth-Password:$INTERNAL_CPS_PWD" -X PUT -T /home/fsp/ca.crt ${ZAPPER_PATH}3rdvms/setup/ca.crt

    In the preceding command, ca.crt specifies the name of the root certificate.

    The server certificate is successfully replaced if information similar to the following is displayed:

    HTTP/1.1 100 Continue 
     
    HTTP/1.1 201 Created 
    Last-Modified: Mon, 28 Sep 2015 14:06:55 GMT 
    Content-Length: 0 
    Etag: c57d2f13fd66905b62c8c0420a20a548 
    Content-Type: text/html; charset=UTF-8 
    X-Trans-Id: txe0257216b17347f3ba2f9-005609497e 
    Date: Mon, 28 Sep 2015 14:06:54 GMT 
    Connection: close

    If the command output does not contain "HTTP/1.1 201 Created", run the following commands to import the server certificate:

    ZAPPER_PATH="https$(echo ${ZAPPER_PATH}|awk -F "http" '{print $2}')"

    INTERNAL_CPS_PWD=`python -c 'from FSSecurity import crypt;import ConfigParser;sys_file_parser = ConfigParser.RawConfigParser();sys_file_parser.read("/etc/huawei/fusionsphere/cfg/sys.ini");print crypt.decrypt(dict(sys_file_parser.items("system_account"))["internal_cps_password"])'`

    curl -k -i -H "X-Auth-User:internal_cps_admin" -H "X-Auth-Password:$INTERNAL_CPS_PWD" -X PUT -T /home/fsp/ca.crt ${ZAPPER_PATH}3rdvms/setup/ca.crt

  8. Perform the following operations to query the hosts accommodating the CPS service, and record the host ID of the active node and the IP address of the External OM network plane:

    cps template-instance-list --service cps cps-server

    In the command output, runsonhost indicates the ID of the host accommodating the CPS service.

    The record with status being active is the active host accommodating the active CPS service. The IP address in the omip column is that of the External OM plane of the host.

  9. Run the following commands to log in to each active host accommodating the CPS service and switch to the root user:

    su - fsp

    ssh fsp@Host External OM IP address

    Enter the private key password as prompted. The default password is Huawei@CLOUD8!. If you have successfully replaced the public and private key files, enter the new private key password. Alternatively, press Enter and type the password of the fsp user to log in.

    The default password of the fsp user is Huawei@CLOUD8.

    su - root

    Run the following command to disable user logout upon system timeout:

    TMOUT=0

    The default password of the root user is Huawei@CLOUD8!.

  10. Run the following commands on the hosts accommodating the CPS service respectively to delete the old certificates in the two directories:

    rm /etc/huawei/fusionsphere/3rdvms/ca.crt > /dev/null 2>&1

    rm /etc/huawei/3rdvms/ca.crt> /dev/null 2>&1

    In the preceding command, ca.crt specifies the name of the old certificate.

  11. Run the following command to delete the uploaded certificate:

    rm /home/fsp/ca.crt

Quickly Collecting Fault Information Using FusionSphere OpenStack

Scenarios

FusionSphere provides the fault scenario information collection function, which does not depend on the external FusionCare tool, thereby accelerating log collection in accident scenarios.

Prerequisites
  • The browser must be Internet Explorer 9.0 to 11.0, Firefox 27.0 to 38.0, or Google Chrome 30.0 to 41.0.
  • You have logged in to the FusionSphere OpenStack web client using the reverse proxy IP address.
    NOTE:

    Before using the reverse proxy IP address to log in to the FusionSphere OpenStack web client, ensure that the reverse proxy has been correctly configured during installation. The login address format is https://Reverse proxy IP address:8890. If the reverse proxy is not configured or cannot be used for login, log in to https://172.28.9.42:8890 on the local PC used for installing the FusionSphere OpenStack.

Procedure
  1. On the FusionSphere OpenStack web client, choose O&M > Routine Maintenance.

    Figure 7-11 Quick collection of fault information

  2. In the Fault Information Quick Collection area, click Export.
  3. After about 5 minutes, check the exported file. If the system generates the FusionSphere_CollectData.tar.gz file, the fault information is collected and exported successfully.

Failure to Log In to a VM When Its System Volume Becomes Faulty

Scenarios

If the system volume of a VM becomes faulty, the VM may fail to be logged in to. You can perform the steps provided in this section to restore the VM.

Procedure

If the VM boots from an image, run the rescue command to restore the VM. If the VM boots from a volume, another VM to restore the VM. Run the nova show VM_ID command to query whether the VM boots from an image or a volume.

Query the VM boot mode.

  1. Use PuTTY to log in to the first host in the AZ through the IP address of the External OM plane.

    The username is fsp and the default password is Huawei@CLOUD8.
    NOTE:
    • The system supports login authentication using a password or private-public key pair. If a private-public key pair is used for login authentication, seeUsing PuTTY to Log In to a Node in Key Pair Authentication Mode.
    • For details about the IP address of the External OM plane, see the LLD generated by FCD sheet of the xxx_export_all.xlsm file exported from FusionCloud Deploy during software installation, and search for the IP addresses corresponding to VMs and nodes.The parameter names in different scenarios are as follows:
      • Cascading layer in the Region Type I scenario : Cascading-ExternalOM-Reverse-Proxy, Cascaded layer : Cascaded-ExternalOM-Reverse-Proxy.
      • Region Type II and Type III scenarios : ExternalOM-Reverse-Proxy.

  2. Run the following command to switch to the root user:

    su - root

    The default password of the root user is Huawei@CLOUD8!.

  3. Run the TMOUT=0 command to disable user logout upon system timeout.
  4. Import environment variables. For details, see Importing Environment Variables.
  5. Determine the VM boot mode.

    1. Run the following command to enter the secure operation mode:

      runsafe

      The following information is displayed:

      Input command:
    2. Run the following command to query the ID of the rescued VM:

      nova list --all-t

      The following information is displayed:

    3. Run the following command to query the VM attributes:

      nova show VM_ID

      VM_ID indicates the ID of the rescued VM.

      • The VM boots from an image if information similar to the following is displayed (d0bd0551-07f2-45f6-8516-f481e0152715 specifies the image ID): Perform 6 to 10 to restore the VM.
        | image | cirros (d0bd0551-07f2-45f6-8516-f481e0152715)|
        If the following information is displayed, the image boots from a volume. Perform 11 to 19 to restore the VM.
        | image | Attempt to boot from volume - no image supplied|

Restore the VM booting from an image.

  1. Upload the rescue image and register it. For details, see "Registering an Image" in FusionCloud 6.3.1.1 O&M Guide.

    NOTE:
    • The image file must be able to guide the VM to properly boot and must meet the restoration requirement.
    • The image file should use a name that is easy to identify.

  2. Set the VM to the rescue mode.

    You can either specify an image or not when you set the VM to the rescue mode.
    • If you choose not to specify an image:
      1. Run the nova rescue VM_ID command to set the VM to enter the rescue mode.

        VM_ID indicates the ID of the rescued VM.

      2. Run the nova list --all-t command to check whether the status of the rescued VM is changed to RESCUE.

    • If you choose to specify an image:
      1. Run the glance image-list command to query the available images and image IDs.

        The image whose Name is cirros is the rescue image registered in 6.

      2. Run the nova rescue --image IMAGE_ID VM_ID command to set the VM to enter the rescue mode.

        VM_ID indicates the ID of the rescued VM. IMAGE_ID indicates the image ID obtained in 7.a. You can select a Linux image or a Windows image. The following uses a Linux image as an example.

      3. Run the nova list --all-t command to check whether the status of the rescued VM is changed to RESCUE.

  3. Run the following command to obtain the VM URL and log in to the VM using this URL:

    nova get-vnc-console VM_ID novnc

    VM_ID indicates the ID of the rescued VM.

    The following information is displayed:

    Replace nova-novncproxy.az1.dc1.domainname.com in the URL in the command output with the reverse proxy IP address of the FusionSphere OpenStack system, such as https://192.168.11.11:8002/vnc_auto.html?token=be540cf4-0185-4992-bafc-60bf6db48191&lang=EN.

    Use the new URL to log in to the rescued VM.

  4. Enter the username and password for logging in to the rescued VM and run the command to restore the faulty system volume.

    • The system disk of the VM is vda before the VM enters the rescue mode, vda/sda for Linux VMs and C for Windows VMs. The mount point is determined based on site requirements.

    • After the VM enters the rescue mode, the disk drive letter of the VM's system volume is changed to the immediate next letter.

    NOTE:

    The rescue interface is designed for restoring the system volume. Only the system volume is displayed after the VM enters the rescue mode. If the data volume of the rescued VM also becomes faulty, first restore the system volume. Then, restore the data volume after the VM starts properly.

  5. Run the nova unrescue VM_ID command to restore the VM status to ACTIVE.

    VM_ID indicates the ID of the rescued VM.

    If you can log in to the VM and start and stop the VM properly, the VM is restored.

Restore the VM booting from a volume.

  1. Log in to the FusionSphere OpenStack web client in tenant view. Choose Resources > Computing > Compute Instances > VMs and locate the VM to be rescued.
  2. On the VMs page, click Create to create a rescue VM.

    Parameter settings of the rescue VM, including the AZ, flavor, and image, must be the same as those of the rescued VM.

  3. Locate the row that contains the rescued VM and choose More > Stop.
  4. Mount the system volume of the rescued VM to the rescue VM.

    1. Perform 1 through 4 to log in to the first host in the AZ and import environment variables.
    2. Run the nova list --all-t command to check whether the status of the rescued VM is changed to SHUTOFF. Record the IDs of the rescued VM and rescue VM.

    3. Run the nova show VM_ID command to query the ID and device name of the system volume for the rescued VM.

      VM_ID indicates the ID of the rescued VM.

    4. Run the nova volume-detach VM_ID VOLUME_ID command to dismount the system volume from the rescued VM.

      VM_ID indicates the ID of the rescued VM. VOLUME_ID indicates the system volume ID of the rescued VM.

    5. Run the nova volume-attachments VM_ID command to check whether the system volume is dismounted from the rescued VM. The command output shows the system volume has been dismounted from the target volume.

      VM_ID indicates the ID of the rescued VM.

    6. Run the nova volume-attach VM_ID VOLUME_ID command to mount the system disk to the rescue VM.

      VM_ID indicates the ID of the rescue VM. VOLUME_ID indicates the system volume ID of the rescued VM.

    7. Run the nova volume-attachments VM_ID command to check whether the system volume is mounted to the rescue VM. The command output shows that the system volume has been mounted to the rescue VM.

      VM_ID indicates the ID of the rescue VM.

  5. Run the following command to obtain the VM URL and log in to the rescue VM using the URL:

    nova get-vnc-console VM_ID novnc

    VM_ID indicates the ID of the rescue VM.

    The following information is displayed:

    Replace nova-novncproxy.az1.dc1.domainname.com in the URL in the command output with the reverse proxy IP address of the FusionSphere OpenStack system, such as https://192.168.11.11:8002/vnc_auto.html?token=7c640a6b-e23a-4885-a651-e6df097f12eb&lang=EN.

    Use the new URL to log in to the rescued VM.

  6. Perform operations to repair the corrupted file system based on site requirements.
  7. After the fault is rectified, log in to the FusionSphere OpenStack web client and switch to the VMs page. Locate the row of the rescue VM and choose More > Stop.
  8. Log in to the FusionSphere OpenStack host, detach the restored system volume from the rescue VM and attach it to the rescued VM.

    1. Run the nova volume-detach VM_ID VOLUME_ID command to detach the system volume from the rescue VM.

      VM_ID indicates the ID of the rescue VM. VOLUME_ID indicates the system volume ID of the rescued VM.

    2. Run the nova volume-attachments VM_ID command to check whether the system volume is detached from the rescue VM.

      VM_ID indicates the ID of the rescue VM.

    3. Run the nova volume-attach VM_ID VOLUME_ID VOLUME_NAME command to attach the restored system volume to the rescued VM.

      VM_ID indicates the ID of the rescued VM. VOLUME_ID indicates the system volume ID of the rescued VM. VOLUME_NAME indicates the name of the system volume.

    4. Run the nova volume-attachments VM_ID command to check whether the volume is attached to the rescued VM.

      VM_ID indicates the ID of the rescued VM.

  9. Log in to the FusionSphere OpenStack web client and switch to the VMs page. Locate the row of the rescued VM and click Start.

    Check the status of the rescued VM. If its status is changed to Running from Stopped, the VM is restored.

Migrating a VM

If a host is faulty, or if it is overloaded or underloaded, migrate the VMs running on it to ensure service continuity or proper resource utilization. Available VM migration modes include live migration and cold migration.

Background
  • VM live migration is not supported when the VM NIC works in SR-IOV, PF, or PCI passthrough mode.
  • VM live migration is not supported for VMs using passthrough-enabled local disks.
  • VM live migration can be performed in the following scenarios:
    • If a host is faulty or overloaded, migrate the VMs running on it to another host to prevent service interruption.
    • If most hosts are underloaded, migrate the VMs to a smaller number of hosts and then power off the idle hosts. This improves resource utilization and reduces power consumption.
  • VM cold migration can be performed in the following scenarios:
    • VMs need to be migrated to prepare for planned upgrade or parts replacement.
    • VMs can be cold migrated if they are in the running state and cannot be live migrated.
NOTE:
  • Migrate VMs one at a time. Start migrating another VM only after the previous one is successfully done.
  • VM live migration may take a long time if the migrated VMs are heavily loaded or have high memory usage. Therefore, perform VM live migration during off-peak hours.
  • The web client displays a timeout prompt when live migration exceeds 12 hours. However, the internal live migration timeout is 24 hours. The system gives up live migration and releases the occupied resources only after the internal timeout expires.
  • When live migrating a VM that boots from a local disk, OpenStack only checks whether the real available disk space on the target host is sufficient, regardless of the remaining disk space displayed on the hypervisor.
  • Live migration requires that the source and destination hosts use the same type of hypervisor (KVM, VMware, or VRM).
  • VM cold migration is not supported if the virtualization type of the AZs housing the VMs is VMware.
  • Batch cold migration is supported if the virtualization type of the AZs housing the VMs is FusionCompute.

For details about the restrictions on VM migration, see Table 7-3.

Table 7-3 Restrictions

Resource-related VM

Cold Migration (Yes/No)

Live Migration (Yes/No)

Local file VM

Yes

No

VM started from the common volume

Yes

Yes

DPDK VM

Yes

Yes

SRIOV/PF passthrough VM

Yes

No

VM with shared volumes

Yes (Not supported for migration of a whole VM)

Yes (Not supported for migration of a whole VM)

VM with SCSI passthrough volumes

Yes

No

NUMA affinity VM

Yes

Yes

Core-binding VM

Yes

Yes

Hugepage memory VM

Yes

Yes

Prerequisites

You have logged in to the ManageOne OM plane.

Procedure
  1. Click O&M Maps on the top tab menu of the ManageOne OM plane.
  2. On the O&M Maps page, choose Configuration > Cloud Services > Compute Services > Compute Instances.
  3. Perform the operations listed in Table 7-4.

    Table 7-4 Migrating a VM

    Operation

    Procedure

    Live migration

    1. Locate the row that contains the target VM, click More, and then click Live Migrate.
    2. Select the destination host and click OK.

    Cold migration

    1. Locate the row that contains the target VM, click More in the Operation column, and select Cold Migrate. If the following alarm is displayed, click OK.

    2. On the Cold Migrate page, select Specify Host or Not Specify Host.
      NOTE:
      • Specify Host: specifies the destination host. You need to specify a destination host in the host list in the lower part of the page.
      • Not Specify Host: The system automatically migrates the VM to a host.

    3. Click OK.

Database Operations

Obtaining Logs as Required
Database Server Logs

Logging In to a Host Running the Database Service and switch to the /var/log/fusionsphere/component/gaussdb directory to obtain the database server logs.

Database Control Script Logs

Logging In to a Host Running the Database Service and switch to the /var/log/fusionsphere/component/gaussdbControl directory to obtain the database control script logs.

System Logs

Logging In to a Host Running the Database Service and switch to the /var/log/uvp-getosstat directory to obtain system status logs recorded by the UVP, and switch to the /var/log/messages directory to obtain system operation logs.

Active and Standby Arbitration Component Logs

Logging In to a Host Running the Database Service and switch to the /var/log/fusionsphere/component/cps-server directory to obtain cps-server logs, and switch to the /var/log/fusionsphere/component/cps-client directory to obtain cps-client logs.

Logging In to a Host Running the Database Service
  1. Use PuTTY to log in to the first host of FusionSphere OpenStack through the IP address of the External OM plane.

    The username is fsp and the default password is Huawei@CLOUD8.
    NOTE:
    • The system supports login authentication using a password or private-public key pair. If a private-public key pair is used for login authentication, seeUsing PuTTY to Log In to a Node in Key Pair Authentication Mode.
    • For details about the IP address of the External OM plane, see the LLD generated by FCD sheet of the xxx_export_all.xlsm file exported from FusionCloud Deploy during software installation, and search for the IP addresses corresponding to VMs and nodes.The parameter names in different scenarios are as follows:
      • Cascading layer in the Region Type I scenario : Cascading-ExternalOM-Reverse-Proxy, Cascaded layer : Cascaded-ExternalOM-Reverse-Proxy.
      • Region Type II and Type III scenarios : ExternalOM-Reverse-Proxy.

  2. Run the following command to switch to user root and enter the password of user root as prompted:

    su - root

    The default password of the root user is Huawei@CLOUD8!.

  3. Run the TMOUT=0 command to disable user logout upon system timeout.
  4. Import environment variables. For details, see Importing Environment Variables.
  5. Run the cps template-instance-list --service gaussdb gaussdb command to query the board where the database is located. The rows containing the active and standby states record the IDs of the active and standby database hosts, respectively. The IP address in the omip column is that of the External OM plane of the host. The command output is as follows:

    +------------+---------------+---------+--------------------------------------+------------+ 
    | instanceid | componenttype | status  | runsonhost                           | omip       | 
    +------------+---------------+---------+--------------------------------------+------------+ 
    | 0          | gaussdb       | active  | 40E3DF67-12BF-E411-8135-000000821800 | 128.26.3.7 | 
    | 1          | gaussdb       | standby | 201A5510-FBDD-E411-A9B3-000000821800 | 128.26.3.4 | 
    +------------+---------------+---------+--------------------------------------+------------+

  6. Run the following commands to log in to the host where the database is located:

    su fsp

    ssh fsp@Host External OM IP address

    Enter the private key password as prompted. The default password is Huawei@CLOUD8!. If you have successfully replaced the public and private key files, enter the new private key password. Alternatively, press Enter and enter the password of user fsp for the login.

    Example:

    ssh fsp@172.28.6.3

    After you log in to the host as user fsp, run the su - root command to switch to user root.

    The default password of user fsp is Huawei@CLOUD8.

    The default password of user root is Huawei@CLOUD8!.

    Import environment variables. For details, see Importing Environment Variables.

Translation
Download
Updated: 2019-06-10

Document ID: EDOC1100063248

Views: 23108

Downloads: 37

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