Manage API attempts to 'retry' manage if the same parameters are used

[Goutham Pacha Ravi]

Steps to reproduce:

Try to manage a share with an invalid share type
When the share goes to 'manage_error', try the same command again

There's this retry logic in the API that is unsettling with the REST way of doing things.

Share Manage API is admin-facing by default, the user experience for these users may not be greatly affected if this retry logic is dropped. They would have to delete the share before attempting to manage again.

[Goutham Pacha Ravi (gouthamr)]

It was decided in the Manila IRC meeting on 14/Apr that this action in the API must be disallowed. Microversion bump is unnecessary.

http://eavesdrop.openstack.org/meetings/manila/2016/manila.2016-04-14-15.00.txt

[NidhiMittalHada (nidhimittal19)]

working on this

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/315966

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/315966
Committed: https://git.openstack.org/cgit/openstack/manila/commit/?id=348fde021f3267ddec7ff77abc4c55fe5a166ffb
Submitter: Jenkins
Branch: master

commit 348fde021f3267ddec7ff77abc4c55fe5a166ffb
Author: nidhimittalhada <email address hidden>
Date: Fri May 13 12:59:54 2016 +0530

    Remove retry logic from manage API

    Lets say, we try to manage a share and share for some reason
    goes in 'manage_error' state. Now, if we try manage command again
    with the same parameters as earlier, then the logic in API,
    tries to reuse old record in DB.

    That is unsettling with the REST way of doing things.

    Share Manage API is admin-facing by default, Admin should
    delete the share before attempting to manage again.

    Hence dropping the retry logic.

    APIImpact

    Change-Id: I0eee6e108fc263e0d2e092eeaa6d0cba47b53dc4
    Closes-Bug: #1566504

[Thierry Carrez (ttx)]

This issue was fixed in the openstack/manila 3.0.0.0b1 development milestone.