Docs for list view and cluster import #65
Conversation
| ---------- | ||
| curl -XPOST -H "Content-Type: application/json" \ | ||
| -d '{ "node_ids": ["7026f371-2d8c-4ad7-895f-4d4dcc70ff1c"], \ | ||
| "sds_type":"ceph"}' http://127.0.0.1/api/1.0/ImportCluster |
There was a problem hiding this comment.
Could you add Gluster related example or mention what is a difference between importing cluster for Gluster and Ceph?
There was a problem hiding this comment.
that depends on the sds_type you send, it could be ceph or gluster.
| ---------- | ||
| Status: 200 OK | ||
| [{ | ||
| "integration_id": "717315a1-2a49-4cb7-826b-d2aa3bc4721a", |
There was a problem hiding this comment.
There should be maybe specified job_id in a response so it would be easier to access one item from jobs list.
There was a problem hiding this comment.
The job_id will be in response to ImportCluster for ex. which can be used the query the job with the Jobs Details API
| Sample Request | ||
| ---------- | ||
| curl -XPOST -d '{"Volume.volname":"Volume_009","Volume.bricks":["dhcp-1.lab.ten | ||
| drl.example:/root/bricks/vol9_b1"]}' http://127.0.0.1/api/1.0/5291c055-70d3-4450-9769-2f6fd4932afb/GlusterCreateVolume |
There was a problem hiding this comment.
Maybe it would be better to add example with more bricks and show how to specify how to create replicated/striped volumes if they are supported. Volume with one brick is a little misleading.
| "status": "finished", | ||
| "flow": "ImportCluster", | ||
| "parameters": { | ||
| "integration_id": "717315a1-2a49-4cb7-826b-d2aa3bc4721a", |
There was a problem hiding this comment.
Is there some way how to get cluster_id from job?
There was a problem hiding this comment.
@fbalak the integration_id is the cluster_id
There was a problem hiding this comment.
Ok, I also overlooked that there is cluster_id in parameters section of response (on line 37). Just why is it different from integration_id if they are same?
|
|
||
| == Import Cluster | ||
|
|
||
| Import an existing Gluster or Ceph cluster. |
There was a problem hiding this comment.
I think documentation and description of arguments will help to use this function. For example where user can get node IDs which can use in this function; where users can get list of available storage types(function or static list in documentation.
This comment is valid for all functions. Please describe function arguments.
| ---------- | ||
| Status: 202 Accepted | ||
|
|
||
| { job_id: "3784922e33e8bec939be5e626e21a174" } |
There was a problem hiding this comment.
Please add to docs where user can use this ID in return message..
|
|
||
| ---------- | ||
| curl -XGET | ||
| http://127.0.0.1/api/1.0/jobs/1cb0694c-8041-4576-95ee-f01f07738ff9 |
There was a problem hiding this comment.
What is 1cb0694c-8041-4576-95ee-f01f07738ff9 and where can user get it? I understand that this is trivial question but answer should be written in docs.
|
|
||
| `DELETE` to delete a resource | ||
|
|
||
| === Authentication |
There was a problem hiding this comment.
Is this accordingly with Tendrl/specifications#128 ?
| `Authorization: Tendrl <API_TOKEN>`. | ||
|
|
||
| ---------- | ||
| curl -H "Authorization: Tendrl <API_TOKEN>" http://127.0.0.1/api/1.0 |
| 3) Sending invalid fields will result in a `422 Unprocessable Entity` | ||
| response. | ||
|
|
||
| === HTTP Verbs |
There was a problem hiding this comment.
Can these verbs be applied to list of resources and single resources? See https://en.wikipedia.org/wiki/Representational_state_transfer#Relationship_between_URL_and_HTTP_methods
|
|
||
| === User Agent | ||
|
|
||
| Every API request must send a `User-Agent` property in the header to identity |
There was a problem hiding this comment.
Is this anyhow related to API_TOKEN? If yes, how?
There was a problem hiding this comment.
After conversation with @anivargi ; approving this PR and getting it ready for merge.
|
@sankarshanmukhopadhyay Is it possible to comment my questions, please? Or should I file documentation issues because this pull request is merged now? |
No description provided.