Skip to main content
  • Place orders quickly and easily
  • View orders and track your shipping status
  • Enjoy members-only rewards and discounts
  • Create and access a list of your products
  • Manage your Dell EMC sites, products, and product-level contacts using Company Administration.

Upgrade Dell PowerFlex to v3.6.x

PDF

Manual upgrade of Linux-based nodes—3-node MDM cluster

Manual upgrade of PowerFlex components, on a Linux-based nodes (all flavors) when the MDM cluster is in 3-node mode.

Prerequisites

  • Ensure that the nodes that will be upgraded meet the requirements described in the Getting to Know Dell PowerFlex Guide. In addition, LIA on each node requires an additional 350 MB RAM. Note that SDS components could require up to 19 GB if the maximum capacity of 128 TB is used.
  • If needed, install the GPG-RPM-KEY file (with the rpm --import <path_to_new_public_key> command). This is used to authenticate the RPM packages.
  • For data networks using IPv6, if you plan to implement a floating virtual IP address for the MDM, disable the IPv6 DAD setting on the node's relevant interfaces, using the command:
     sysctl net.ipv6.conf.<interface_name>.dad_transmits=0
  • If you intend to use the CHAP component authentication feature after the upgrade, make the necessary preparations to ensure that you upgrade all system components now. Systems containing any components running PowerFlex versions lower than v3.5 do not support CHAP authentication.
    NOTE:Upgrade to v3.5 sets the CHAP authentication feature to DISABLED, and passwords will be all-zero. To enable the feature, after the upgrade is complete on all system components, follow the post-installation procedures described in the Configure and Customize Dell PowerFlex.

About this task

When SNMP is enabled in the 2.x base version, SNMP MDM credentials remain in the lockbox after the upgrade.

When performing this procedure on XenServer nodes, use siocli in place of scli.

For the purposes of this procedure, the following naming conventions are used:

MDM 1—primary MDM

MDM 2—secondary MDM

TB 1—tiebreaker 1

NOTE:During this procedure, the finalize_upgrade command initiates an MDM service restart. This should have no other effect on the system.

Steps

  1. Download and extract the latest PowerFlex packages to /tmp/.
  2. Upgrade (or install) LIA on all machines in the system:
    rpm -U EMC-ScaleIO-lia-3.6-X.<build>.elX.x86_64.rpm --nosignature
  3. Switch to MDM single mode, by running the following commands on MDM 1:
    scli --login --username <USER> --password <PASSWORD>
    scli --start_upgrade
    scli --allow_commands_during_upgrade
    scli --switch_cluster_mode --cluster_mode 1_node --remove_slave_mdm_ip <MDM2 IP> --remove_tb_ip <TB1 IP>

    In the --switch_cluster_mode command, type the data (not management) IP addresses.

  4. Upgrade MDM 2 and TB 1, by running the following commands on both:
    1. Stop service:
      /opt/emc/scaleio/mdm/bin/delete_service.sh
    2. Upgrade RPM:
      I_AM_SURE=1 UPGRADE_PRECHECK_SCRIPT_DONE=1 rpm -U EMC-ScaleIO-mdm-3.6-X.<build>.<flavor>.x86_64.rpm
    3. Start service:
      /opt/emc/scaleio/mdm/bin/create_service.sh
  5. Add the upgraded MDM 2 and TB 1, by running the following commands on MDM 1:
    1. Add the components to the cluster:
      scli --switch_cluster_mode --cluster_mode 3_node --add_slave_mdm_ip <MDM2 IP> --add_tb_ip <TB1 IP>
    2. Verify that the cluster is in normal state:
      scli --query_cluster
  6. Switch MDM ownership, by running the following command on MDM 1:
    scli --switch_mdm_ownership --new_master_mdm_ip <MDM2 IP>
  7. Run the following commands on MDM 2:
    1. Log in:
      scli --login --username <USER> --password <PASSWORD>
    2. Enable commands:
      scli --allow_commands_during_upgrade
    3. View cluster information:
      scli --query_cluster
    4. Switch to single mode:
      scli --switch_cluster_mode --cluster_mode 1_node --remove_slave_mdm_ip <MDM1 IP> --remove_tb_ip <TB1 IP>
  8. Upgrade MDM 1, by running the following commands on MDM 1:
    1. Stop service:
      /opt/emc/scaleio/mdm/bin/delete_service.sh
    2. Upgrade RPM:
      I_AM_SURE=1 UPGRADE_PRECHECK_SCRIPT_DONE=1 rpm -U EMC-ScaleIO-mdm-3.6-X.<build>.<flavor>.x86_64.rpm
    3. Start service:
      /opt/emc/scaleio/mdm/bin/create_service.sh
  9. Add the MDM as a standby MDM, and switch to 3-node cluster mode, by running these commands from the primary MDM (MDM 2):
    1. Log in (if not logged in):
      scli --login --username <USER> --password <PASSWORD>
    2. Enable commands (if not already done):
      scli --allow_commands_during_upgrade
    3. Switch mode:
      scli --switch_cluster_mode --cluster_mode 3_node --add_slave_mdm_ip <MDM1 IP> --add_tb_ip <TB1 IP>
    4. Wait for the cluster to finish synchronizing the repository:
      scli --query_cluster
      Wait until cluster state is Normal.
    5. Switch cluster ownership:
      scli --switch_mdm_ownership --new_master_mdm_ip <MDM1 IP>
  10. Display list of all SDSs, by running the following commands from primary MDM 1:
    scli --login --username <USER> --password <PASSWORD>
    scli --query_all_sds
  11. Upgrade SDSs. Repeat this step on each SDS:
    1. Enter an SDS into maintenance mode, by running the following command from MDM 1:
      scli --enter_maintenance_mode --sds_id <SDS ID>
      Validate that the operation succeeded.
    2. Disable RFcache (if installed), by running the following command from MDM 1:
      scli --disable_rfcache --sds_id <SDS ID>
    3. Upgrade the SDS, by running the following command on the machine where the SDS resides:
      rpm -U EMC-ScaleIO-sds-3.6-X.<build>.<flavor>.x86_64.rpm
    4. Upgrade RFcache (if installed), by running the following command on the machine where the SDS resides:
      rpm -U EMC-ScaleIO-xcache-3.6-X.<build>.<flavor>.x86_64.rpm
    5. Enable RFcache (if installed), by running the following command from MDM 1:
      scli --enable_rfcache --sds_id <SDS ID>
    6. Exit the SDS from maintenance mode, by running the following command from MDM 1:
      scli --exit_maintenance_mode --sds_id <SDS ID>
    7. Before continuing, validate that the SDS is in the proper mode, by running the following command from MDM 1:
      scli --query_properties --object_type SDS --object_id <SDS ID> --properties MAINTENANCE_MODE_STATE
      Ensure that the state is NO_MAINTENANCE
    The SDS is upgraded. Repeat these steps for every SDS.
  12. Upgrade all SDCs:
    rpm -U EMC-ScaleIO-sdc-3.6-X.<build>.<flavor>.x86_64.rpm
    If the SDC kernel module is processing I/O, its upgrade will occur only after the node is restarted.
    NOTE:An SDC may appear to be running I/O even though no I/O is actually occurring. In such cases, perform a reboot and return to operation of each SDC one by one, preferably after all other system components have been upgraded.
    The SDC is upgraded. Repeat this step for every SDC.
  13. Validate the upgrade's current status, by running the following command on MDM 1:
    scli --query_upgrade
    Ensure that the output includes the following:
    0 SDSs need to be upgraded in total
  14. Define the expected device media type for each Storage Pool, by running the following command for each one:
    scli --set_storage_pool_media_type --protection_domain_name <PROTECTION_DOMAIN_NAME> --storage_pool_name <STORAGE_POOL_NAME> --media_type <MEDIA_TYPE>
    Device media type for a Storage Pool has been set. Repeat this step for every Storage Pool.
  15. Finalize the upgrade, by running the following command on MDM 1:
    scli --finalize_upgrade
    Ensure that the output includes the following:
    Successfully finalized the upgrade process
  16. Upgrade the currently installed PowerFlex Gateway by running the following command (optional):
    rpm -U EMC-ScaleIO-gateway-3.6-X.<build>.x86_64.rpm

    The following variables can be added to the command, if required:

    • SIO_GW_KEYTOOL=<keytool_path> allows override of the location for the Java keytool utility. For example: SIO_GW_KEYTOOL=/usr/java/jre1.8.0_XX/bin/
    • SIO_JAVA=<Java version> specifies a different Java version
    Example:
    rpm -U /tmp/EMC-ScaleIO-gateway-3.6-X.<build>.x86_64.rpm
    A success message is displayed, and the PowerFlex Installer interface now reflects the updated PowerFlex Gateway version.
    NOTE:If your PowerFlex Gateway uses a non-default certificate, you must copy the certificate and other files before upgrading the PowerFlex Gateway. For more information, see the "Upgrading the PowerFlex Gateway when a non-default certificate is used" topic.

Next steps

If you are upgrading from v2.6.x, use the CLI or the upgraded GUI to add RFcache devices back to the system. First, create an RFcache SSD Acceleration Pool for the RFcache devices, and then add the devices to the Acceleration Pool. For more information, see the Configure and Customize PowerFlex guide.
NOTE:From v3.0 and later, the commands and workflow for adding RFcache are different from earlier versions.

Rate this content

Accurate
Useful
Easy to understand
Was this article helpful?
0/3000 characters
  Please provide ratings (1-5 stars).
  Please provide ratings (1-5 stars).
  Please provide ratings (1-5 stars).
  Please select whether the article was helpful or not.
  Comments cannot contain these special characters: <>()\