1. Query 'scope=system' and 'system=true' is neccessary when fetching all users
or groups from Keystone.
Without these, users and groups in other domains (non-caller domains),
as well as system-level users, cannot be detected.
2. Query 'scope=system' is neccessary when fetching all project from Keystone.
1. Add same return value with contact GET interface when updating
2. Add pull params for query to pull subcontact by mobile number.
For example, pull user_id from dingtalk throught user's mobile
number.
3. Add corresponding feature based on above for climc.
4. UserCache will try FetchByName if there is no id matched user when
fetching users from KeyStone.
1. Uname is availbale now in api via adding "uname=true" in query.
So we can use uname in climc which is much more visual and use uid in
web client which avoid time-consuming remote calls that may occur.
2. Previously, the template for the message was managed by the
individual sending components. Now, these templats was managed uniformly
by Notify Service so that we can easily add and remove.
3. Now, failure to send a validation message will trigger a notification
to the user via websocket.
- image list filter by uefi
- fix instance sanpshot delete
- guest clean expire server exclude pending deleted servers
- host register abort on network not match
1. Support generate_name for guest image
2. Only allow subimage of guest image to update description
3. Add support for creating vm from guest image in fetchDiskInfo(disk.go)
4. Fix the bug vm can't start automatically after build guest image
5. Fix the bug that project of guest image is incorrect
6. Fix the bug about updating guest image
Feature:
1. Add apis: list server by group, list group by server, bind
servers, unbind servers, bind groups, unbind groups.
2. Add climc command and docs corresponding to the above newly added apis
3. Add log about binding and unbinding operator.
4. Support enable and disable instance group.
Fix:
1. Set capacity zeor if there are not enough resource aboue forced
instance group in one host.
2. Change the stage(real delete => pending delete) that guest leave all groups.
