Get replica state from backend before promotion
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| OpenStack Shared File Systems Service (Manila) |
Triaged
|
Medium
|
kiran pawar | ||
Bug Description
Description
===========
The promote share replica s using the NetApp driver.
Steps to reproduce
==================
* Create a share replica of the created share
* Wait for the replica to achieve 'in_sync' state
* Promote the share replica
* Wait for error
* sometime replica gets promoted, but sometime it fails.
filed bug to address failure scenario.
Possible description
=======
when replica state is retrieved from DB, it might be in IN_SYNC state. However this state is updated every 5 minute, it is possible that replica becomes OUT_SYNC after its state is read from backend. Then user tries to promote replica assuming its in IN_SYNC state. Also when retrieve from DB, state is IN_SYNC, but thereafter promotion fails since replica is actually OUT_SYNC. So retrieve current state from backend, then update in db and then try to promote only if its in IN_SYNC.
| tags: | added: netapp |
| tags: | added: replication |
| Changed in manila: | |
| status: | New → In Progress |
| Vida Haririan (vhariria) wrote | #1 |
| Vida Haririan (vhariria) wrote | #2 |
| Changed in manila: | |
| milestone: | none → antelope-rc1 |
| importance: | Undecided → Medium |
| assignee: | nobody → kiran pawar (kpdev) |
| OpenStack Infra (hudson-openstack) wrote Change abandoned on manila (master) | #3 |
| Changed in manila: | |
| milestone: | antelope-rc1 → bobcat-1 |
| Changed in manila: | |
| milestone: | bobcat-1 → bobcat-2 |
| Changed in manila: | |
| milestone: | bobcat-2 → bobcat-3 |
| Changed in manila: | |
| milestone: | bobcat-3 → bobcat-rc1 |
| Changed in manila: | |
| milestone: | bobcat-rc1 → caracal-1 |
| Changed in manila: | |
| status: | In Progress → Triaged |
| Vida Haririan (vhariria) wrote | #4 |
Link to the proposed fix https:/