[
https://issues.apache.org/jira/browse/FLINK-12296?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16826787#comment-16826787
]
Stefan Richter commented on FLINK-12296:
----------------------------------------
[~klion26] I was just taking a look into the code and in my opinion the paths
should be already good as they are, each operator goes into a separate
directory already inside the {{chk_x}} directory. However, the problem is with
{{RocksIncrementalSnapshotStrategy#prepareLocalSnapshotDirectory}}, which a bit
to ambitiously does things like {{if (directory.exists())
FileUtils.deleteDirectory(directory)}}. What about just not deleting the
existing directory here, just checking this for the sub-directory of the
operator itself, but not the chk-root? If I am not missing anything here, that
would should be a more appropriate fix. What do you think?
> Data loss silently in RocksDBStateBackend when more than one operator(has
> states) chained in a single task
> -----------------------------------------------------------------------------------------------------------
>
> Key: FLINK-12296
> URL: https://issues.apache.org/jira/browse/FLINK-12296
> Project: Flink
> Issue Type: Bug
> Components: Runtime / State Backends
> Affects Versions: 1.6.3, 1.6.4, 1.7.2, 1.8.0
> Reporter: Congxian Qiu(klion26)
> Assignee: Congxian Qiu(klion26)
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 1.7.3, 1.9.0, 1.8.1
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> As the mail list said[1], there may be a problem when more than one operator
> chained in a single task, and all the operators have states, we'll encounter
> data loss silently problem.
> Currently, the local directory we used is like below
> ../local_state_root_1/allocation_id/job_id/vertex_id_subtask_idx/chk_1/(state),
>
> if more than one operator chained in a single task, and all the operators
> have states, then all the operators will share the same local
> directory(because the vertext_id is the same), this will lead a data loss
> problem.
>
> The path generation logic is below:
> {code:java}
> // LocalRecoveryDirectoryProviderImpl.java
> @Override
> public File subtaskSpecificCheckpointDirectory(long checkpointId) {
> return new File(subtaskBaseDirectory(checkpointId),
> checkpointDirString(checkpointId));
> }
> @VisibleForTesting
> String subtaskDirString() {
> return Paths.get("jid_" + jobID, "vtx_" + jobVertexID + "_sti_" +
> subtaskIndex).toString();
> }
> @VisibleForTesting
> String checkpointDirString(long checkpointId) {
> return "chk_" + checkpointId;
> }
> {code}
> [1]
> [http://mail-archives.apache.org/mod_mbox/flink-user/201904.mbox/%3cm2ef5tpfwy.wl-nings...@gmail.com%3E]
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
