Service by
Service
This method allows you to upgrade one service at a time.
Pro s: Rather than a single large service outage, you are able to stage
outages to specific services. For example, you could have the Identity
service running at the Havana release while Compute runs at the Grizzly
release.
You can schedule potentially longer upgrades, such as the Compute
service upgrade in a large environment, separately from upgrades that take
less time.
Co n s:You will still have an interruption to your Nova APIs and Compute
nodes.
Section 3.3, “ Method - Upgrade Service-by-Service”
Service by
Service with
Parallel
Computes
This method is a variation of the service-by-service upgrade, with a change in
how the Compute service is upgraded. Rather than upgrading your existing
Compute environment as part of the process, you deploy new nodes running
the Havana Compute services. You then wait for existing workloads on your
Grizzly Compute nodes to complete (or migrate them by hand), and when a
Grizzly Compute node is no longer hosting any instances, you upgrade the
Compute service on that node.
Pro s: This method minimizes interruptions to your Compute service, with
only a few minutes for the smaller services, and a longer migration interval
for the workloads moving to newly-upgraded Compute hosts. Existing
workloads can run indefinitely, and you do not need to wait for a database
migration.
Co n s: Additional hardware resources are required to bring up the Havana
Compute nodes.
Section 3.4, “ Method - Upgrade Service-by-Service with Parallel Computes”
Met h o d Descrip t io n
For all methods:
Ensure you have subscribed to the correct channels for this release on all hosts (see Section 1.3,
“ RHN/CD N Channels” ).
The upgrade will involve some service interruptions.
Running instances will not be affected by the upgrade process unless you either reboot a
Compute node or explicitly shut down an instance.
Comentários a estes Manuais