RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-24 20:13:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

doc: extra \ in CEPH\_AUTH\_UNKNOWN

Browse Source 
Suppress all \

fixes: #11097

Signed-off-by: DEHU Robin <robindehu@gmail.com>
This commit is contained in:
dehu robin 2015-03-13 12:53:57 +01:00
parent 6e56438b01
commit 38bc2986d8
1 changed files with 51 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

102
doc/dev/cephx_protocol.rst
View File

@ -70,7 +70,7 @@ Getting Started With Authorization
When the client first needs to get service, it contacts the monitor. At the moment, it has
no tickets. Therefore, it uses the "unknown" protocol to talk to the monitor. This protocol
is specified as ``CEPH\_AUTH\_UNKNOWN``. The monitor also takes on the authentication server
is specified as ``CEPH_AUTH_UNKNOWN``. The monitor also takes on the authentication server
role, A. The remainder of the communications will use the cephx protocol (most of whose code
will be found in files in ``auth/cephx``). This protocol is responsible for creating and
communicating the tickets spoken of above.
@ -90,7 +90,7 @@ Phase I:
The client is set up to know that it needs certain things, using a variable called ``need``,
which is part of the ``AuthClientHandler`` class, which the ``CephxClientHandler`` inherits
from. At this point, one thing that's encoded in the ``need`` variable is
``CEPH\_ENTITY\_TYPE\_AUTH``, indicating that we need to start the authentication protocol
``CEPH_ENTITY_TYPE_AUTH``, indicating that we need to start the authentication protocol
from scratch. Since we're always talking to the same authorization server, if we've gone
through this step of the protocol before (and the resulting ticket/session hasn't timed out),
we can skip this step and just ask for client tickets. But it must be done initially, and
@ -106,9 +106,9 @@ in the ``need`` flag as necessary. Then we call ``ticket.get_handler()``. This
authorization) in the ticket map, creates a ticket handler object for it, and puts the
handler into the right place in the map. Then we hit specialized code to deal with individual
cases. The case here is when we still need to authenticate to A (the
``if (need & CEPH\_ENTITY\_TYPE\_AUTH)`` branch).
``if (need & CEPH_ENTITY_TYPE_AUTH)`` branch).
We now create a message of type ``CEPHX\_GET\_AUTH\_SESSION\_KEY``. We need to authenticate
We now create a message of type ``CEPH_AUTH_UNKNOWN``. We need to authenticate
this message with C's secret key, so we fetch that from the local key repository. (It's
called a key server in the code, but it's not really a separate machine or processing entity.
It's more like the place where locally used keys are kept.) We create a
@ -124,12 +124,12 @@ challenges, gets put into the message. Then we return from this function, and t
message is sent.
We now switch over to the authenticator side, A. The server receives the message that was
sent, of type ``CEPHX\_GET\_AUTH\_SESSION\_KEY``. The message gets handled in ``prep_auth()``,
sent, of type ``CEPH_AUTH_UNKNOWN``. The message gets handled in ``prep_auth()``,
in ``mon/AuthMonitor.cc``, which calls ``handle_request()`` is ``CephxServiceHandler.cc`` to
do most of the work. This routine, also, handles multiple cases.
The control flow is determined by the ``request_type`` in the ``cephx_header`` associated
with the message. Our case here is ``CEPHX\_GET\_AUTH\_SESSION\_KEY``. We need the
with the message. Our case here is ``CEPH_AUTH_UNKNOWN``. We need the
secret key A shares with C, so we call ``get_secret()`` from out local key repository to get
it. We should have set up a server challenge already with this client, so we make sure
we really do have one. (This variable is specific to a ``CephxServiceHandler``, so there
@ -153,32 +153,32 @@ If the attempt to decode his old ticket fails (most probably because he didn't h
the name of C, the global ID provided in the method call (unless there was an old ticket), and
his ``auid``, obtained from the ``eauth`` structure obtained above. We need a new session key
to help the client communicate securely with us, not using his permanent key. We set the
service ID to ``CEPH\_ENTITY\_TYPE\_AUTH``, which will tell the client C what to do with the
service ID to ``CEPH_ENTITY_TYPE_AUTH``, which will tell the client C what to do with the
message we send it. We build a cephx response header and call
``cephx\_build\_service\_ticket\_reply()``.
``cephx_build_service_ticket_reply()``.
``cephx\_build\_service\_ticket\_reply()`` is in ``auth/cephx/CephxProtocol.cc``. This
``cephx_build_service_ticket_reply()`` is in ``auth/cephx/CephxProtocol.cc``. This
routine will build up the response message. Much of it copies data from its parameters to
a message structure. Part of that information (the session key and the validity period)
gets encrypted with C's permanent key. If the ``should\_encrypt\_ticket`` flag is set,
gets encrypted with C's permanent key. If the ``should_encrypt_ticket`` flag is set,
encrypt it using the old ticket's key. Otherwise, there was no old ticket key, so the
new ticket is not encrypted. (It is, of course, already encrypted with A's permanent key.)
Presumably the point of this second encryption is to expose less material encrypted with
permanent keys.
Then we call the key server's ``get\_service\_caps()`` routine on the entity name, with a
flag ``CEPH\_ENTITY\_TYPE\_MON``, and capabilities, which will be filled in by this routine.
Then we call the key server's ``get_service_caps()`` routine on the entity name, with a
flag ``CEPH_ENTITY_TYPE_MON``, and capabilities, which will be filled in by this routine.
The use of that constant flag means we're going to get the client's caps for A, not for some
other data server. The ticket here is to access the authorizer A, not the service S. The
result of this call is that the caps variable (a parameter to the routine we're in) is
filled in with the monitor capabilities that will allow C to access A's authorization services.
``handle\_request()`` itself does not send the response message. It builds up the
``result\_bl``, which basically holds that message's contents, and the capabilities structure,
but it doesn't send the message. We go back to ``prep\_auth()``, in ``mon/AuthMonitor.cc``,
``handle_request()`` itself does not send the response message. It builds up the
``result_bl``, which basically holds that message's contents, and the capabilities structure,
but it doesn't send the message. We go back to ``prep_auth()``, in ``mon/AuthMonitor.cc``,
for that. This routine does some fiddling around with the caps structure that just got
filled in. There's a global ID that comes up as a result of this fiddling that is put into
the reply message. The reply message is built here (mostly from the ``response\_bl`` buffer)
the reply message. The reply message is built here (mostly from the ``response_bl`` buffer)
and sent off.
This completes Phase I of the protocol. At this point, C has authenticated himself to A, and A has generated a new session key and ticket allowing C to obtain server tickets from A.
@ -190,16 +190,16 @@ This phase starts when C receives the message from A containing a new ticket and
The goal of this phase is to provide A with a session key and ticket allowing him to
communicate with S.
The message A sent to C is dispatched to ``build\_request()`` in ``CephxClientHandler.cc``,
The message A sent to C is dispatched to ``build_request()`` in ``CephxClientHandler.cc``,
the same routine that was used early in Phase I to build the first message in the protocol.
This time, when ``validate\_tickets()`` is called, the ``need`` variable will not contain
``CEPH\_ENTITY\_TYPE\_AUTH``, so a different branch through the bulk of the routine will be
This time, when ``validate_tickets()`` is called, the ``need`` variable will not contain
``CEPH_ENTITY_TYPE_AUTH``, so a different branch through the bulk of the routine will be
used. This is the branch indicated by ``if (need)``. We have a ticket for the authorizer,
but we still need service tickets.
We must send another message to A to obtain the tickets (and session key) for the server
S. We set the ``request\_type`` of the message to ``CEPHX\_GET\_PRINCIPAL\_SESSION\_KEY`` and
call ``ticket\_handler.build\_authorizer()`` to obtain an authorizer. This routine is in
S. We set the ``request_type`` of the message to ``CEPHX_GET_PRINCIPAL_SESSION_KEY`` and
call ``ticket_handler.build_authorizer()`` to obtain an authorizer. This routine is in
``CephxProtocol.cc``. We set the key for this authorizer to be the session key we just got
from A,and create a new nonce. We put the global ID, the service ID, and the ticket into a
message buffer that is part of the authorizer. Then we create a new ``CephXAuthorize``
@ -207,20 +207,20 @@ structure. The nonce we just created goes there. We encrypt this ``CephXAuthor
structure with the current session key and stuff it into the authorizer's buffer. We
return the authorizer.
Back in ``build\_request()``, we take the part of the authorizer that was just built (its
Back in ``build_request()``, we take the part of the authorizer that was just built (its
buffer, not the session key or anything else) and shove it into the buffer we're creating
for the message that will go to A. Then we delete the authorizer. We put the requirements
for what we want in ``req.keys``, and we put ``req`` into the buffer. Then we return, and
the message gets sent.
The authorizer A receives this message which is of type ``CEPHX\_GET\_PRINCIPAL\_SESSION\_KEY``.
The authorizer A receives this message which is of type ``CEPHX_GET_PRINCIPAL_SESSION_KEY``.
The message gets handled in ``prep_auth()``, in ``mon/AuthMonitor.cc``, which again calls
``handle\_request()`` in ``CephxServiceHandler.cc`` to do most of the work.
``handle_request()`` in ``CephxServiceHandler.cc`` to do most of the work.
In this case, ``handle\_request()`` will take the ``CEPHX\_GET\_PRINCIPAL\_SESSION\_KEY`` case.
It will call ``cephx\_verify\_authorizer()`` in ``CephxProtocol.cc``. Here, we will grab
In this case, ``handle_request()`` will take the ``CEPHX_GET_PRINCIPAL_SESSION_KEY`` case.
It will call ``cephx_verify_authorizer()`` in ``CephxProtocol.cc``. Here, we will grab
a bunch of data out of the input buffer, including the global and service IDs and the ticket
for A. The ticket contains a ``secret\_id``, indicating which key is being used for it.
for A. The ticket contains a ``secret_id``, indicating which key is being used for it.
If the secret ID pulled out of the ticket was -1, the ticket does not specify which secret
key A should use. In this case, A should use the key for the specific entity that C wants
to contact, rather than a rotating key shared by all server entities of the same type.
@ -236,8 +236,8 @@ this message. Use that session key to decrypt the rest of the message.
Create a ``CephXAuthorizeReply`` to hold our reply. Extract the nonce (which was in the stuff
we just decrypted), add 1 to it, and put the result in the reply. Encrypt the reply and
put it in the buffer provided in the call to ``cephx\_verify\_authorizer()`` and return
to ``handle\`_request()``. This will be used to prove to C that A (rather than an attacker)
put it in the buffer provided in the call to ``cephx_verify_authorizer()`` and return
to ``handle`_request()``. This will be used to prove to C that A (rather than an attacker)
created this response.
Having verified that the message is valid and from C, now we need to build him a ticket for S.
@ -245,43 +245,43 @@ We need to know what S he wants to communicate with and what services he wants.
ticket request that describes those things out of his message. Now run through the ticket
request to see what he wanted. (He could potentially be asking for multiple different
services in the same request, but we will assume it's just one, for this discussion.) Once we
know which service ID he's after, call ``build\_session\_auth\_info()``.
know which service ID he's after, call ``build_session_auth_info()``.
``build\_session\_auth\_info()`` is in ``CephxKeyServer.cc``. It checks to see if the
secret for the ``service\_ID`` of S is available and puts it into the subfield of one of
the parameters, and calls the similarly named ``\_build\_session\_auth\_info()``, located in
the same file. This routine loads up the new ``auth\_info`` structure with the
``build_session_auth_info()`` is in ``CephxKeyServer.cc``. It checks to see if the
secret for the ``service_ID`` of S is available and puts it into the subfield of one of
the parameters, and calls the similarly named ``_build_session_auth_info()``, located in
the same file. This routine loads up the new ``auth_info`` structure with the
ID of S, a ticket, and some timestamps for that ticket. It generates a new session key
and puts it in the structure. It then calls ``get\_caps()`` to fill in the
``info.ticket`` caps field. ``get\_caps()`` is also in ``CephxKeyServer.cc``. It fills the
``caps\_info`` structure it is provided with caps for S allowed to C.
and puts it in the structure. It then calls ``get_caps()`` to fill in the
``info.ticket`` caps field. ``get_caps()`` is also in ``CephxKeyServer.cc``. It fills the
``caps_info`` structure it is provided with caps for S allowed to C.
Once ``build\_session\_auth\_info()`` returns, A has a list of the capabilities allowed to
Once ``build_session_auth_info()`` returns, A has a list of the capabilities allowed to
C for S. We put a validity period based on the current TTL for this context into the info
structure, and put it into the ``info\_vec`` structure we are preparing in response to the
structure, and put it into the ``info_vec`` structure we are preparing in response to the
message.
Now call ``build\_cephx\_response\_header()``, also in ``CephxServiceHandler.cc``. Fill in
the ``request\_type``, which is ``CEPHX\_GET\_PRINCIPAL\_SESSION\_KEY``, a status of 0,
Now call ``build_cephx_response_header()``, also in ``CephxServiceHandler.cc``. Fill in
the ``request_type``, which is ``CEPHX_GET_PRINCIPAL_SESSION_KEY``, a status of 0,
and the result buffer.
Now call ``cephx\_build\_service\_ticket\_reply()``, which is in ``CephxProtocol.cc``. The
Now call ``cephx_build_service_ticket_reply()``, which is in ``CephxProtocol.cc``. The
same routine was used towards the end of A's handling of its response in phase I. Here,
the session key (now a session key to talk to S, not A) and the validity period for that
key will be encrypted with the existing session key shared between C and A.
The ``should\_encrypt\_ticket`` parameter is false here, and no key is provided for that
The ``should_encrypt_ticket`` parameter is false here, and no key is provided for that
encryption. The ticket in question, destined for S once C sends it there, is already
encrypted with S's secret. So, essentially, this routine will put ID information,
the encrypted session key, and the ticket allowing C to talk to S into the buffer to
be sent to C.
After this routine returns, we exit from ``handle\_request()``, going back to ``prep\_auth()``
After this routine returns, we exit from ``handle_request()``, going back to ``prep_auth()``
and ultimately to the underlying message send code.
The client receives this message. The nonce is checked as the message passes through
``Pipe::connect()``, which is in ``msg/SimpleMessager.cc``. In a lengthy ``while(1)`` loop in
the middle of this routine, it gets an authorizer. If the get was successful, eventually
it will call ``verify\_reply()``, which checks the nonce. ``connect()`` never explicitly
it will call ``verify_reply()``, which checks the nonce. ``connect()`` never explicitly
checks to see if it got an authorizer, which would suggest that failure to provide an
authorizer would allow an attacker to skip checking of the nonce. However, in many places,
if there is no authorizer, important connection fields will get set to zero, which will
@ -289,16 +289,16 @@ ultimately cause the connection to fail to provide data. It would be worth test
it looks like failure to provide an authorizer, which contains the nonce, would not be helpful
to an attacker.
The message eventually makes its way through to ``handle\_response()``, in
``CephxClientHandler.cc``. In this routine, we call ``get\_handler()`` to get a ticket
The message eventually makes its way through to ``handle_response()``, in
``CephxClientHandler.cc``. In this routine, we call ``get_handler()`` to get a ticket
handler to hold the ticket we have just received. This routine is embedded in the definition
for a ``CephXTicketManager`` structure. It takes a type (``CEPH\_ENTITY\_TYPE\_AUTH``, in
this case) and looks through the ``tickets\_map`` to find that type. There should be one, and
for a ``CephXTicketManager`` structure. It takes a type (``CEPH_ENTITY_TYPE_AUTH``, in
this case) and looks through the ``tickets_map`` to find that type. There should be one, and
it should have the session key of the session between C and A in its entry. This key will
be used to decrypt the information provided by A, particularly the new session key allowing
C to talk to S.
We then call ``verify\_service\_ticket\_reply()``, in ``CephxProtocol.cc``. This routine
We then call ``verify_service_ticket_reply()``, in ``CephxProtocol.cc``. This routine
needs to determine if the ticket is OK and also obtain the session key associated with this
ticket. It decrypts the encrypted portion of the message buffer, using the session key
shared with A. This ticket was not encrypted (well, not twice - tickets are always encrypted,
@ -309,7 +309,7 @@ The stuff we decrypted with the session key shared between C and A included the
key. That's our current session key for this ticket, so set it. Check validity and
set the expiration times. Now return true, if we got this far.
Back in ``handle\_response()``, we now call ``validate\_tickets()`` to adjust what we think
Back in ``handle_response()``, we now call ``validate_tickets()`` to adjust what we think
we need, since we now have a ticket we didn't have before. If we've taken care of
everything we need, we'll return 0.