<div dir="ltr"><div><div><br><div class="gmail_quote"><div dir="ltr">On Fri, Sep 28, 2018 at 1:03 PM Harry Rybacki <<a href="mailto:hrybacki@redhat.com">hrybacki@redhat.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On Fri, Sep 28, 2018 at 1:57 PM Morgan Fainberg<br>
<<a href="mailto:morgan.fainberg@gmail.com">morgan.fainberg@gmail.com</a>> wrote:<br>
><br>
> Ideally I would like to see it in the form of least specific to most specific. But more importantly in a way that there is no additional delimiters between the service type and the resource. Finally, I do not like the change of plurality depending on action type.<br>
><br>
> I propose we consider<br>
><br>
> <service-type>:<resource>:<action>[:<subaction>]<br>
><br>
> Example for keystone (note, action names below are strictly examples I am fine with whatever form those actions take):<br>
> identity:projects:create<br>
> identity:projects:delete<br>
> identity:projects:list<br>
> identity:projects:get<br>
><br>
> It keeps things simple and consistent when you're looking through overrides / defaults.<br>
> --Morgan<br>
+1 -- I think the ordering if `resource` comes before<br>
`action|subaction` will be more clean.<br></blockquote><div><br></div><div>++</div><div><br></div><div>These are excellent points. I especially like being able to omit the convention about plurality. Furthermore, I'd like to add that I think we should make the resource singular (e.g., project instead or projects). For example:<div><br></div><div>compute:server:list<br><div>compute:server:update</div><div>compute:server:create</div><div>compute:server:delete</div><div>compute:server:action:reboot</div><div>compute:server:action:confirm_resize (or confirm-resize)<br></div><div><br></div><div>Otherwise, someone might mistake compute:servers<span class="gmail-gr_ gmail-gr_21 gmail-gr-alert gmail-gr_gramm gmail-gr_inline_cards gmail-gr_run_anim gmail-Style gmail-replaceWithoutSep" id="gmail-21" style="display:inline;border-bottom:2px solid transparent;background-repeat:no-repeat;color:inherit;font-size:inherit">:get</span>, <span class="gmail-gr_ gmail-gr_18 gmail-gr-alert gmail-gr_gramm gmail-gr_inline_cards gmail-gr_run_anim gmail-Grammar gmail-multiReplace" id="gmail-18" style="display:inline;border-bottom:2px solid transparent;background-repeat:no-repeat;color:inherit;font-size:inherit">as</span> "list". This is ultra-nick-picky, but something I thought of when seeing the usage of "get_all" in policy names in favor of "list."</div><div><br></div><div>In summary, the new convention based on the most recent feedback should be:</div><div><br></div><div><b><service-type>:<resource>:<action>[:<subaction>]</b></div><div><br></div><div>Rules:</div><div><ul><li>service-type is always defined in the service types authority</li><li>resources are always singular</li></ul></div><div>Thanks to all for sticking through this tedious discussion. I appreciate it.</div></div></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
/R<br>
<br>
Harry<br>
><br>
> On Fri, Sep 28, 2018 at 6:49 AM Lance Bragstad <<a href="mailto:lbragstad@gmail.com">lbragstad@gmail.com</a>> wrote:<br>
>><br>
>> Bumping this thread again and proposing two conventions based on the discussion here. I propose we decide on one of the two following conventions:<br>
>><br>
>> <service-type>:<action>:<resource><br>
>><br>
>> or<br>
>><br>
>> <service-type>:<action>_<resource><br>
>><br>
>> Where <service-type> is the corresponding service type of the project [0], and <action> is either create, get, list, update, or delete. I think decoupling the method from the policy name should aid in consistency, regardless of the underlying implementation. The HTTP method specifics can still be relayed using oslo.policy's DocumentedRuleDefault object [1].<br>
>><br>
>> I think the plurality of the resource should default to what makes sense for the operation being carried out (e.g., list:foobars, create:foobar).<br>
>><br>
>> I don't mind the first one because it's clear about what the delimiter is and it doesn't look weird when projects have something like:<br>
>><br>
>> <service-type>:<action>:<subaction>:<resource><br>
>><br>
>> If folks are ok with this, I can start working on some documentation that explains the motivation for this. Afterward, we can figure out how we want to track this work.<br>
>><br>
>> What color do you want the shed to be?<br>
>><br>
>> [0] <a href="https://service-types.openstack.org/service-types.json" rel="noreferrer">https://service-types.openstack.org/service-types.json</a><br>
>> [1] <a href="https://docs.openstack.org/oslo.policy/latest/reference/api/oslo_policy.policy.html#default-rule" rel="noreferrer">https://docs.openstack.org/oslo.policy/latest/reference/api/oslo_policy.policy.html#default-rule</a><br>
>><br>
>> On Fri, Sep 21, 2018 at 9:13 AM Lance Bragstad <<a href="mailto:lbragstad@gmail.com">lbragstad@gmail.com</a>> wrote:<br>
>>><br>
>>><br>
>>> On Fri, Sep 21, 2018 at 2:10 AM Ghanshyam Mann <<a href="mailto:gmann@ghanshyammann.com">gmann@ghanshyammann.com</a>> wrote:<br>
>>>><br>
>>>> ---- On Thu, 20 Sep 2018 18:43:00 +0900 John Garbutt <<a href="mailto:john@johngarbutt.com">john@johngarbutt.com</a>> wrote ----<br>
>>>> > tl;dr+1 consistent names<br>
>>>> > I would make the names mirror the API... because the Operator setting them knows the API, not the codeIgnore the crazy names in Nova, I certainly hate them<br>
>>>><br>
>>>> Big +1 on consistent naming which will help operator as well as developer to maintain those.<br>
>>>><br>
>>>> ><br>
>>>> > Lance Bragstad <<a href="mailto:lbragstad@gmail.com">lbragstad@gmail.com</a>> wrote:<br>
>>>> > > I'm curious if anyone has context on the "os-" part of the format?<br>
>>>> ><br>
>>>> > My memory of the Nova policy mess...* Nova's policy rules traditionally followed the patterns of the code<br>
>>>> > ** Yes, horrible, but it happened.* The code used to have the OpenStack API and the EC2 API, hence the "os"* API used to expand with extensions, so the policy name is often based on extensions** note most of the extension code has now gone, including lots of related policies* Policy in code was focused on getting us to a place where we could rename policy** Whoop whoop by the way, it feels like we are really close to something sensible now!<br>
>>>> > Lance Bragstad <<a href="mailto:lbragstad@gmail.com">lbragstad@gmail.com</a>> wrote:<br>
>>>> > Thoughts on using create, list, update, and delete as opposed to post, get, put, patch, and delete in the naming convention?<br>
>>>> > I could go either way as I think about "list servers" in the API.But my preference is for the URL stub and POST, GET, etc.<br>
>>>> > On Sun, Sep 16, 2018 at 9:47 PM Lance Bragstad <<a href="mailto:lbragstad@gmail.com">lbragstad@gmail.com</a>> wrote:If we consider dropping "os", should we entertain dropping "api", too? Do we have a good reason to keep "api"?I wouldn't be opposed to simple service types (e.g "compute" or "loadbalancer").<br>
>>>> > +1The API is known as "compute" in api-ref, so the policy should be for "compute", etc.<br>
>>>><br>
>>>> Agree on mapping the policy name with api-ref as much as possible. Other than policy name having 'os-', we have 'os-' in resource name also in nova API url like /os-agents, /os-aggregates etc (almost every resource except servers , flavors). As we cannot get rid of those from API url, we need to keep the same in policy naming too? or we can have policy name like compute:agents:create/post but that mismatch from api-ref where agents resource url is os-agents.<br>
>>><br>
>>><br>
>>> Good question. I think this depends on how the service does policy enforcement.<br>
>>><br>
>>> I know we did something like this in keystone, which required policy names and method names to be the same:<br>
>>><br>
>>> "identity:list_users": "..."<br>
>>><br>
>>> Because the initial implementation of policy enforcement used a decorator like this:<br>
>>><br>
>>> from keystone import controller<br>
>>><br>
>>> @controller.protected<br>
>>> def list_users(self):<br>
>>> ...<br>
>>><br>
>>> Having the policy name the same as the method name made it easier for the decorator implementation to resolve the policy needed to protect the API because it just looked at the name of the wrapped method. The advantage was that it was easy to implement new APIs because you only needed to add a policy, implement the method, and make sure you decorate the implementation.<br>
>>><br>
>>> While this worked, we are moving away from it entirely. The decorator implementation was ridiculously complicated. Only a handful of keystone developers understood it. With the addition of system-scope, it would have only become more convoluted. It also enables a much more copy-paste pattern (e.g., so long as I wrap my method with this decorator implementation, things should work right?). Instead, we're calling enforcement within the controller implementation to ensure things are easier to understand. It requires developers to be cognizant of how different token types affect the resources within an API. That said, coupling the policy name to the method name is no longer a requirement for keystone.<br>
>>><br>
>>> Hopefully, that helps explain why we needed them to match.<br>
>>><br>
>>>><br>
>>>><br>
>>>> Also we have action API (i know from nova not sure from other services) like POST /servers/{server_id}/action {addSecurityGroup} and their current policy name is all inconsistent. few have policy name including their resource name like "os_compute_api:os-flavor-access:add_tenant_access", few has 'action' in policy name like "os_compute_api:os-admin-actions:reset_state" and few has direct action name like "os_compute_api:os-console-output"<br>
>>><br>
>>><br>
>>> Since the actions API relies on the request body and uses a single HTTP method, does it make sense to have the HTTP method in the policy name? It feels redundant, and we might be able to establish a convention that's more meaningful for things like action APIs. It looks like cinder has a similar pattern [0].<br>
>>><br>
>>> [0] <a href="https://developer.openstack.org/api-ref/block-storage/v3/index.html#volume-actions-volumes-action" rel="noreferrer">https://developer.openstack.org/api-ref/block-storage/v3/index.html#volume-actions-volumes-action</a><br>
>>><br>
>>>><br>
>>>><br>
>>>> May be we can make them consistent with <service-type>:<resource>:<action_with_snake_case> or any better opinion.<br>
>>>><br>
>>>> > From: Lance Bragstad <<a href="mailto:lbragstad@gmail.com">lbragstad@gmail.com</a>>> The topic of having consistent policy names has popped up a few times this week.<br>
>>>> ><br>
>>>> > I would love to have this nailed down before we go through all the policy rules again. In my head I hope in Nova we can go through each policy rule and do the following:<br>
>>>> > * move to new consistent policy name, deprecate existing name* hardcode scope check to project, system or user** (user, yes... keypairs, yuck, but its how they work)** deprecate in rule scope checks, which are largely bogus in Nova anyway* make read/write/admin distinction** therefore adding the "noop" role, amount other things<br>
>>>><br>
>>>> + policy granularity.<br>
>>>><br>
>>>> It is good idea to make the policy improvement all together and for all rules as you mentioned. But my worries is how much load it will be on operator side to migrate all policy rules at same time? What will be the deprecation period etc which i think we can discuss on proposed spec - <a href="https://review.openstack.org/#/c/547850" rel="noreferrer">https://review.openstack.org/#/c/547850</a><br>
>>><br>
>>><br>
>>> Yeah, that's another valid concern. I know at least one operator has weighed in already. I'm curious if operators have specific input here.<br>
>>><br>
>>> It ultimately depends on if they override existing policies or not. If a deployment doesn't have any overrides, it should be a relatively simple change for operators to consume.<br>
>>><br>
>>>><br>
>>>><br>
>>>><br>
>>>> -gmann<br>
>>>><br>
>>>> > Thanks,John __________________________________________________________________________<br>
>>>> > OpenStack Development Mailing List (not for usage questions)<br>
>>>> > Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
>>>> > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
>>>> ><br>
>>>><br>
>>>><br>
>>>><br>
>>>> __________________________________________________________________________<br>
>>>> OpenStack Development Mailing List (not for usage questions)<br>
>>>> Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
>>>> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
>><br>
>> __________________________________________________________________________<br>
>> OpenStack Development Mailing List (not for usage questions)<br>
>> Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
>> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
><br>
> __________________________________________________________________________<br>
> OpenStack Development Mailing List (not for usage questions)<br>
> Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</blockquote></div></div></div></div>