Release Notes

NDU
Non Disruptive Upgrade (NDU) of the software can be performed on the metro node using inbuilt design. The basic principle is
system is divided into two halves that are called as first upgraders (A directors) and second upgraders (B directors).
All the nodes of the first upgraders are upgraded to new software, and when it is stable and hosts recognize it, then the set of
nodes in the second upgraders are upgraded.
NOTE: NDU start must be triggered from the second upgraders (B directors) only as it initiates a reboot for the first
upgraders (A directors).
The NDU consist of the following steps:
● service@director-1-1-B:~>: To upgrade the software on all the nodes, install the new version of NDU module and
tools.
● VPlexcli:/> ndu pre-check: It performs various pre-checks. It verifies the health of the system, and also verifies that
whether configurations are followed to avoid DU situation during NDU.
● From second upgrader: VPlexcli:/> ndu start -u /home/service/vplex-node-upgrade-VPlex-
<version>.tar: ndu start triggers on second upgrader and upgrades the nodes of first upgraders.
● From first upgrader: VPlexcli:/> ndu complete: ndu complete upgrades the second upgraders and completes the
upgrade procedure.
○ Once the NDU is complete, run the following commands:
￭ systemctl start notifications
￭ systemctl start kafka
For more details, see the NDU related documents available on SolVe online at https://solveonline.emc.com/solve/home.
UI Notifications Page
The Notification page is the consolidated area of all the appliance alerts and events for the cluster. Events provide information
about changes to the system, which indicate there is a problem with the system. Alerts are categorized by severity which
indicates the urgency of the alert.
On the notifications tab, there are two alert tabs :
Platform Alerts: You can monitor the status of metro node platform which includes alerts at director level and cluster level. In
the Live alerts drop-down, you can view alerts that were created during the last 24 hours.
Hardware Alerts: You can monitor the status of metro node hardware which includes alerts that are generated at hardware
level. In the Live alerts drop-down, you can view alerts that were created during the last 24 hours.
In the details of each alert, additional information can be found including the Severity, Message, Corrective Action, and other
properties. This information is useful in troubleshooting scenarios and allows users to remediate issues seen on the system.
When alerts are no longer relevant or are resolved, the system automatically clears the alerts with no user intervention. This
action ensures that cleared alerts are hidden from the default view so that only relevant issues are displayed. Acknowledging an
alert does not indicate that the issue is resolved. Acknowledged alerts can be viewed as well through table filtering options.
Configure Alerts
Under settings, go to Notifications. The user can also configure alerts at various levels.
● Enable or disable alert at alert condition ID level
○ For example: 0x30004 is the condition ID for Storage Volume Unreachable alert and if the customer wants to disable
this alert, select that particular alert and disable 0x30004. With it, whenever metro node issues 0x30004, it will not be
listed in UI, and also a dial home event will not be generated.
● Configure alerts at component level
○ When the user clicks Configure notifications button, then side panel shows the options to disable the alerts at
components level. For example: If System Volume is disabled, then whenever metro node presses any system volume-
related alerts, it will not be listed in UI, and also a dial home event will not be generated.
6
Release Notes