gsadmin/rustfs
1
0
Fork 0
mirror of https://github.com/rustfs/rustfs.git synced 2026-08-29 00:17:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(capacity): honor high-latency timeout profile (#6611)

Browse Source 
Co-authored-by: Henry Guo <marshawcoco@users.noreply.github.com>
This commit is contained in:
Henry Guo
2026-08-26 10:03:34 +08:00
committed by GitHub
parent 45c03ca37f
commit be22175035
4 changed files with 106 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/operations/drive-timeout-tuning.md
+2
View File
@@ -62,6 +62,8 @@ Explicit per-operation overrides always win over the profile, so you can select
| `RUSTFS_DRIVE_METADATA_TIMEOUT_SECS` | `5` | `60` | Timeout for metadata reads such as `read_metadata`. |
| `RUSTFS_DRIVE_DISK_INFO_TIMEOUT_SECS` | `5` | `60` | Timeout for `disk_info()` calls. |
| `RUSTFS_OBJECT_DISK_READ_TIMEOUT` | `10` | `60` | Per-read stall budget while streaming an object body from disk. |
| `RUSTFS_CAPACITY_STAT_TIMEOUT` | `3` | `60` | Base cooperative budget for an object-capacity disk scan. |
| `RUSTFS_CAPACITY_MAX_TIMEOUT` | `15` | `60` | Upper bound for the dynamically adjusted object-capacity scan budget. |
| `RUSTFS_DRIVE_MAX_TIMEOUT_DURATION` | `30` | — | Deprecated global fallback for every per-operation knob without an explicit override. Prefer the per-operation knobs. |
The health-transition and probe knobs (`RUSTFS_DRIVE_TIMEOUT_HEALTH_ACTION`,