HDDS-15323. Add storage capacity planning administrator guide.#436
HDDS-15323. Add storage capacity planning administrator guide.#436jojochuang wants to merge 2 commits into
Conversation
Document datanode reserved space, full-disk thresholds, and inspection workflows. Link from the Datanodes architecture page. Co-authored-by: Cursor <cursoragent@cursor.com>
|
|
||
| - **Purpose**: Protects the Datanode from OS-level disk-full errors. | ||
| - **Configuration**: | ||
| - `hdds.datanode.volume.min.free.space`: A fixed value (e.g., `20GB`). |
There was a problem hiding this comment.
@jojochuang Should we also mention about volume min free space hard limit?
- Soft limit (min.free.space.percent, default 2%): used for what is reported to SCM in heartbeats as freeSpaceToSpare. This governs SCM placement decisions.
- Hard limit (min.free.space.hard.limit.percent, default 1.5%): used for local write rejection on the Datanode.
There was a problem hiding this comment.
Ah yes. I missed these recent additions.
| To see the status of individual disks on a specific Datanode: | ||
|
|
||
| ```bash | ||
| ozone admin datanode usageinfo --uuid <datanode-uuid> |
There was a problem hiding this comment.
uuid is deprecated.
| ozone admin datanode usageinfo --uuid <datanode-uuid> | |
| ozone admin datanode usageinfo --node-id <datanode-uuid> |
|
|
||
| ### 3. Detailed Volume Info | ||
|
|
||
| To see the status of individual disks on a specific Datanode: |
There was a problem hiding this comment.
This is a bit misleading as datanode usageInfo it does not give individual disks info on specific Datanode.
Rather it should be,
| To see the status of individual disks on a specific Datanode: | |
| To see the aggregated capacity, usage, and remaining space for a specific Datanode: |
| | `hdds.datanode.volume.min.free.space.percent` | `0.02` (2%) | Percentage-based minimum free space. | | ||
| | `hdds.datanode.dir.du.reserved` | (unset) | Fixed bytes reserved for non-Ozone use. | | ||
| | `ozone.scm.container.size` | `5GB` | The target size for containers. | | ||
| | `hdds.datanode.storage.utilization.critical.threshold` | `0.95` (95%) | Threshold at which SCM logs a "Critical" space warning. | |
There was a problem hiding this comment.
We also mention about warning utilisation threshold :
<property>
<name>hdds.datanode.storage.utilization.warning.threshold</name>
<value>0.75</value>
<tag>OZONE, SCM, MANAGEMENT</tag>
<description>
If a datanode overall storage utilization exceeds more than this
value, a warning will be logged while processing the nodeReport in SCM.
</description>
</property>
Document soft/hard min free space limits, warning utilization threshold, --node-id CLI flag, and clarify per-datanode usageinfo scope. Co-authored-by: Cursor <cursoragent@cursor.com>
What changes were proposed in this pull request?
HDDS-15323. Add storage capacity planning administrator guide.
Please describe your PR in detail:
What is the link to the Apache Jira?
https://issues.apache.org/jira/browse/HDDS-15323
How was this patch tested?