- What is going on?
- Why do the servers need to be migrated?
- When will the servers be migrated?
- Which servers will be migrated?
- Will there be a service interruption during this process? If so, how long will the servers be down?
- Will there be any network, IP Address or firewall changes?
- How will the servers be migrated?
- Do you have a fallback plan in case there is a problem with the migration?
- What do service owners need to do prior to and during the migration process?
- Will there be any difference in how my service and applications run or how I manage my servers?
- Our servers have attached or mounted storage. Will the external storage need to be moved too?
What is going on?
The Klinck Data Centre will be decommissioned and shutdown in 2015. UBC IT will be moving production virtual servers and storage from the Klinck Data Center to the new University Data Centre in the Pharmaceutical Sciences Building. Migrations of production environments will take place between October and December 2013, development and test environments will be moved by the end of 2015.
Why do the servers need to be migrated?
The Klinck Data Centre will be decommissioned and shutdown in 2015. The virtual infrastructure’s hardware in the Klinck Data Centre is several years old and vendor support will be expiring at the end of 2013.
UBC IT will be moving production virtual servers and storage from the Klinck Data Center to the new University Data Centre in the Pharmaceutical Sciences Building. Migrations of production environments will take place between October and December 2013, development and test environment will be moved by the end of 2015.
When will the servers be migrated?
The UBC IT Systems team will have provided you with a list of your group and department’s affected servers. UBC IT Systems will work with you to map out your server dependencies and to identify an acceptable window for migrating your servers. The service owners will need to advise their own clients about the service outage.
More information about each individual migration will be posted here: http://bulletins.it.ubc.ca/
Which servers and storage will be migrated?
In Phase 1 of the project, all production data and workload will be evacuated from the Klinck Data Centre. In Phase 2 of the project, all development and test data and workload will be moved.
Will there be a service interruption during this process? If so, how long will the servers be down?
Yes, unfortunately, the migration process involves upgrading certain packages and components and, if applicable, remounting storage. These changes can't be completed without shutting down/rebooting the servers.
The service outage will depend on the complexity of the environment being moved.
Will there be any network, IP Address or firewall changes?
No, since the related networks have been configured to span the data centres. Therefore no network configuration or IP address changes will be necessary and all firewall rules will continue to apply.
How will the servers be migrated?
Servers will exported from the Klinck Data Centre and imported into the University Data Centre (UDC). After being registered in the new environment, the servers will need to be shut off and rebooted to perform the required upgrades.
Do you have a fallback plan in case there is a problem with the migration?
Yes, A virtual server snapshot will be taken prior to each move so we can revert back to a current copy of the virtual server should we need to roll back.
What do service owners need to do prior to and during the migration process?
The UBC IT Systems team will work with the service owners to identify all steps to be performed during the migration process. These steps will include the listing of impacted servers, scheduling appropriate outage windows, and establishing the shutdown and start up sequence of interrelated servers and applications. The service owner will need to advise all clients to whom services are extended, have someone available during migration to shutdown and bring up application in the right sequence during migration, and to perform post migration quality assurance and validation testing of the environment.
Will there be any difference in how my service and applications run or how I manage my servers?
After the migration, services and applications will run the same way and provide equal or better performance. In terms of managing your servers, UDC has a new backup and recovery solution that will allow you to initiate your own restore process. More information will soon be available about the new backup and recovery solution.
Our servers have attached or mounted storage. Will the external storage need to be moved too?
Yes, storage will be moved to the UDC as well. A number of challenges exist with our current storage infrastructure; these challenges are making the management and scalability of our current storage environment very difficult. In order to meet the current and future needs of the university, we will be using the data and workload migration project as an opportunity to also complete a consolidation and re-organization of our storage infrastructure.
The goal is to consolidate and re-organize our storage from the current multiple island configuration to a storage cluster. The storage cluster configuration allows us to group the many existing storage islands into one physical and logical unit – this gives our infrastructure the scalability, performance, fault tolerance and manageability required to meet current and future needs. The storage cluster will also eliminate the need to disrupt services when hardware upgrades are required.
The conversion and migration of your storage will depend on what your storage needs are. The UBC IT Systems team will work with you to identify and implement the most appropriate storage solution for your needs.
Please send an email to it.sys.services@ubc.ca if you have any questions or comments about the migration.