add release note

This commit is contained in:
Christian McHugh 2020-05-04 10:13:47 +01:00 committed by Daniel Wozniak
parent 075fef3b3d
commit fc8189928a

View file

@ -155,12 +155,33 @@ Vault Module
------------
The :py:func:`vault module <salt.modules.vault>` has been updated with the ability
to cache generated tokens. By specifying ``ttl`` or ``uses`` the token generated on
to cache generated tokens. By specifying ``uses`` and optionally ``ttl``, the token generated on
behalf of the minion will be allowed to persist and function for the defined time period
or number of uses.
or number of uses. Setting ``uses: 0`` creates an unlimited use token, that is only constrained by
the ``ttl``.
.. code-block:: yaml
vault:
auth:
uses: 25
This functionality is configured by default on the master and is thus shared behavior for all minion token generation.
To delegate use count to individual minions, specify ``allow_minion_override: True`` in the master config, and define
``uses`` and ``ttl`` in the minion config as directed above.
.. code-block:: yaml
vault:
auth:
method: token
allow_minion_override: True
Additionally, the vault module now supports Vault secrets backend version 2. The approperate secrets backend will be
automatically detected, and cached in the same credentials file as long lived vault tokens mentioned above. For any
configurations that worked around KV v2 handling by adding a manual data key to the end of vault lookups,
``salt['vault'].read_secret('secret/my/secret')['data']``, these are automatically detected and will continue to
function, but will generate a debug log message and can be removed.
The long lived token and secret metadata cache file can be cleared with the new ``vault.clear_token_cache``
execution function.