[
https://issues.apache.org/jira/browse/IGNITE-21805?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17849819#comment-17849819
]
Mikhail Efremov edited comment on IGNITE-21805 at 5/27/24 8:10 PM:
-------------------------------------------------------------------
*Done work*
# {{RaftManager}} is moved into {{ReplicaManager}} completely.
# Almost all calls from {{RaftManager}} are made as internal
{{{}ReplicaManager{}}}'s calls.
# {{ReplicaManager#replica()}} method is created and it returns a future
replica by given replication group ID; the method allows to get access to
raft-clients outside replication layer, but through it.
# The follow RAFT-related entities are moved into {{{}ReplicaManager{}}}:
{{{}RaftGroupOptions{}}}, {{RaftGroupEventsListener}}
({{{}RebalanceRaftGroupEventsListener{}}}), {{{}PartitionMover{}}},
{{{}TopologyAwareRaftGroupServiceFactory{}}}, {{{}LogStorageFactoryCreator{}}}.
# The creation place of raft-client in {{TableManager}} through
{{ReplicaManager#startRaftClient}} is localized now and will be removed in
IGNITE-22315
*Related next tickets*
# IGNITE-22315 is about making code to a state where if a node is related in
assignments, then and only then there will be created only one raft-node, only
one raft-client and only one replica. {{ReplicaManager#startRaftClient}} will
be removed as it's single call.
# IGNITE-22218 is about {{TableRaftService}} removing from the code and it's
calls for raft-clients will be replaced with a chains like
{{{}replicaMgr.replica(replicaGrpId).thenApply(Replica::raftClient).thenCompose(raftClient
-> ...){}}}.
# IGNITE-22292 is about {{ReplicaManager#getLogSyncer}} removal.
was (Author: JIRAUSER303791):
*Done work*
# {{RaftManager}} is moved into {{ReplicaManager}} completely.
# Almost all calls from {{RaftManager}} are made as internal
{{{}ReplicaManager{}}}'s calls.
# {{ReplicaManager#replica()}} method is created and it returns a future
replica by given replication group ID; the method allows to get access to
raft-clients outside replication layer, but through it.
# The follow RAFT-related entities are moved into {{{}ReplicaManager{}}}:
{{{}RaftGroupOptions{}}}, {{RaftGroupEventsListener}}
({{{}RebalanceRaftGroupEventsListener{}}}), {{{}PartitionMover{}}},
{{{}TopologyAwareRaftGroupServiceFactory{}}}, {{{}LogStorageFactoryCreator{}}}.
# The creation place of raft-client in {{TableManager}} through
{{ReplicaManager#startRaftClient}} is localized now and will be removed in
IGNITE-22315
*Related next tickets*
# IGNITE-22315 is about making code to a state where if a node is related in
assignments, then and only then there will be created only one raft-node, only
one raft-client and only one replica. {{ReplicaManager#startRaftClient}} will
be removed as it's single call.
# IGNITE-22218 is about {{TableRaftService}} removing from the code and it's
calls for raft-clients will be replaced with a chains like
{{{}replicaMgr.replica(replicaGrpId).thenCompose(raftClient -> ...){}}}.
# IGNITE-22292 is about {{ReplicaManager#getLogSyncer}} removal.
> Refactor TableManager and move all RAFT related pieces to Replica
> -----------------------------------------------------------------
>
> Key: IGNITE-21805
> URL: https://issues.apache.org/jira/browse/IGNITE-21805
> Project: Ignite
> Issue Type: Improvement
> Reporter: Kirill Gusakov
> Assignee: Mikhail Efremov
> Priority: Major
> Labels: ignite-3
> Time Spent: 2h 40m
> Remaining Estimate: 0h
>
> *Motivation*
> At the moment we have some places inside the TableManager, which:
> * Use RaftManager to start/stop RAFT nodes
> (handleChangePendingAssignmentEvent/handleChangeStableAssignmentEvent)
> * Use RaftGroupService through
> table.internalTable().tableRaftService().partitionRaftGroupService calls
> This fact prevents us on the track of zone-based collocation. The further
> collocation steps will be easier, if we will move the whole RAFT connected
> operation to the Replica class. Moreover, it should be there semantically
> *Definition of done*
> * All code inside the handleChangePendingAssignmentEvent connected with the
> start of raft groups (PartitionListener/RebalanceRaftGroupEventsListener) and
> raft clients must be moved to the start of the Replica itself
> * The same about handleChangeStableAssignmentEvent - the stop of Replica
> must stop appropriate raft node
> * All calls for
> table.internalTable().tableRaftService().partitionRaftGroupService must be
> replaced by the replicaMgr.replica(replicaGrpdId).raftClient()
> *Implementation notes*
> * The new temporary methods must be implemented and remove after IGNITE-22036
> ** ReplicaManager.replica(ReplicationGroupId replicaGrpId) - which returns
> the appropriate Replica by group id
> ** Replica.raftClient() - which return replica's RaftGroupService (raft
> client)
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
