Tom Beerbower created AMBARI-7349:
-------------------------------------
Summary: Admin : LDAP Sync API to support sync plans
Key: AMBARI-7349
URL: https://issues.apache.org/jira/browse/AMBARI-7349
Project: Ambari
Issue Type: Task
Reporter: Tom Beerbower
Assignee: Tom Beerbower
Fix For: 1.7.0
Revised LDAP_SYNC API Proposal...
The ldap sync request resource contains the information needed to perform an
ldap sync. This includes the principal type (user, group), the sync type (all,
existing, specific), and an optional set of principal names. Wildcards will be
supported for principal names. You can mix and match specs, so it is possible
to have a request that syncs all users but specific groups, for example.
-- Create ldap sync requests ...
{code}
POST http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests
{
"specs" : [
{
"principal-type":"users",
"sync-type":"all"
},
{
"principal-type":"groups",
"sync-type":"all"
}
]
}
POST http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests
{
"specs" : [
{
"principal-type":"users",
"sync-type":"existing"
},
{
"principal-type":"groups",
"sync-type":"existing"
}
]
}
POST http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests
{
"specs" : [
{
"principal-type":"users",
"names":"fred,joe,admin*",
"sync-type":"specific"
},
{
"principal-type":"groups",
"names":"group1,group2,admin*",
"sync-type":"specific"
}
]
}
{code}
When you create a request it is assigned a unique id.
The response of each post will contain the href to the new ldap sync request
resource. For example,
{code}"href" :
"http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/1"{code}
The requests are not persisted in the DB. We will keep them in memory only for
now, possibly with an expiry value. If there is a need to persist them beyond
a stop then we can add the DB persistence later.
When a request is added, it will be placed on a queue. Requests will be
serviced sequentially (for now we will not handle concurrent syncing).
Request will be run as they are taken off the queue. In the future we may add
the ability to schedule requests.
-- Get all of the sync requests ...
{code}
GET http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests
{
"href" : "http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/",
"items" : [
{
"href" :
"http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/1",
"Request" : {
"id":1
},
"href" :
"http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/2",
"Request" : {
"id":2
},
"href" :
"http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/3",
"Request" : {
"id":3
},
}
]
}
{code}
You can view a specific sync request.
-- Get a specific sync request ...
{code}
GET http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/1
{
"href" : "http://c6401.ambari.apache.org:8080/api/v1/ldap-sync-requests/1/",
"Request":{
"id":1,
"state":"COMPLETE",
"sync-start-time":"10/12/2004 12:14:26.002"
"sync-end-time":"10/12/2004 12:15:36.006"
"users-fetched":200,
"users-created":12,
"users-updated":4,
"users-removed":5,
"groups-fetched":10,
"groups-created":3,
"groups-updated":2,
"groups-removed":1,
"membership-fetched":20,
"membership-created":1,
"membership-updated":5,
"specs" : [
{
"principal-type":"users",
"sync-type":"all"
},
{
"principal-type":"groups",
"sync-type":"all"
}
]
}
}
{code}
The values for the state property will be PENDING, RUNNING, COMPLETE. The
result values will be null until the state is COMPLETE.
When a request is COMPLETE, the ldap user / group / member resources of Ambari
should have been updated accordingly.
Note that the results do not show the specific users and groups that were
synced, only the totals. The specific users and groups touched by the sync can
be added if required.
Also note that this revised proposal removes the LDAP_SYNC resource as a root
level service. It doesn't seem to be needed to meet the current requirements.
We can add it later, if needed.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)