fix(delete,retention): clear DuckDB cache and free OS memory after execution#372
Merged
fix(delete,retention): clear DuckDB cache and free OS memory after execution#372
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request addresses a memory retention issue by clearing DuckDB's internal caches and triggering garbage collection after delete and retention operations, while also optimizing memory usage during file rewrites by setting a row group size. The review highlights that the current implementation misses several exit paths and dry-run scenarios where DuckDB caches are still populated, suggesting the use of a defer block to ensure comprehensive cache cleanup.
…ecution Neither the delete handler nor the retention handler cleared DuckDB's parquet metadata/data cache after completing their file operations, causing memory to accumulate and never be released — requiring container restarts. Mirrors the cleanup pattern already used by compaction. - Call ClearHTTPCache() after delete/retention completes (always, including dry runs and no-match paths — read_parquet populates cache regardless) - Add freeOSMemoryThrottled() — debounced via atomic CAS, fires at most once every 30s in a goroutine to prevent GC storms under concurrent requests - Add ROW_GROUP_SIZE 122880 constant (parquetRowGroupSize) to cap write buffering during file rewrites, matching compaction's row group size - Fix stale "after compaction" log messages in ClearHTTPCache()
xe-nvdk
force-pushed
the
fix/delete-retention-memory
branch
from
bcaa2b0 to
9bb839f
Compare
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Summary
delete.goandretention.gowere not clearing DuckDB's parquet metadata/data cache after completing file operations, causing memory to accumulate permanently and requiring container restarts to recoverClearHTTPCache+FreeOSMemory)FreeOSMemoryruns in a goroutine to avoid blocking the HTTP response with a GC pauseROW_GROUP_SIZE 122880to delete COPY queries to match compaction's write bufferingClearHTTPCache()Test plan
go build ./internal/...passesdocker stats— memory should return to baseline after completion!req.DryRun)Context
Reported by a user running Arc in Docker with Garage S3 backend, who observed memory climbing by several GBs each night during nightly retention runs, with no recovery until container restart. Reproducible on v26.03.1 and v26.04.1.