[GH-ISSUE #2475] [Bug]: Offload bucket eviction policy defaults to none, leading to disk exhaustion #655

New issue

Closed

opened 2026-10-29 21:17:43 +00:00 by bsakaguchi · 1 comment

bsakaguchi commented

2026-10-29 21:17:43 +00:00

Owner

Originally created by @NUABO on GitHub (Jun 15, 2026).
Original GitHub issue: https://github.com/kvcache-ai/Mooncake/issues/2475

Originally assigned to: @NUABO on GitHub.

Bug Report

When MOONCAKE_OFFLOAD_BUCKET_EVICTION_POLICY is not set, its default value is "none". This means no automatic eviction happens when the offload bucket reaches capacity. As a result, continued writes will eventually fail with disk full errors, even though the bucket has reached its configured limit.

Before submitting...

Ensure you searched for relevant issues and read the [documentation]

Originally created by @NUABO on GitHub (Jun 15, 2026). Original GitHub issue: https://github.com/kvcache-ai/Mooncake/issues/2475 Originally assigned to: @NUABO on GitHub. ### Bug Report When `MOONCAKE_OFFLOAD_BUCKET_EVICTION_POLICY` is not set, its default value is `"none"`. This means no automatic eviction happens when the offload bucket reaches capacity. As a result, continued writes will eventually fail with disk full errors, even though the bucket has reached its configured limit. ### Before submitting... - [ ] Ensure you searched for relevant issues and read the [documentation]

bsakaguchi

2026-10-29 21:17:43 +00:00

closed this issue
added the
bug
label

bsakaguchi commented

2026-10-29 21:17:44 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (Jun 15, 2026):

Thanks for opening this issue, @NUABO!

Field	Value
Issue	#2475
GitHub user ID	`29769310`
Reporter	@NUABO

A maintainer will triage this when possible. To help us respond faster, please include:

Mooncake version or commit SHA
Environment (OS, CUDA/driver, RDMA stack if relevant)
Steps to reproduce and expected vs. actual behavior

Useful links: Documentation · Contributing guide

This message was posted automatically by the issue bot.

@github-actions[bot] commented on GitHub (Jun 15, 2026): Thanks for opening this issue, @NUABO! | Field | Value | |-------|-------| | **Issue** | #2475 | | **GitHub user ID** | `29769310` | | **Reporter** | @NUABO | A maintainer will triage this when possible. To help us respond faster, please include: - Mooncake version or commit SHA - Environment (OS, CUDA/driver, RDMA stack if relevant) - Steps to reproduce and expected vs. actual behavior Useful links: [Documentation](https://kvcache-ai.github.io/Mooncake/) · [Contributing guide](https://github.com/kvcache-ai/Mooncake/blob/main/CONTRIBUTING.md) > This message was posted automatically by the issue bot.

bsakaguchi referenced this issue

2026-10-29 21:29:48 +00:00

[PR #655] [MERGED] fix(store) : fix disk-backed replicas in size calculation and slice allocation(#653) #1722