HDDS-7787. GetChecksum for EC files can fail intermittently with IndexOutOfBounds exception #4180
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…xOutOfBounds exception
adoroszlai
approved these changes
Jan 17, 2023
jojochuang
pushed a commit
to jojochuang/ozone
that referenced
this pull request
Feb 6, 2023
…xOutOfBounds exception (apache#4180) (cherry picked from commit c61ee3e) Change-Id: I0e7964c9ce10aaa29ca3b85c1407bd289ea2c467
errose28
added a commit
to errose28/ozone
that referenced
this pull request
Apr 20, 2023
* master: (209 commits) HDDS-7097. Container scanner log output lacks useful information (apache#4169) HDDS-7813. Handle Mismatched Replicas (OPEN or CLOSING) of QUASI-CLOSED containers in RM (apache#4195) HDDS-7625. Do not compress OM/SCM checkpoints (apache#4130) HDDS-7801. Bucket not found when calling getKeyInfo with tenant context (apache#4189) HDDS-7807. TarContainerPacker closes streams multiple times (apache#4193) HDDS-7755. Ensure that acquired locks are always released. (apache#4191) HDDS-7804. UNHEALTHY replicas will not contribute to sufficient replication in RatisContainerReplicaCount (apache#4192) HDDS-7748. Rename OMFileRequest.addToOpenFileTable() to avoid misuse. (apache#4176) HDDS-7723. Refresh Keys and Certificate used in OzoneSecretManager after certificate renewed (apache#4179) HDDS-7788. Ratis OverReplicationHandler should exclude stale replicas (apache#4183) HDDS-7718. Bump Netty to 4.1.86 and gRPC to 1.51.1 (apache#4139) HDDS-7542. Refactor DefaultReplicationConfig (apache#4005) HDDS-7787. GetChecksum for EC files can fail intermittently with IndexOutOfBounds exception (apache#4180) HDDS-7754. Download of container is failing with SSL/TLS error during re-replication (apache#4174) HDDS-7455. ClassCastException: OzoneTokenIdentifier cannot be cast to String (apache#4159) HDDS-7441. Rename function names of retrieving metadata keys (apache#3918) HDDS-7722. FSO buckets fail to invalidate open file table cache when committing a key (apache#4156) HDDS-7774. Update outdated Trash documentation (apache#4172) HDDS-7761. EC: ReplicationManager - Use placementPolicy.replicasToRemoveToFixOverreplication in EC Over replication handler (apache#4166) HDDS-7775. EC: Exception encountered while deleting UNHEALTHY replica in Datanode (apache#4173) ...
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
When calculating a checksum for an EC file with Rack Topology enabled, you can get the following error intermittently:
This is because the wrong nodes are used to obtain the stripe checksum sometimes as the node does not correctly use the replicaIndex in the pipeline to order the nodes.
What is the link to the Apache JIRA
https://issues.apache.org/jira/browse/HDDS-7787
How was this patch tested?
An existing test covers the checksum validate, so it confirms this change has not broken anything. The actual problem is difficult to reproduce in a unit test as the rack awareness is not easy to setup in such a way to affect the node order in the pipeline. We do have a reproducible test with a Hive workload that causes this, so we can validate the fix that way after this has been committed.