gsadmin/rustfs
1
0
Fork 0
mirror of https://github.com/rustfs/rustfs.git synced 2026-07-26 08:18:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(ilm): bound manual transition duration (#5174)

Browse Source 
* feat(ilm): bound manual transition duration

Add maxDurationSeconds handling for manual transition runs so operators can bound long scoped scans without changing the existing enqueue_only job contract.

Co-Authored-By: heihutu <heihutu@gmail.com>

* docs(ilm): document manual transition run limits

Document the enqueue_only manual transition contract, secret-handling guidance, and best-effort duration budget while pinning the new duration flag in the e2e response model.

Co-Authored-By: heihutu <heihutu@gmail.com>

---------

Co-authored-by: heihutu <heihutu@gmail.com>
This commit is contained in:
houseme
2026-07-24 15:54:52 +08:00
committed by GitHub
parent 358caa23cb
commit 3132637294
4 changed files with 110 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/operations/tier-ilm-debugging.md
+25
View File
@@ -73,6 +73,31 @@ If both `x-rustfs-internal-transitioned-versionID` and
`x-minio-internal-transitioned-versionID` are the **empty string**, the object
was transitioned to a non-versioned tier bucket and no versionId must be sent.
## Manual transition run
Manual transition run is an operator trigger for the existing lifecycle transition evaluator. It does not force objects that are not due under the bucket lifecycle rule, and it does not bypass versioning, replication, delete-marker, directory-marker, tier, or in-flight transition checks.
The admin endpoint is:
```text
POST /rustfs/admin/v3/ilm/transition/run?bucket=<bucket>&prefix=<prefix>&tier=<tier>&dryRun=true&maxObjects=10000&maxDurationSeconds=30
```
Only `bucket` is required. `prefix`, `tier`, `dryRun`, `maxObjects`, and `maxDurationSeconds` narrow or bound the run. `maxObjects` defaults to `10000` and is capped at `100000`. `maxDurationSeconds` is optional, capped at `3600`, and enforced as a best-effort budget checked between listed object versions and pages; an in-flight listing call is not cancelled.
The current contract is `enqueue_only`: the response reports what this bounded scan evaluated and enqueued into the in-memory transition queue. `state=completed` means the bounded scan reached the end of its current scope without queue pressure or configured budget truncation. It does not mean every remote tier PUT has completed. `state=partial` means the run stopped early because it hit `maxObjects`, `maxDurationSeconds`, or queue pressure (`skipped_queue_full`, `skipped_queue_closed`, or `skipped_queue_timeout`).
`job_id` and `status_endpoint` are currently `null`. There is no durable background job, no restart recovery cursor, no cluster-wide single-flight admission, no status polling endpoint, and no cancel endpoint for this trigger yet. Re-running the command is safe only in the normal lifecycle sense: already in-flight object versions are deduplicated in the local process, but separate nodes do not yet share a durable manual-run admission record.
Recommended operator flow:
```bash
rc admin ilm transition run local/mybucket --prefix logs/ --tier cold --dry-run --max-objects 1000 --max-duration-seconds 30
rc admin ilm transition run local/mybucket --prefix logs/ --tier cold --max-objects 1000 --max-duration-seconds 30
```
Inspect the aggregate counters before widening scope. Full object-key lists are intentionally not returned by the admin response. If `RUSTFS_RPC_SECRET` or other credentials were pasted into an issue, chat, log, or ticket while debugging tiering, rotate them on every node, restart the cluster with the new value, and redact the exposed copy before sharing more diagnostics.
## Historical fixes (for context, already merged)
- Expire/GET race (`NoSuchVersion` during expiry of a tiered object):