(Ongoing) Planned Maintenance region FRA1 (Frankfurt) from 2020-01-20 00:00 to 2020-01-29 17:00 CET
2020-01-28 09.20 CET: Next up is Galera and RabbitMQ
2020-01-28 09.15 CET: Deplyoing servers works again.
2020-01-28 08.45 CET: Todays work is started. Disturbance deploying new servers can occur.
2020-01-27 17.15 CET: Work done for today, tomorrow upgrade to Train will continue.
2020-01-27 14.00 CET: LBaaSv2 is now removed, and next step is OpenStack upgrade from version Stein to Train.
2020-01-27 13.00 CET: The work with converting LBaaSv2 to Octavia is still ongoing for the nearest hour.
2020-01-27 08.30 CET: This morning we will start to convert LBaaSv2 to Octavia. Expect small disturbances on load balances up to 30 minutes.
2020-01-24 14.30 CET: We have now completed the upgrade to Openstack version Stein. Because of some delays in the beginning of this maintenance together with circumstances beyond our control today, we have to extend the maintenance between Monday 27/1 08:00 to Wednesday 29/1 17:00.
In the plan for next week, we have conversion of LBaaSv2 to Octavia and upgrade from Stein to the latest Openstack version Train.
2020-01-23 09.45 CET: Neutron upgrade is ongoing, and converting of LBs is next. Short’s network disturbance can occur.
2020-01-21 11.30 CET: All network related issues is now solved, and we will continue with the maintenance.
2020-01-21 09.00 CET: We experience network issues for new created servers. We are working on it.
2020-01-21 02.00 CET: All load balancers should now be functional.
2020-01-21 01.30 CET: Load balancers are coming online and troubleshooting continues for the remaining ones.
2020-01-21 00.20 CET: We are facing issues with bringing up the load balancers, troubleshooting has started.
2020-01-21 00.02 CET: The virtual instances are back online and the planned downtime has finished.
2020-01-20 23.58 CET: Network upgrades are done and we are bringing the virtual instances back online now.
2020-01-20 23.00 CET: We are preparing to reboot the hypervisors – planned downtime starts now.
2020-01-20 12.00 CET: Maintenance window have been opened and the first activities will be the planned downtime at 23:00.
Dear Valued Customer,
City Network would like to inform you that we have planned maintenance as scheduled below. This post is an update to previously communicated information.
Customer Intervention WILL BE required.
Region/datacenter: Openstack region FRA1 (Frankfurt, Germany)
Maintenance window starts: 2020-01-20 08.00 CET
Planned downtime starts: 2020-01-20 23.00 CET
Planned downtime ends: 2020-01-20 24.00 CET
Maintenance window ends: 2020-01-24 17.00 CET
Extension of Maintenance
Maintenance window starts: 2020-01-27 08.00 CET
Maintenance window ends: 2020-01-29 17.00 CET
Throughout the maintenance window, a number of improvements will be implemented – from functionality to security alignment.
The major part of the scope is an Openstack upgrade from Rocky to Train, which will offer certain new functions but also address a list of known bugs. Some functions will improve seen to maturity level and Magnum (Container Orchestation) in particular.
There will, in parallell, be work security wise to further increase protection and to address vulnerabilities. Furthermore there will be improvements to operations automation and those improvement will allow a high flexbility for us to operate the cloud more seamlessly.
Our ambition is to carry out the majority of the maintenance window with little to no impact on running services, with the exception of the stated downtime of 60 minutes.
Expected customer impact:
Due to the circumstances and the datacenter elements involved, City Network has after careful consideration, decided to introduce a major halt of all Openstack resources during a period of 60 minutes.
The planned downtime will take place Monday Jan 20th at 23.00 CET and lasting for 60 minutes.
While our maintenance strategy is to handle one compute node at a time – to allow for anti-affinity services (redundant customer services) to still be operational – we felt that the scenario with the slightest risks was to shut down all running instances throughout the duration of the network maintenance (outage)
The decision was based on that we see risks of network and storage disruptions due to the nature of the network upgrades, where longer periods of potential disruptions or even outages could occur – hence the decision to stop all virtual instances for a more controlled period of downtime
Customers are expected to ensure virtual instances (virtual servers) along with any services hosted on those, will properly start up after a reboot of an virtual instance (virtual server).
Furthermore, Load Balancer as a Service v2 will be deprecated and removed in the Openstack Train release – why conversion of all existing LBaaSv2 services will be carried out throughout the maintenance. Affected customers will be contacted individually with additional information.
Follow status at http://cnstatus.com for updates.
If you have further questions, please contact our support at email@example.com.