This removes the Cartesian product in the DB for GrantsForUser in favor
of returning the actual grant scope information and dealing with it in
application code.
(cherry picked from commit 913cd536823d73994acc4f4d833208c1cc05f91b)
This ensures we don't have competing update processes running. When doing the
inline RefreshForSearch, a specific error will be returned so we can hint in
the API that we're in the middle of a refresh.
(cherry picked from commit 54776091e32af2f2f110f2bf10d0617bd5677aba)
This allows getting a list of scope IDs known by the cache via cached
targets and sessions. It is not refreshed from the controller. The name
contains `implicit` in case we ever want e.g. `scopes`.
(cherry picked from commit c0a318c58367d8ec55365447f41cf007eab64ce3)
A query parameter `max_result_set_size` can be specified to set a
per-request limit, with `-1` meaning all values. If there were more
entries than the max size used for the request, an `incomplete` boolean
is set in the response.
(cherry picked from commit 3cdd07889be5e4e6f21d044cb60fad1face4be08)
* Replace `interface{}` with `any`
* For options that are bools, don't always set true; set to the actual
value passed in (some options already behaved this way, but they all
should)
This adds grant_scope_ids in the output of role listing. The lookup for grant scope IDs was changed to allow taking in a set of roles instead of just a single role, and the results are collated afterwards.
---------
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
When wrapper.Wrap was put into place, the argument going into it was no
longer part of a closure but was instead evaluated at function call
time. As a result this would create 197 extra goroutines and
channels on startup - one for each command using wrapper.Wrap. Among
other things this made stack traces much more difficult to parse.
At some point we migrated to grpc ConnectParams to control how often
retries were happening. One value that we didn't set was the
MinConnectTimeout. It turns out that if you don't set this, it won't use
the default; instead if will actually use a value of zero, which gRPC
internally ups to something that seems to be around 25ms to meet a
minimum threshold. This is not documented in the function comments.
This commit simply sets the value of MinConnectTimeout to the unexported
gRPC default.
This does not currently remove the field from the message on the API but
since it will now always show `pki` it's meaningless and causing
confusion for some users.
* Add new GrantsForUser test
This exercises a wide array of granting possibilities and ensures that
we see all of the desired grants and none that we don't expect.
* Address review feedback
This allows arbitrary programs to have an easy mechanism to make
connections through Boundary using Go. Internal cluster tests and test
helpers have been migrated to use this package, as well as the `connect`
command.
---------
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
Some platforms use `@` for a Unix socket address, and some leave it
empty. In the empty case this causes the rate limiting code to throw a
500 error. This commit standardizes on using `@` for incoming
connections over a Unix socket.
This has been superceded by the nodeenrollment-based KMS worker auth
mechanism. This commit only removes the ability to connect with this
mechanism; some upgrade logic will remain intact for a bit and
can be removed later.
* Add generic read/update/delete commands
This adds three generic commands that use the resource prefix to
determine the subcommand to run and pass through args over to that
subcommand. Since this occurs very early on and to avoid loading
essentially every internal resource package, I moved subtype definitions
to globals.
This replaces #3967
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
* De-duplicate Subtype code (#4031)
Co-authored-by: Timothy Messier <tim.messier@gmail.com>
---------
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
Co-authored-by: Timothy Messier <tim.messier@gmail.com>
* If passing a private key without specifying it in URL syntax, give a
proper error message (previously we were only making sure it was a
scheme we knew how to handle, but we need to also be more informative
if it's not a URL at all)
* If the handshake from the proxy fails, don't hang, just exit
* Update some error messages to use capitalization in line with our
standard CLI error format
* Minor test fixes
The auth rotation test fix allows us to keep checking if we did not yet
see the expected bytes in case the call to lookup succeeds but we are
late in rotating due to parallelism.
The api credentials test fix allows us to check if at _least_ one
rotation has happened. That way if two rotations happen before we check
again due to parallelism, this is deemed a successful result.
* Add timeout to auth rotation test
* Add generic read/update/delete commands
This adds three generic commands that use the resource prefix to
determine the subcommand to run and pass through args over to that
subcommand. Since this occurs very early on and to avoid loading
essentially every internal resource package, I moved subtype definitions
to globals.
* Update main.go
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
* Update changelog and remove another deprecated command
---------
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
In #2076 the refactor changed the position of a parenthesis, causing
the dev key to truncate the base64 representation instead of specifying
which bytes to base64 encode.
We don't need this slicing anymore anyways since we now do individual
key generation calls so just remove it.
In preparation for putting proxy code in the api package this creates some necessary fields in the target output and it pre-moves credential parsing and proxy handshake code to their new places. This will make the diff for the actual proxy code move quite a bit smaller.
Co-authored-by: Johan Brandhorst-Satzkorn <johan.brandhorst@gmail.com>
With a suitably slow system -- say, if you're waiting on a Windows
Terminal to output text inside a virtual machine -- the expected order
of things can be thrown off. One specific way this can happen is that
the worker can attempt to dial and have its dial fail because of the
slowness causing the dial context to time out. If, because of the
slowness, we also haven't released the log gate yet, then the event with
the failure information will be queued, along with the context that was
used.
Unfortunately, in some of these error cases, the context that was used
was the dial context instead of the system base context. In many other
places in the function it was the system base context so this is just a
mismatch, probably from code written at different times.
Normally this wouldn't be a problem as we'd fall back to the underlying
logger, but when we release the log gate, things happen differently:
it's a synchronous function and on error it causes us to abandon
system startup entirely.
This commit fixes the issue with using the incorrect context. It's an
open question whether we should change the behavior around errors when
replaying queued events, falling back to the underlying logger instead
of erroring.
With the KMS changes this should never have worked at all (because the
different databases means the credentials conveyed after the KMS-based
authentication would never be valid). However, it was sometimes passing
because of a race condition: the worker could be found in the database
before the new credentials were returned to the worker, causing the
worker to start at the KMS step with the second controller.
This changes the test to use the same database for two controllers, but
to only ever have one running at a time, and uses a sequence of two
changed (or unchanged, depending) status updates from the worker to
ensure that the reload was successful.
This avoids a situation where teardown results in an error due to the
worker not accepting the connection (since the credentials are invalid).
Teardown is a best-effort way to avoid sessions that have long lifetimes
ending up having to wait for expiration before cleanup, but if we're
near expiration it doesn't really matter.
Fixes ICU-9633
This allows specifying multiple IDs in grants via an `ids` field, which
will eventually become the normal way to specify IDs. Internally, each
grant with multiple IDs turns into multiple ACL grants each with a
single ID; to make this distinction clearer and allow for better
separation of concerns an AclGrant type has been introduced, so really
multiple IDs in a Grant turn into multiple AclGrants.
The nice thing about doing it this way is that it becomes purely a
parsing concern; no actual validation logic has changed.
For backwards compatibility with grants already stored in roles, we do
not error out if `id` is specified. However, eventually we will simply
not support grants with `id` coming via the API; they will fail API
validation, such that new grants will require the use of the updated
`ids` field.
A follow-on PR will plumb this through the service handlers/API/CLI, but
I decided to do it in separate chunks to make reviewing easier, given
that they are easily separable into the internal and external aspects of
the change.
* Add CLI/API checks around using ID in grants (#3264)
* Plumb through some contexts to remove a lot of deprecated errors (#3268)
This requires specifying `**` as the selector in the proto, which will
match any segments and simply requires that the verb is the last part of
the URL, which it already is.
Most of the changes here are changes or enhancements that are useful for
tests. The crux of the issue was that I had pulled up one var to the
package level so it could be read from tests, but that var was not being
reset between tests. It's not even used in tests anymore, so I put it
back to function-local.
These are deps either maintained by the Boundary team or approved by
members of the team. A few specific notes on some of the non-HC and
non-DB ones:
github.com/fatih/color: fixes some windows issues
github.com/zalando/go-keyring: fixes a windows crash
* Add more logging to worker auth rotation and test flags
This makes it much easier to figure out when and why rotation is
happening and control it for tests/debugging.
* Update auth rotation logic
Because we were looking at validity period of client certs, which may
not have matched the actual rotation interval, it was possible (mostly
in test scenarios) for the rotation to happen multiple times on the
roots before we actually hit the worker rotation interval.
This changes how the interval is calculated by looking at the difference
between not-after values on the certs to understand the rotation
interval being used by the controller. We then use that as the starting
point for calculating a next-rotation time.
This makes it much easier to figure out when and why rotation is
happening and control it for tests/debugging.
Remove cert skew
Cherry-picked from release/0.12.x
Jeff Mitchell
