Equinix continually strives to improve Network Edge to provide a cutting-edge platform that customers use to enable their multi/hybrid cloud architectures. Equinix will perform a major Network Edge infrastructure upgrade to significantly enhance Network Edge capabilities relating to features, services, and device types. This upgrade will be performed for all 27 Network Edge service locations. The upgrade will be service impacting. This topic summarizes details of the upgrade, its impact, and instructions to recover the VNF in the rare scenario where auto recovery of the device fails
Overall Infrastructure Upgrade Flow
Network Edge Infrastructure upgrade for 27 metro locations will be upgraded during Q4 2022 and Q2 2023. This table outlines the overall flow for the upgrade of each location.
|Customer notification prior to upgrade
|Equinix representatives will be reaching out to users to discuss upgrade details, required actions based on VNF type and possible mitigation plan. Email notifications containing upgrade details and instructions in case of failures are sent to users two weeks prior to the upgrade and a reminder notification will be sent a day before the upgrade. Users will receive two types of notification. One is for pre-upgrade notification. Another one is for RMA process described in section below. It is highly recommended that users subscribe to the status.equinix.com to receive upgrade schedule changes and upgrade status.
|It’s strongly recommended that device configuration be backed up before the upgrade to a remote location from where it can be accessed after upgrade, through a VNF interface if required. Equinix will be backing up configurations for all Equinix-Configured devices.
|Equinix performs the infrastructure upgrade. The complete upgrade is expected to take approximately 45-60 hours. See Service Interruption Details.
Users won’t be able to create any new devices during maintenance window. Status.equinix.com provides most up-to-date information and status of the infra upgrade.
|Upgrade completion notification
|Email notification is sent after the upgrade containing a checklist and including instructions if an unexpected incident occurs requiring recovery flow.
|If needed: VNF notification
|User + Equinix
|Some devices might require specific action to be taken post upgrade. See Troubleshooting VNF Failure.
|If needed: open support case
|Open a support case if unexpected incident happens after the upgrade.
See Network Edge Infrastructure Update Schedule for the tentative sequence in which we intend to upgrade the NE infrastructure in the 27 Network Edge locations. The Metros are listed in order of target dates. The update sequence is subject to change.
This is a major upgrade of the infrastructure, requiring management stack and firmware upgrades to all servers. All efforts were made to minimize service interruptions for redundant and clustered deployments. However, due to the breadth of the upgrade, redundant and cluster devices in the same metro location might experience service interruptions as listed in the table below. Some customers may experience traffic degradation of approximately 3 hours.
Currently a total of 48-60 hours maintenance window is expected for this infrastructure upgrade. During the maintenance window, the user will not be able to create, modify, or delete any VNFs and associated Network Edge Services. If any new VNFs are ordered, the provisioning of the virtual device is queued, and the provisioning process will start after the upgrade is completed.
For Ashburn (DC), the maintenance window time is expected to be longer than other metros due to the size of the infrastructure, and it is expected to be approximately 60 hours before the Maintenance window is closed.
|Single device deployments will experience a total service interruption of up to 80 minutes. Interruptions will be comprised of multiple smaller interruptions over the total upgrade period.
|Redundant device –Single Metro
|Single metro, redundant devices may experience a total service interruption of up to 20 minutes. Interruptions will be comprised of multiple interruptions over the total upgrade period due to simultaneous software process restarts.
|Redundant device – Multiple Metros
|If a primary device is installed in the upgrading metro, and a secondary device is designed to re-route traffic while the primary device is down, there will be no downtime expected during this upgrade, other than the normal router re-convergence time.
|Clustered devices may experience total service interruption of up to 20 minutes. Interruptions will be comprised of multiple smaller interruptions over the total upgrade period due to simultaneous software process restarts.
The new infrastructure management stack is incompatible with software below the versions listed in the table for Palo Alto Networks Prisma and Aruba SD-WAN. Ensure that the software is at (or above) the versions listed in the table. If the software is not updated before this infrastructure upgrade, VNFs are expected to be non-operational post upgrade.
|Required Software Version
|Palo Alto Networks
|5.5.3.b2 or above
|188.8.131.52 or above
We expect all VNFs to be operational upon the completion of the upgrade. However, we want you to be prepared to take action in the unforeseen event that your VNFs are not operational. See the Troubleshooting documentation to identify your device type and follow the instructions to recover your VNFs if required.
This infrastructure upgrade will result in a new internal identifier required for all devices. If this identifier is being used by an SD-WAN device to generate the serial number used in licensing/registration, you will be required to go through a vendor specific replacement process (also-know-as the RMA process).
We retain all the old serial numbers after the upgrade. However, when the VNF eventually reloads (due to reasons unrelated to this upgrade), the VNF will get a new serial number which will cause communication between the device and the SD-WAN controller to break. You will be advised to proactively address the issue by going through a vendor-specific replacement process for the devices listed in this section to generate a registration key for the new serial number. Once the key is retrieved, you can reprovision the VNF from the Network Edge Portal and all configurations will be restored from the SD-WAN controller.
For documentation about the RMA process, see: