CertificatesV1beta1 was removed in Kube 1.22, so the tests cannot
blindly rely on it anymore. Use CertificatesV1 whenever the server
reports that is available, and otherwise use the old
CertificatesV1beta1.
Note that CertificatesV1 was introduced in Kube 1.19.
Changes made to both components:
1. Logs are always flushed on process exit
2. Informer cache sync can no longer hang process start up forever
Changes made to concierge:
1. Add pre-shutdown hook that waits for controllers to exit cleanly
2. Informer caches are synced in post-start hook
Changes made to supervisor:
1. Add shutdown code that waits for controllers to exit cleanly
2. Add shutdown code that waits for active connections to become idle
Waiting for controllers to exit cleanly is critical as this allows
the leader election logic to release the lock on exit. This reduces
the time needed for the next leader to be elected.
Signed-off-by: Monis Khan <mok@vmware.com>
Instead of blindly waiting long enough for a disruptive change to
have been observed by the old leader and followers, we instead rely
on the approximation that checkOnlyLeaderCanWrite provides - i.e.
only a single actor believes they are the leader. This does not
account for clients that were in the followers list before and after
the disruptive change, but it serves as a reasonable approximation.
Signed-off-by: Monis Khan <mok@vmware.com>
Those images that are pulled from Dockerhub will cause pull failures
on some test clusters due to Dockerhub rate limiting.
Because we already have some images that we use for testing, and
because those images are already pre-loaded onto our CI clusters
to make the tests faster, use one of those images and always specify
PullIfNotPresent to avoid pulling the image again during the integration
test.
OpenShift has good defaults for these duration fields that we can
use instead of coming up with them ourselves:
e14e06ba8d/pkg/config/leaderelection/leaderelection.go (L87-L109)
Copied here for easy future reference:
// We want to be able to tolerate 60s of kube-apiserver disruption without causing pod restarts.
// We want the graceful lease re-acquisition fairly quick to avoid waits on new deployments and other rollouts.
// We want a single set of guidance for nearly every lease in openshift. If you're special, we'll let you know.
// 1. clock skew tolerance is leaseDuration-renewDeadline == 30s
// 2. kube-apiserver downtime tolerance is == 78s
// lastRetry=floor(renewDeadline/retryPeriod)*retryPeriod == 104
// downtimeTolerance = lastRetry-retryPeriod == 78s
// 3. worst non-graceful lease acquisition is leaseDuration+retryPeriod == 163s
// 4. worst graceful lease acquisition is retryPeriod == 26s
if ret.LeaseDuration.Duration == 0 {
ret.LeaseDuration.Duration = 137 * time.Second
}
if ret.RenewDeadline.Duration == 0 {
// this gives 107/26=4 retries and allows for 137-107=30 seconds of clock skew
// if the kube-apiserver is unavailable for 60s starting just before t=26 (the first renew),
// then we will retry on 26s intervals until t=104 (kube-apiserver came back up at 86), and there will
// be 33 seconds of extra time before the lease is lost.
ret.RenewDeadline.Duration = 107 * time.Second
}
if ret.RetryPeriod.Duration == 0 {
ret.RetryPeriod.Duration = 26 * time.Second
}
Signed-off-by: Monis Khan <mok@vmware.com>
This change fixes a small race condition that occurred when the
current leader failed to renew its lease. Before this change, the
leader would first release the lease via the Kube API and then would
update its in-memory status to reflect that change. Now those
events occur in the reverse (i.e. correct) order.
Signed-off-by: Monis Khan <mok@vmware.com>
Even though a client may hold the leader election lock in the Kube
lease API, that does not mean it has had a chance to update its
internal state to reflect that. Thus we retry the checks in
checkOnlyLeaderCanWrite a few times to allow the client to catch up.
Signed-off-by: Monis Khan <mok@vmware.com>
In the upstream dynamiccertificates package, we rely on two pieces
of code:
1. DynamicServingCertificateController.newTLSContent which calls
- clientCA.CurrentCABundleContent
- servingCert.CurrentCertKeyContent
2. unionCAContent.VerifyOptions which calls
- unionCAContent.CurrentCABundleContent
This results in calls to our tlsServingCertDynamicCertProvider and
impersonationSigningCertProvider. If we Unset these providers, we
subtly break these consumers. At best this results in test slowness
and flakes while we wait for reconcile loops to converge. At worst,
it results in actual errors during runtime. For example, we
previously would Unset the impersonationSigningCertProvider on any
sync loop error (even a transient one caused by a network blip or
a conflict between writes from different replicas of the concierge).
This would cause us to transiently fail to issue new certificates
from the token credential require API. It would also cause us to
transiently fail to authenticate previously issued client certs
(which results in occasional Unauthorized errors in CI).
Signed-off-by: Monis Khan <mok@vmware.com>
- Add `AllowPasswordGrant` boolean field to OIDCIdentityProvider's spec
- The oidc upstream watcher controller copies the value of
`AllowPasswordGrant` into the configuration of the cached provider
- Add password grant to the UpstreamOIDCIdentityProviderI interface
which is implemented by the cached provider instance for use in the
authorization endpoint
- Enhance the IDP discovery endpoint to return the supported "flows"
for each IDP ("cli_password" and/or "browser_authcode")
- Enhance `pinniped get kubeconfig` to help the user choose the desired
flow for the selected IDP, and to write the flow into the resulting
kubeconfg
- Enhance `pinniped login oidc` to have a flow flag to tell it which
client-side flow it should use for auth (CLI-based or browser-based)
- In the Dex config, allow the resource owner password grant, which Dex
implements to also return ID tokens, for use in integration tests
- Enhance the authorize endpoint to perform password grant when
requested by the incoming headers. This commit does not include unit
tests for the enhancements to the authorize endpoint, which will come
in the next commit
- Extract some shared helpers from the callback endpoint to share the
code with the authorize endpoint
- Add new integration tests
At a high level, it switches us to a distroless base container image, but that also includes several related bits:
- Add a writable /tmp but make the rest of our filesystems read-only at runtime.
- Condense our main server binaries into a single pinniped-server binary. This saves a bunch of space in
the image due to duplicated library code. The correct behavior is dispatched based on `os.Args[0]`, and
the `pinniped-server` binary is symlinked to `pinniped-concierge` and `pinniped-supervisor`.
- Strip debug symbols from our binaries. These aren't really useful in a distroless image anyway and all the
normal stuff you'd expect to work, such as stack traces, still does.
- Add a separate `pinniped-concierge-kube-cert-agent` binary with "sleep" and "print" functionality instead of
using builtin /bin/sleep and /bin/cat for the kube-cert-agent. This is split from the main server binary
because the loading/init time of the main server binary was too large for the tiny resource footprint we
established in our kube-cert-agent PodSpec. Using a separate binary eliminates this issue and the extra
binary adds only around 1.5MiB of image size.
- Switch the kube-cert-agent code to use a JSON `{"tls.crt": "<b64 cert>", "tls.key": "<b64 key>"}` format.
This is more robust to unexpected input formatting than the old code, which simply concatenated the files
with some extra newlines and split on whitespace.
- Update integration tests that made now-invalid assumptions about the `pinniped-server` image.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This may be a temporary fix. It switches the manual auth code prompt to use `promptForValue()` instead of `promptForSecret()`. The `promptForSecret()` function no longer supports cancellation (the v0.9.2 behavior) and the method of cancelling in `promptForValue()` is now based on running the blocking read in a background goroutine, which is allowed to block forever or leak (which is not important for our CLI use case).
This means that the authorization code is now visible in the user's terminal, but this is really not a big deal because of PKCE and the limited lifetime of an auth code.
The main goroutine now correctly waits for the "manual prompt" goroutine to clean up, which now includes printing the extra newline that would normally have been entered by the user in the manual flow.
The text of the manual login prompt is updated to be more concise and less scary (don't use the word "fail").
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This test is asynchronously waiting for the controller to do something, and in some of our test environments it will take a bit longer than we'd previously allowed.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
this will hopefully fix some flakes where aws provisioned a host for the
load balancer but the tests weren't able to resolve it.
Signed-off-by: Margo Crawford <margaretc@vmware.com>
It seems like page.ClearCookies() only clears cookies for the current
domain, so there doesn't seem to be a function to clear all browser
cookies. Instead, we'll just start a whole new browser each test.
They start fast enough that it shouldn't be a problem.
Our actual CLI code behaved correctly, but this test made some invalid assumptions about the "upstream" IDP we're testing. It assumed that the upstream didn't support `response_mode=form_post`, but Okta does. This means that when we end up on the localhost callback page, there are no URL query parameters.
Adjusting this regex makes the test pass as expected.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
Using the same fake TTY trick we used to test LDAP login, this new subtest runs through the "manual"/"jump box" login flow. It runs the login with a `--skip-listen` flag set, causing the CLI to skip opening the localhost listener. We can then wait for the login URL to be printed, visit it with the browser and log in, and finally simulate "manually" copying the auth code from the browser and entering it into the waiting CLI prompt.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
For some reason our headless Chrome test setup behaves slightly differently on Linux and macOS hosts. On Linux, the emoji characters are not recognized as valid text, so they are URL encoded. This change updates the test to cope with both cases correctly.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This adds a new login flow that allows manually pasting the authorization code instead of receiving a browser-based callback.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This is a new pacakge internal/oidc/provider/formposthtml containing a number of static files embedded using the relatively recent Go "//go:embed" functionality introduced in Go 1.16 (https://blog.golang.org/go1.16).
The Javascript and CSS files are minifiied and injected to make a single self-contained HTML response. There is a special Content-Security-Policy helper to calculate hash-based script-src and style-src rules.
This new code is covered by a new integration test that exercises the JS/HTML functionality in a real browser outside of the rest of the Supervisor.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This test would occasionally flake for me when running locally. This change moves more of the assertions into the "eventually" loop, so they can temporarily fail as long as they converge on the expected values.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This test did not tolerate this connection failing, which can happen for any number of flaky networking-related reasons. This change moves the connection setup into an "eventually" retry loop so it's allowed to fail temporarily as long as it eventually connects.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
TestSimultaneousLDAPRequestsOnSingleProvider proved to be unreliable
on AKS due to some kind of kubectl port-forward issue, so only
run the LDAP client's integration tests on Kind. They are testing
the integration between the client code and the OpenLDAP test server,
not testing anything about Kubernetes, so running only on Kind should
give us sufficient test coverage.
This fixes some rare test flakes caused by a data race inherent in the way we use `assert.Eventually()` with extra variables for followup assertions. This function is tricky to use correctly because it runs the passed function in a separate goroutine, and you have no guarantee that any shared variables are in a coherent state when the `assert.Eventually()` call returns. Even if you add manual mutexes, it's tricky to get the semantics right. This has been a recurring pain point and the cause of several test flakes.
This change introduces a new `library.RequireEventually()` that works by internally constructing a per-loop `*require.Assertions` and running everything on a single goroutine (using `wait.PollImmediate()`). This makes it very easy to write eventual assertions.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This change updates the impersonator to always authorize every
request instead of relying on the Kuberentes API server to perform
the check on the impersonated request. This protects us from
scenarios where we fail to correctly impersonate the user due to
some bug in our proxy logic. We still rely completely on the API
server to perform admission checks on the impersonated requests.
Signed-off-by: Monis Khan <mok@vmware.com>
This change updates the impersonation proxy code to run as a
distinct service account that only has permission to impersonate
identities. Thus any future vulnerability that causes the
impersonation headers to be dropped will fail closed instead of
escalating to the concierge's default service account which has
significantly more permissions.
Signed-off-by: Monis Khan <mok@vmware.com>
When anonymous authentication is disabled, the impersonation proxy
will no longer authenticate anonymous requests other than calls to
the token credential request API (this API is used to retrieve
credentials and thus must be accessed anonymously).
Signed-off-by: Benjamin A. Petersen <ben@benjaminapetersen.me>
Signed-off-by: Monis Khan <mok@vmware.com>
There was nothing to guarantee that _all_ Supervisor pods would be ready to handle this request. We saw a rare test flake where the LDAPIdentityProvider was marked as ready but one of the Supervisor pods didn't have it loaded yet and returned an HTTP 422 error (`Unprocessable Entity: No upstream providers are configured`).
Signed-off-by: Matt Moyer <moyerm@vmware.com>
The `require.Eventually()` function runs the body of the check in a separate goroutine, so it's not safe to use other `require` assertions as we did here. Our `library.RequireEventuallyWithoutError()` function does not spawn a goroutine, so it's safer to use here.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
The LastUpdateTime is no longer updated on every resync. It only changes if the underlying status has changed, so that it effectively shows when the transition happened.
This change happened in ab750f48aa, but we missed this test. It only fails when it has been more than ten minutes since the CredentialIssuer transitioned into a healthy state, but that can happen in our long-running CI environments.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
This test felt overly complex and some of the cleanup logic wasn't 100% correct (it didn't clean up in all cases).
The new code is essentially the same flow but hopefully easier to read.
Signed-off-by: Matt Moyer <moyerm@vmware.com>
