[
https://issues.apache.org/jira/browse/FLINK-8360?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16365582#comment-16365582
]
ASF GitHub Bot commented on FLINK-8360:
---------------------------------------
Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/5239#discussion_r168479432
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/state/TaskExecutorLocalStateStoresManager.java
---
@@ -44,11 +44,21 @@
private final Map<JobID, Map<JobVertexSubtaskKey, TaskLocalStateStore>>
taskStateManagers;
/** This is the root directory for all local state of this task manager
/ executor. */
- private final File localStateRootDirectory;
+ private final File[] localStateRootDirectories;
- public TaskExecutorLocalStateStoresManager(File
localStateRootDirectory) {
+ public TaskExecutorLocalStateStoresManager(File[]
localStateRootDirectories) {
this.taskStateManagers = new HashMap<>();
- this.localStateRootDirectory =
Preconditions.checkNotNull(localStateRootDirectory);
+ this.localStateRootDirectories =
Preconditions.checkNotNull(localStateRootDirectories);
+
+ for (File localStateRecoveryRootDir :
localStateRootDirectories) {
+ if (!localStateRecoveryRootDir.exists()) {
+
+ if (!localStateRecoveryRootDir.mkdirs()) {
+ throw new IllegalStateException("Could
not create root directory for local recovery: " +
--- End diff --
Isn't this more of an `IOException`. This would make it a bit more explicit
what can happen if one instantiates the `TaskExecutorLocalStateStoresManager`.
> Implement task-local state recovery
> -----------------------------------
>
> Key: FLINK-8360
> URL: https://issues.apache.org/jira/browse/FLINK-8360
> Project: Flink
> Issue Type: New Feature
> Components: State Backends, Checkpointing
> Reporter: Stefan Richter
> Assignee: Stefan Richter
> Priority: Major
> Fix For: 1.5.0
>
>
> This issue tracks the development of recovery from task-local state. The main
> idea is to have a secondary, local copy of the checkpointed state, while
> there is still a primary copy in DFS that we report to the checkpoint
> coordinator.
> Recovery can attempt to restore from the secondary local copy, if available,
> to save network bandwidth. This requires that the assignment from tasks to
> slots is as sticky is possible.
> For starters, we will implement this feature for all managed keyed states and
> can easily enhance it to all other state types (e.g. operator state) later.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
