Upgrading the System Software

Use the instructions in this topic to upgrade the Automated Uptime Layer with Smart Exchange system software on a Stratus ztC Endurance system from Release 1.0.x.x to Release 1.1.x.x. Upgrading the system includes upgrading the following:

  • Management VM software (Ubuntu)

  • VMware ESXi hypervisor software and Stratus proprietary add-ins

  • Compute module BMC firmware, BIOS firmware, and Standby OS software (Ubuntu)

Notes:  
  • For the latest information about upgrade paths and restrictions, see the Release Notes: Automated Uptime Layer with Smart Exchange Release 1.1.0.0 (VMware).
  • Consider scheduling a maintenance window during which you can complete the upgrade process, which takes approximately 3 hours. Because the system needs to upgrade software and firmware on both the A- and B-sides of the system, it must shut down the virtual machines and the ESXi host halfway through the upgrade to perform a reboot exchangeClosed A process for switching from the active compute module to the standby compute module when a Smart Exchange is not possible due to firmware mismatches between the two modules. During a reboot exchange, the system: 1) Instructs the standby compute module to become active. 2) Shuts down the original active compute module. 3) Reboots the new active compute module, which automatically takes ownership of all I/O and storage connected to the original active compute module. 4) Starts the system software on the new active compute module. A reboot exchange is typically initiated during software upgrades, firmware updates, and troubleshooting procedures. to the upgraded side and continue upgrading software and firmware on the other side. (The ESXi host restarts only once during the upgrade process. The Standby OS in each compute module restarts additional times, but it does not affect the availability of the virtual machines.)

    You can manage the downtime for your virtual machines in one of two ways:

    • To minimize downtime for your virtual machines, you can allow them to keep running during the upgrade. About 1.5 hours through the upgrade, the system automatically shuts down both the virtual machines and the ESXi host, performs a reboot exchange to reprovision the storage and devices​, and starts the ESXi host and virtual machines on the side that was already upgraded. If the virtual machines are properly configured to autostart, downtime for this reboot exchange process may take approximately 20-25 minutes, depending on how many virtual machines are present and how long it usually takes to shut down and restart them.

    • Alternatively, you can shut down the virtual machines before the upgrade starts. If the virtual machines are properly configured to autostart, they start immediately after the reboot exchange for a total downtime of approximately 1.5 hours.

  • To minimize downtime, it is important to confirm that your virtual machines are properly configured to autostart when the ESXi host starts. If you do not want the virtual machines to start until after you have verified the success of the upgrade, you can optionally disable autostart, but remember to re-enable it after the upgrade. For information about managing autostart settings, see the VMware vSphere documentation.

    Do not change the autostart setting on the Management VM, which must run at all times to ensure the proper fault-tolerant operation of your system and manage upgrade operations.

  • Before performing an upgrade on a system in use, ensure that you have backed up all of your virtual machines.

  • To ensure the success of the upgrade process, remove snapshots for all virtual machines in the ESXi Host Client or vSphere Web Client before starting the upgrade. The upgrade will not proceed if any snapshots are present.

  • Verify that the system is in good health before upgrading the system software. All modules must be running and not in maintenance mode. In the ztC Endurance console​, resolve any alerts on the Dashboard page before continuing with the upgrade. The masthead must display a green check mark () and System is Healthy with no outstanding issues; otherwise, the upgrade will not proceed.

  • During the upgrade process, you temporarily lose access to the ztC Endurance console​ for 1-2 minutes when the Management VM restarts on the newly upgraded software, and later for about 15-20 minutes when the system performs the reboot exchange to the newly upgraded side of the system. When the ztC Endurance console​ returns, refresh your web browser and log in again.

  • The upgrade to Automated Uptime Layer with Smart Exchange Release 1.1.x.x adds a new Upgrade Kits page to the ztC Endurance console​ for managing upgrade kits and monitoring upgrades in progress. The Upgrade Kits page becomes available after the system upgrades and restarts the Management VM, about 10-15 minutes after you start the upgrade. Thereafter, you can use the Upgrade Kits page to monitor the completion of the upgrade as described in the upgrade procedure.

  • If a software upgrade is in progress, do not upload another upgrade kit on the Upgrade Kits page of the ztC Endurance console​. Doing so will disrupt the upgrade.

  • After upgrading, you need to register your ztC Endurance system, download a license file, and apply the license file as described in Registering the System. (If you need to use ASNConnect to allow service personnel to remotely diagnose your system as described in Enabling ASNConnect, the system must have an active license installed before you enable the ASNConnect software.)

  • In Automated Uptime Layer with Smart Exchange Release 1.0.0.0, the system enabled Ubuntu OS unattended upgrades in the Management VM by default. If your system was connected to the Internet at any time after you installed Release 1.0.0.0, and it has downloaded updates later than the original Release 1.0.0.0 Ubuntu software packages, the system requires access to the Internet during the upgrade process to Release 1.1.0.0 or higher to resolve package dependencies. (If your system has never been connected to the Internet, you do not need to connect it for the upgrade.)

  • When you upgrade to Release 1.1.0.0 or higher, the system permanently disables Ubuntu unattended upgrades in the Management VM. To prevent incompatibilities and ensure the proper fault-tolerant operation of your system, do not enable Ubuntu unattended upgrades in the Management VM or Standby OS.

  • When you upgrade to Release 1.1.0.0 or higher, the system completely reinstalls and overwrites the Standby OS in each compute module. If needed, save any custom files from the Standby OS before the upgrade, and always avoid modifying the Standby OS and Management VM unless instructed to do so.

Caution: Do not update the Ubuntu system software of the Management VM or Standby OS from any source other than Stratus unless instructed to do so. To prevent incompatibilities and ensure the proper fault-tolerant operation of your system, use only the release and software packages that are installed with the Automated Uptime Layer with Smart Exchange software.
Caution: If you upgrade any software or firmware outside of the documented process for upgrades, the ztC Endurance system automatically reinstalls and reverts to the versions of software or firmware that are compatible with Automated Uptime Layer with Smart Exchange. If the system needs to revert to an earlier BIOS revision, it automatically triggers a reboot exchange with downtime of approximately 20-25 minutes for the virtual machines as previously described.

The upgrade steps are:

I. To download the upgrade files

II. To upload the upgrade files to the system

III. To verify the ESXi network configuration

IV. To upgrade the system software

V. To complete post-upgrade tasks