Andrés Beck-Ruiz created CASSSIDECAR-371:
--------------------------------------------
Summary: Support coordination for cluster-wide operations in
operational job framework
Key: CASSSIDECAR-371
URL: https://issues.apache.org/jira/browse/CASSSIDECAR-371
Project: Sidecar for Apache Cassandra
Issue Type: Improvement
Reporter: Andrés Beck-Ruiz
The operational job framework currently only supports local jobs on nodes that
a Sidecar instance manages. In order to perform multi-node, up to cluster-wide
operations on Cassandra clusters, we need to add a coordination interface.
In the default implementation, the {{StorageOperationalJobCoordinator}} will be
used for coordination. This coordinator will use the {{StorageProvider}} to set
new operational jobs as active and ensure that a single operational job is
running at the same time. Other implementations of an operational job
coordinator can use different methods for setting operational jobs as active
and coordinating local node jobs between Sidecar instances.
This ticket also tracks the work to introduce a {{LocalJob}}, which are jobs on
local nodes that make up a larger operational job. For example, restarting a
three node cluster would result in one operational job and three local jobs,
one for each node. The {{LocalJobCoordinator}} will be responsible for polling
the {{StorageProvider}} for new operational jobs, and the {{LocalJobManager}}
will create a {{LocalJob}} once it is the Sidecar instance’s turn to execute an
operation on a local node.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
