mirror of
https://github.com/nspcc-dev/neofs-api.git
synced 2026-03-01 04:28:54 +00:00
Bearer token in request classification #82
Labels
No labels
I0
I1
I2
I3
I4
S0
S1
S2
S3
S3
S4
U2
U3
U4
bug
discussion
discussion
documentation
enhancement
enhancement
enhancement
feature
good first issue
question
task
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
nspcc-dev/neofs-api#82
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @alexvanin on GitHub (Sep 5, 2022).
Description
NeoFS service applications are middleware between users and native NeoFS protocol. All NeoFS requests are signed with the key. Service applications do not have access to user's keys, so they use configured keys (middleware keys) to sign requests.
In general, public key is used to identify request sender (owner) to apply access control rules. NeoFS supports token mechanism to allow access to the requests signed by middleware keys: bearer tokens and session tokens.
In the request handler, storage node fetches public key and compares it with container owner, alphabet or container nodes. Here is the simplified scheme of fetch function from v0.32.0 release: RequestOwner().
There are two cases where token scheme is not applicable.
1. Service application requires access to a private container on behalf of the user
WebUI + REST gateway should provide UX similar to CLI. To access private container, REST Gateway's request should be considered as user's request. The only way to impersonate user is to use session token signed by user's key.
However, session token can't be reused for different requests: object context requires verb and object address.
Session token is a nice solution for peer-to-peer communication, when application uses private key to resign session token for every request. It is not applicable for service application, because token usually is signed once in a while:
2. User grants access to the container, which is accessed through gateway
S3 gateway user wants to provide object access for specific user. To do that, container's extended ACL is modified to grant access for user's public key.
When S3 gateway sends request on behalf of granted user, access control record can't be matched, because S3 signed request with middleware key.
Session token is a nice solution for peer-to-peer communication, when client uses private key to resign session token for evert request.
It is not applicable for service application, because token usually is signed once for a while:
- signing every NeoFS request might be very bad UX (e.g. with Wallet Connect),
- session token is not transitive, so the client has to create session with unknown NeoFS Node, instead of the known gateway.
Service applications usually work with bearer tokens, while session is established between gateway and the node.
Private container denies bearer tokens. So service application have to create public containers with restricted extended ACL to imitate private container.
Proposal
To solve this issue, modify bearer token processing behaviour. Provide
impersonateflag in the token body.On
false(default), behaviour is the same.On
true:When token allows impersonating, there is no meaning in replacing container extended ACL table with token extended ACL table. Node should behave as the request was originally signed by token owner. Then S3 Gateway users will be able to provide object access to specific users.
Tree service
Changes above are applied for object service. However, tree service is also affected by this change.
On
false(default), behaviour is the same.On
true:(3) looks very important here. Now S3 Gateway accesses user bucket by attaching bearer token. To access other user buckets it does not attach bearer token, because token issuer is always going to be different from container owner.
To grant access to the container, S3 gateway will specify public key of the user in container extended ACL. To match this rule, S3 gateway has to attach bearer token with impersonate flag. So token issuer check should gone in this case. It makes sense, because impersonate flag can be treated as if "there is no bearer tokens at all, request is signed on behalf of token issuer".
🟢 It is backward compatible
🔴 Classification becomes less clear
@alexvanin commented on GitHub (Oct 26, 2022):
We did some implementation prototyping for service checks:
neofs-api-go/poc/impersonate
neofs-sdk-go/poc/impersonate
neofs-node/poc/impersonate
@KirillovDenis commented on GitHub (Oct 26, 2022):
neofs-rest-gw
neofs-s3-gw
@alexvanin commented on GitHub (Oct 27, 2022):
Let's check scheme with putting object to node outside of the container.
@cthulhu-rider commented on GitHub (Oct 27, 2022):
If in scheme above REST Gateway forms object on its side (lets imagine this 🪄) and stores it in the network, the bearer token will allow the gateway to act on user's behalf only within this particular request. Subsequently, this power of attorney will be lost, and there will not be able to justify why the object was saved.
On the contrary, session token is sewn into the object. Overall, IMO session mechanism fits better since we need to transfer the power of attorney.
@KirillovDenis commented on GitHub (Nov 10, 2022):
The
impersonateflag doesn't affect the integrity problem (object owner doesn't match to session token issuer in general) in the following case:wallet1.jsonwallet2.jsonwallet1.json, session token issuer will be fromwallet2.json)@alexvanin commented on GitHub (Nov 17, 2022):
Probably we still have bearer token rules support in auth procedure. It is relevant when all credentials are issued by the same key.
@roman-khimov commented on GitHub (Mar 17, 2024):
I'm not sure what's so good about bearer to extend it this way effectively completely changing its meaning. Likely session tokens can be extended or even changed to:
@roman-khimov commented on GitHub (Mar 18, 2024):
The only positive side is one token instead of two tokens (with possibility to deprecate session ones). But currently session and bearer can be combined for a request, unlike in this proposal (where impersonation disables eacl).
@cthulhu-rider commented on GitHub (Apr 10, 2024):
Union
currently tokens implement completely independent mechanisms:
they currently converge in:
in total, tokens can be combined only structurally, not conceptually. Moreover, even being in the same data structure, their payload would be mutex
Proposal
im leaving an example for now. A more detailed format will be offered later
Public key (JWK)
Proxy token payload (JWT)
Proxy token (JWS Compact serialization)
it may also be more flex to unite
trustedSubjectsandtrustedActionsinto one member and have list of them asproxyList@roman-khimov commented on GitHub (Oct 10, 2025):
Summarizing our discussions around this. We don't need to modify bearer tokens, they're good for the problem they solve allowing container owner to create some access token for third party use. The problem we really have in all of the app-gateway-sn chain scenarios is delegation, power of attorney and it should be solved with updated session token mechanism. Critical things for it:
Preferably:
To be decided:
Optional:
@roman-khimov commented on GitHub (Oct 15, 2025):
A note on JSON. While currently it's only relevant for REST, there are other possibilities as well if we're to follow JSON route, like https://github.com/neo-project/neo/issues/2866. It is possible to parse JSON in a smart contract (unlike protobuf) and this can create possibilities for contract-level action check and traceability.
@roman-khimov commented on GitHub (Dec 6, 2025):
Fixed in #350.