Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove PRRL creation/deletion in peer recovery of remote store enabled replica #4954

Merged
merged 10 commits into from
Nov 16, 2022

Conversation

ashking94
Copy link
Member

@ashking94 ashking94 commented Oct 27, 2022

Description

PRRL is used to keep lucene operation history and is also used in recovery for replaying these operations once the segment files are sent to the target. During replica recovery, PRRL is acquired for sending these operations to the replica after the segment files (from most recent safe commit) are copied over to replicas. This is not required when remote translog is enabled. This PR aims to remove the unnecessary code around PRRL in case of recovery of remote store enabled replicas.

Issues Resolved

#4502

Check List

  • New functionality includes testing.
    • All tests pass
  • New functionality has been documented.
    • New functionality has javadoc added
  • Commits are signed per the DCO using --signoff
  • Commit changes are listed out in CHANGELOG.md file (See: Changelog)

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@ashking94 ashking94 changed the title Add RecoverySourceHandlerFactory for extensibility Remove PRRL creation/deletion in replica peer recovery Oct 27, 2022
@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@ashking94 ashking94 force-pushed the 4502-1 branch 2 times, most recently from 6f88c46 to 79c1d1a Compare October 27, 2022 09:19
@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@codecov-commenter
Copy link

codecov-commenter commented Oct 27, 2022

Codecov Report

Merging #4954 (9e5eb24) into main (0eb6b2f) will increase coverage by 0.00%.
The diff coverage is 84.69%.

@@            Coverage Diff            @@
##               main    #4954   +/-   ##
=========================================
  Coverage     70.87%   70.88%           
+ Complexity    58105    58071   -34     
=========================================
  Files          4708     4711    +3     
  Lines        277559   277604   +45     
  Branches      40189    40190    +1     
=========================================
+ Hits         196729   196780   +51     
+ Misses        64722    64690   -32     
- Partials      16108    16134   +26     
Impacted Files Coverage Δ
...overy/RemoteStoreReplicaRecoverySourceHandler.java 77.77% <77.77%> (ø)
...indices/recovery/DefaultRecoverySourceHandler.java 80.68% <80.68%> (ø)
...search/indices/recovery/RecoverySourceHandler.java 77.96% <90.90%> (-0.16%) ⬇️
...org/opensearch/index/seqno/ReplicationTracker.java 68.70% <100.00%> (+0.34%) ⬆️
...ch/indices/recovery/PeerRecoverySourceService.java 60.00% <100.00%> (+3.67%) ⬆️
...indices/recovery/RecoverySourceHandlerFactory.java 100.00% <100.00%> (ø)
...org/opensearch/index/shard/IndexShardTestCase.java 94.02% <100.00%> (+0.07%) ⬆️
...g/opensearch/index/analysis/CharFilterFactory.java 0.00% <0.00%> (-100.00%) ⬇️
.../java/org/opensearch/node/NodeClosedException.java 50.00% <0.00%> (-50.00%) ⬇️
.../opensearch/client/indices/CloseIndexResponse.java 56.25% <0.00%> (-35.00%) ⬇️
... and 501 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

@ashking94 ashking94 changed the title Remove PRRL creation/deletion in replica peer recovery Remove PRRL creation/deletion in peer recovery of remote store enabled replica Oct 27, 2022
@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@ashking94 ashking94 force-pushed the 4502-1 branch 2 times, most recently from e6cd4a9 to 77d0d3d Compare October 27, 2022 15:57
@ashking94 ashking94 marked this pull request as ready for review October 27, 2022 15:57
@ashking94 ashking94 requested review from a team and reta as code owners October 27, 2022 15:57
@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@ashking94
Copy link
Member Author

@Bukhtawar @sachinpkale Can you take a look at this?

@gbbafna
Copy link
Collaborator

gbbafna commented Nov 2, 2022

Changes look good to me .

Few clarifying questions :

  1. How did we test out these changes ? Will we do end to end testing later ?
  2. What is the impact of not removing PRRL for primary's recovery ? I do understand we do need PRRL for primary. Trying to understand the end goal here .

@ashking94
Copy link
Member Author

@gbbafna thanks for the review -

How did we test out these changes ? Will we do end to end testing later ?

There are units tests (OpenSearchTestCase) - ReplicaRecoveryWithRemoteTranslogOnPrimaryTests. Apart from that, validated the same by running the recovery with tweaking some changes locally. There would be more IT tests added once we have remote translog changes merged.

What is the impact of not removing PRRL for primary's recovery ? I do understand we do need PRRL for primary. Trying to understand the end goal here.

PRRL would still be used for doing primary-primary peer recovery. PRRL helps in doing sequence based recovery by replaying the same lucene operations on the recovering replica (which would be bumped to primary).

@@ -324,6 +324,7 @@ void awaitEmpty() {

private final class ShardRecoveryContext {
final Map<RecoverySourceHandler, RemoteRecoveryTargetHandler> recoveryHandlers = new HashMap<>();
private final RecoverySourceHandlerFactory recoverySourceHandlerFactory = new RecoverySourceHandlerFactory();
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: This can be a singleton.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Java does not allow creating static fields within inner classes. It's supported, I guess, from 16/17+.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why is the factory an instance at all? It contains no state. I think you can make RecoverySourceHandlerFactory#create static and call it below without creating an instance.

StartRecoveryRequest request,
RecoverySettings recoverySettings
) {
boolean isReplicaRecoveryWithRemoteTranslog = request.isPrimaryRelocation() == false && shard.isRemoteTranslogEnabled();
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why are we checking request.isPrimaryRelocation() == false?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

request.isPrimaryRelocation() == false is replica recovery. We want to use the RemoteStoreReplicaRecoverySourceHandler only for recovering the replicas. When request.isPrimaryRelocation() == true, it is primary-primary recovery for which it would be DefaultRecoverySourceHandler. Primary-primary recovery would involve translogs and lucene operation replay during the peer recovery.

@@ -215,7 +215,7 @@ public void writeFileChunk(
});
}
};
RecoverySourceHandler handler = new RecoverySourceHandler(
RecoverySourceHandler handler = new DefaultRecoverySourceHandler(
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These tests essentially become tests for DefaultRecoverySourceHandler. Don't we need tests for RemoteStoreReplicaRecoverySourceHandler as well?

Copy link
Member Author

@ashking94 ashking94 Nov 3, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The RecoverySourceHandlerTests now covers both the RecoverySourceHandler and DefaultRecoverySourceHandler. The tests for RemoteStoreReplicaRecoverySourceHandler were written in an older PR and is present in ReplicaRecoveryWithRemoteTranslogOnPrimaryTests.java. Let me decompose it and create another test class dedicated to the new Handler.

@@ -902,7 +902,7 @@ private boolean invariant() {
if (primaryMode && indexSettings.isSoftDeleteEnabled() && hasAllPeerRecoveryRetentionLeases) {
// all tracked shard copies have a corresponding peer-recovery retention lease
for (final ShardRouting shardRouting : routingTable.assignedShards()) {
if (checkpoints.get(shardRouting.allocationId().getId()).tracked) {
if (checkpoints.get(shardRouting.allocationId().getId()).tracked && !indexSettings().isRemoteTranslogStoreEnabled()) {
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is required as currently if a shard is tracked then it is expected to have a corresponding retention lease.

@ashking94 ashking94 force-pushed the 4502-1 branch 2 times, most recently from 561d668 to 5146fdc Compare November 7, 2022 06:37
@ashking94 ashking94 requested a review from sachinpkale November 7, 2022 06:37
@github-actions
Copy link
Contributor

github-actions bot commented Nov 7, 2022

Gradle Check (Jenkins) Run Completed with:

@github-actions
Copy link
Contributor

github-actions bot commented Nov 7, 2022

Gradle Check (Jenkins) Run Completed with:

@github-actions
Copy link
Contributor

github-actions bot commented Nov 7, 2022

Gradle Check (Jenkins) Run Completed with:

Signed-off-by: Ashish Singh <[email protected]>
Signed-off-by: Ashish Singh <[email protected]>
@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@ashking94 ashking94 requested a review from Bukhtawar November 15, 2022 08:50
Copy link
Member

@andrross andrross left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good, mostly minor comments

@@ -106,21 +102,21 @@
*
* @opensearch.internal
*/
public class RecoverySourceHandler {
public abstract class RecoverySourceHandler {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you make the constructor package-private to prevent this from being extended in ways not intended? I'm assuming this should only ever be instantiated via one of the subclasses in this package.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ack

import org.opensearch.index.shard.IndexShard;

/**
* Factory that supplies {@link RecoverySourceHandler}.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please add @opensearch.internal

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ack

import java.util.function.Consumer;

/**
* This handler is used when peer recovery target is a remote store enabled replica.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please add @opensearch.internal

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ack


/**
* This handler is used for the peer recovery when there is no remote store available for segments/translogs. TODO -
* Add more details as this is refactored further.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please add @opensearch.internal

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ack

@@ -324,6 +324,7 @@ void awaitEmpty() {

private final class ShardRecoveryContext {
final Map<RecoverySourceHandler, RemoteRecoveryTargetHandler> recoveryHandlers = new HashMap<>();
private final RecoverySourceHandlerFactory recoverySourceHandlerFactory = new RecoverySourceHandlerFactory();
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why is the factory an instance at all? It contains no state. I think you can make RecoverySourceHandlerFactory#create static and call it below without creating an instance.

import java.util.function.Consumer;

/**
* This handler is used for the peer recovery when there is no remote store available for segments/translogs. TODO -
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is probably better to document this based on what it is/what it does, versus what it is not. e.g. "this handler is used for node-to-node recovery..."

It might also be better to give it a name more descriptive that "default". "Default" makes sense when you are adding a new alternative but will be less obvious over time. Maybe something like "RetentionLeaseRecoverySourceHandler" or "LocalRecoverySourceHandler"?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Makes sense. Will make the change.

@Override
protected void innerRecoveryToTarget(ActionListener<RecoveryResponse> listener, Consumer<Exception> onFailure) throws IOException {
// A replica of an index with remote translog does not require the translogs locally and keeps receiving the
// updated segments file on refresh, flushes, and merges. We plan to make the existing replication call to
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The "We plan to" comment is confusing here. It suggests something here needs to change in the future. Are these sentences needed beyond the first sentence that says a local translog isn't needed?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, the "we plan to" comment tells us what is coming as this is an incremental PR to the bigger effort. However, I get that we can make it shorter and concise. I have reworded the "we plan to" section. Let me know if this makes sense.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The update looks good, thanks!

I know this is nit-picky but comments like "we plan to..." very often become stale and will still be present in the code even after all the planned work is done, because the future work may not touch this exact piece of code again. So unless there is a strong need to document future intent I find it is better to try to avoid it.

@ashking94 ashking94 requested a review from andrross November 16, 2022 14:37
@github-actions
Copy link
Contributor

Gradle Check (Jenkins) Run Completed with:

@andrross andrross merged commit 0be3afb into opensearch-project:main Nov 16, 2022
@andrross
Copy link
Member

andrross commented Nov 16, 2022

@ashking94 Should this be backported to 2.x? I've added the label but want to get confirmation from you.

@andrross andrross added the backport 2.x Backport to 2.x branch label Nov 16, 2022
@opensearch-trigger-bot
Copy link
Contributor

The backport to 2.x failed:

The process '/usr/bin/git' failed with exit code 128

To backport manually, run these commands in your terminal:

# Fetch latest updates from GitHub
git fetch
# Create a new working tree
git worktree add ../.worktrees/backport-2.x 2.x
# Navigate to the new working tree
pushd ../.worktrees/backport-2.x
# Create a new branch
git switch --create backport/backport-4954-to-2.x
# Cherry-pick the merged commit of this pull request and resolve the conflicts
git cherry-pick -x --mainline 1 0be3afbd249f771057663059279b4d02e4d282df
# Push it to GitHub
git push --set-upstream origin backport/backport-4954-to-2.x
# Go back to the original working tree
popd
# Delete the working tree
git worktree remove ../.worktrees/backport-2.x

Then, create a pull request where the base branch is 2.x and the compare/head branch is backport/backport-4954-to-2.x.

@ashking94
Copy link
Member Author

@ashking94 Should this be backported to 2.x? I've added the label but want to get confirmation from you.

Yes, the plan is to get request level durability in 2.5 for which this is an incremental step.

@ashking94
Copy link
Member Author

ashking94 commented Nov 17, 2022

Thanks for the review, @andrross, @sachinpkale & @gbbafna!

@andrross
Copy link
Member

@ashking94 Auto-backport failed. Can you go ahead and create the backport PR?

ashking94 added a commit to ashking94/OpenSearch that referenced this pull request Jan 6, 2023
…d replica (opensearch-project#4954)

* Add RecoverySourceHandlerFactory for extensibility

Signed-off-by: Ashish Singh <[email protected]>

* recoverToTarget made extensible to allow multiple implementations

Signed-off-by: Ashish Singh <[email protected]>

* Remove PRRL after SendFileStep in Peer Recovery

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Remove CHANGELOG entry as this is incremental PR

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>
ashking94 added a commit to ashking94/OpenSearch that referenced this pull request Jan 6, 2023
…d replica (opensearch-project#4954)

* Add RecoverySourceHandlerFactory for extensibility

Signed-off-by: Ashish Singh <[email protected]>

* recoverToTarget made extensible to allow multiple implementations

Signed-off-by: Ashish Singh <[email protected]>

* Remove PRRL after SendFileStep in Peer Recovery

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Remove CHANGELOG entry as this is incremental PR

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>
gbbafna pushed a commit that referenced this pull request Jan 9, 2023
…5731)

* Remove PRRL creation/deletion in peer recovery of remote store enabled replica (#4954)

* Add RecoverySourceHandlerFactory for extensibility

Signed-off-by: Ashish Singh <[email protected]>

* recoverToTarget made extensible to allow multiple implementations

Signed-off-by: Ashish Singh <[email protected]>

* Remove PRRL after SendFileStep in Peer Recovery

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Remove CHANGELOG entry as this is incremental PR

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>

* Enhance CheckpointState to support no-op replication (#5282)

* CheckpointState enhanced to support no-op replication

Signed-off-by: Ashish Singh <[email protected]>
Co-authored-by: Bukhtawar Khan<[email protected]>
Signed-off-by: Ashish Singh <[email protected]>

* Add transport action for primary term validation for remote-backed indices (#5616)

Add transport action for primary term validation for remote-backed indices

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>
sachinpkale pushed a commit to sachinpkale/OpenSearch that referenced this pull request Jan 10, 2023
…pensearch-project#5731)

* Remove PRRL creation/deletion in peer recovery of remote store enabled replica (opensearch-project#4954)

* Add RecoverySourceHandlerFactory for extensibility

Signed-off-by: Ashish Singh <[email protected]>

* recoverToTarget made extensible to allow multiple implementations

Signed-off-by: Ashish Singh <[email protected]>

* Remove PRRL after SendFileStep in Peer Recovery

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Remove CHANGELOG entry as this is incremental PR

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>

* Enhance CheckpointState to support no-op replication (opensearch-project#5282)

* CheckpointState enhanced to support no-op replication

Signed-off-by: Ashish Singh <[email protected]>
Co-authored-by: Bukhtawar Khan<[email protected]>
Signed-off-by: Ashish Singh <[email protected]>

* Add transport action for primary term validation for remote-backed indices (opensearch-project#5616)

Add transport action for primary term validation for remote-backed indices

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>
gbbafna pushed a commit that referenced this pull request Jan 10, 2023
* Remove PRRL creation/deletion in peer recovery of remote store enabled replica (#4954)

* Add RecoverySourceHandlerFactory for extensibility

Signed-off-by: Ashish Singh <[email protected]>

* recoverToTarget made extensible to allow multiple implementations

Signed-off-by: Ashish Singh <[email protected]>

* Remove PRRL after SendFileStep in Peer Recovery

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Remove CHANGELOG entry as this is incremental PR

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>

* Enhance CheckpointState to support no-op replication (#5282)

* CheckpointState enhanced to support no-op replication

Signed-off-by: Ashish Singh <[email protected]>
Co-authored-by: Bukhtawar Khan<[email protected]>
Signed-off-by: Ashish Singh <[email protected]>

* Add transport action for primary term validation for remote-backed indices (#5616)

Add transport action for primary term validation for remote-backed indices

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>
Co-authored-by: Ashish <[email protected]>
kotwanikunal pushed a commit that referenced this pull request Jan 25, 2023
…5731)

* Remove PRRL creation/deletion in peer recovery of remote store enabled replica (#4954)

* Add RecoverySourceHandlerFactory for extensibility

Signed-off-by: Ashish Singh <[email protected]>

* recoverToTarget made extensible to allow multiple implementations

Signed-off-by: Ashish Singh <[email protected]>

* Remove PRRL after SendFileStep in Peer Recovery

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Empty-Commit

Signed-off-by: Ashish Singh <[email protected]>

* Remove CHANGELOG entry as this is incremental PR

Signed-off-by: Ashish Singh <[email protected]>

* Incorporate PR review feedback

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>

* Enhance CheckpointState to support no-op replication (#5282)

* CheckpointState enhanced to support no-op replication

Signed-off-by: Ashish Singh <[email protected]>
Co-authored-by: Bukhtawar Khan<[email protected]>
Signed-off-by: Ashish Singh <[email protected]>

* Add transport action for primary term validation for remote-backed indices (#5616)

Add transport action for primary term validation for remote-backed indices

Signed-off-by: Ashish Singh <[email protected]>

Signed-off-by: Ashish Singh <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backport 2.x Backport to 2.x branch skip-changelog
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants