Single node META FATE data #5127
Merged
Conversation
This file contains 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
- Moved all the fate data for a `META` transaction into a single ZK node - Pushed all the data into `NodeValue` (renamed to `FateData`) - Previously, `FateData` stored `TStatus`, `FateReservation`, and `FateKey`. Now additionally stores the `REPO` stack and `TxInfo`. - Status enforcement added to `MetaFateStore` (previously only possible for `UserFateStore`). - Moved `testFateInitialConfigCorrectness()` from `UserFateStoreIT` to `UserFateIT` - Renamed `UserFateStoreIT` to `UserFateStatusEnforcementIT` (now extends a new class `FateStatusEnforcementIT`) - Now only tests status enforcement (previously status enforcement + `testFateInitialConfigCorrectness()`) - Created `MetaFateStatusEnforcementIT` (extends `FateStatusEnforcementIT`) - Tests that the new status enforcement in `MetaFateStore` works - Created `FateStoreUtil`, moving the `createFateTable()` util here, created MetaFateZKSetup inner class here (the counterpart to `createFateTable()` for `UserFateStore` but sets up ZooKeeper for use in `MetaFateStore`) - Deleted `UserFateStoreIT`s (now `UserFateStatusEnforcementIT`) method `injectStatus` replacing with the existing `setStatus` which does the same thing
kevinrr888
commented
Dec 2, 2024
test/src/main/java/org/apache/accumulo/test/fate/meta/MetaMultipleStoresIT.java
Show resolved
Hide resolved
test/src/main/java/org/apache/accumulo/test/fate/FateStatusEnforcementIT.java
Show resolved
Hide resolved
test/src/main/java/org/apache/accumulo/test/fate/user/UserFateStoreIT.java
Show resolved
Hide resolved
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Show resolved
Hide resolved
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Show resolved
Hide resolved
keith-turner
reviewed
Dec 5, 2024
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Outdated
Show resolved
Hide resolved
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Outdated
Show resolved
Hide resolved
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Outdated
Show resolved
Hide resolved
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Outdated
Show resolved
Hide resolved
test/src/main/java/org/apache/accumulo/test/fate/meta/MetaMultipleStoresIT.java
Show resolved
Hide resolved
test/src/main/java/org/apache/accumulo/test/fate/FateStatusEnforcementIT.java
Show resolved
Hide resolved
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Show resolved
Hide resolved
- Got rid of END_REPO_DATA marker, replacing with simply writing the number of repos to read - Avoid AtomicBoolean in `MetaFateStore.FateTxStoreImpl.push(repo)` by changing StackOverflowException to instead be a RuntimeException (previously Exception) - Deleted unnecessary preexisting catch and immediate re-throw of a StackOverflowException in `MetaFateStore.FateTxStoreImpl.push(repo)` - Cleaned up and refactored `MetaFateStore` methods which mutate existing FateData; now reuse same pattern across these methods: all call new method `MetaFateStore.mutate()`
|
Addressed in 29edaa2:
|
kevinrr888
commented
Dec 9, 2024
core/src/main/java/org/apache/accumulo/core/fate/zookeeper/MetaFateStore.java
Show resolved
Hide resolved
|
Resolved the conflicts from the recent seed transaction speedup. Thankfully, those changes were mostly for UserFateStore and these are for MetaFateStore, so conflicts were very easy |
keith-turner
approved these changes
Dec 15, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
closes #4905
METAtransaction into a single ZK nodeNodeValue(renamed toFateData)FateDatastoredTStatus,FateReservation, andFateKey. Now additionally stores theREPOstack andTxInfo.MetaFateStore(previously only possible forUserFateStore).testFateInitialConfigCorrectness()fromUserFateStoreITtoUserFateITUserFateStoreITtoUserFateStatusEnforcementIT(now extends a new classFateStatusEnforcementIT)testFateInitialConfigCorrectness())MetaFateStatusEnforcementIT(extendsFateStatusEnforcementIT)MetaFateStoreworksFateStoreUtil, moving thecreateFateTable()util here, created MetaFateZKSetup inner class here (the counterpart tocreateFateTable()forUserFateStorebut sets up ZooKeeper for use inMetaFateStore)UserFateStoreITs (nowUserFateStatusEnforcementIT) methodinjectStatusreplacing with the existingsetStatuswhich does the same thingI did not end up using JSON for the fate data. I was not sure what the benefit of this would be since the data needs to be stored in ZK as a byte array anyways and it seemed simpler to me to just keep the same serialization/deserialization strategy for the fate data. Maybe this could be changed to JSON in a follow on if there would be a benefit of using JSON over the current strategy.