Occasionally we had seen the TestWatchServers_ACLToken_PermissionDenied be flagged as flaky in circleci. This change should fix that.
Why it fixes it is complicated. The test was failing with a panic when a mocked ACL Resolver was being called more times than expected. I struggled for a while to determine how that could be. This test should call authorize once and only once and the error returned should cause the stream to be terminated and the error returned to the gRPC client. Another oddity was no amount of running this test locally seemed to be able to reproduce the issue. I ran the test hundreds of thousands of time and it always passed.
It turns out that there is nothing wrong with the test. It just so happens that the panic from unexpected invocation of a mocked call happened during the test but was caused by a previous test (specifically the TestWatchServers_StreamLifecycle test)
The stream from the previous test remained open after all the test Cleanup functions were run and it just so happened that when the EventPublisher eventually picked up that the context was cancelled during cleanup, it force closes all subscriptions which causes some loops to be re-entered and the streams to be reauthorized. Its that looping in response to forced subscription closures that causes the mock to eventually panic. All the components, publisher, server, client all operate based on contexts. We cancel all those contexts but there is no syncrhonous way to know when they are stopped.
We could have implemented a syncrhonous stop but in the context of an actual running Consul, context cancellation + async stopping is perfectly fine. What we (Dan and I) eventually thought was that the behavior of grpc streams such as this when a server was shutting down wasn’t super helpful. What we would want is for a client to be able to distinguish between subscription closed because something may have changed requiring re-authentication and subscription closed because the server is shutting down. That way we can send back appropriate error messages to detail that the server is shutting down and not confuse users with potentially needing to resubscribe.
So thats what this PR does. We have introduced a shutting down state to our event subscriptions and the various streaming gRPC services that rely on the event publisher will all just behave correctly and actually stop the stream (not attempt transparent reauthorization) if this particular error is the one we get from the stream. Additionally the error that gets transmitted back through gRPC when this does occur indicates to the consumer that the server is going away. That is more helpful so that a client can then attempt to reconnect to another server.
OSS portion of enterprise PR 1857.
This removes (most) references to the `cache.UpdateEvent` type in the
`proxycfg` package.
As we're going to be direct usage of the agent cache with interfaces that
can be satisfied by alternative server-local datasources, it doesn't make
sense to depend on this type everywhere anymore (particularly on the
`state.ch` channel).
We also plan to extract `proxycfg` out of Consul into a shared library in
the future, which would require removing this dependency.
Aside from a fairly rote find-and-replace, the main change is that the
`cache.Cache` and `health.Client` types now accept a callback function
parameter, rather than a `chan<- cache.UpdateEvents`. This allows us to
do the type conversion without running another goroutine.
- Introduce a new telemetry configurable parameter retry_failed_connection. User can set the value to true to let consul agent continue its start process on failed connection to datadog server. When set to false, agent will stop on failed start. The default behavior is true.
Co-authored-by: Dan Upton <daniel@floppy.co>
Co-authored-by: Evan Culver <eculver@users.noreply.github.com>
Treat each exported service as a "discovery chain" and replicate one
synthetic CheckServiceNode for each chain and remote mesh gateway.
The health will be a flattened generated check of the checks for that
mesh gateway node.
By changing to use WatchCtx we will actually block for changes to the peering list. WatchCh creates a goroutine to collect errors from WatchCtx and returns immediately.
The existing behavior wouldn't result in a tight loop because of the rate limiting in the surrounding function, but it would still lead to more work than is necessary.
Changes include:
- Add diagrams of the operation of different consistency modes
- Note that only stale reads benefit from horizontal scaling
- Increase scannability with headings
- Document consistency mode defaults and how to override for
DNS and HTTP API interfaces
- Document X-Consul-Effective-Consistency response header
Changes to how the version string was handled created small regression with the release of consul 1.12.0 enterprise.
Many tools use the Config:Version field reported by the agent/self resource to determine whether Consul is an enterprise or OSS instance, expect something like 1.12.0+ent for enterprise and simply 1.12.0 for OSS. This was accidentally broken during the runup to 1.12.x
This work fixes the value returned by both the self endpoint in ["Config"]["Version"] and the metrics consul.version field.
Signed-off-by: Mark Anderson <manderson@hashicorp.com>
Add validation to ensure connect native services have a port or socketpath specified on catalog registration.
This was the only missing piece to ensure all mesh services are validated for a port (or socketpath) specification on catalog registration.
* Support vault namespaces in connect CA
Follow on to some missed items from #12655
From an internal ticket "Support standard "Vault namespace in the
path" semantics for Connect Vault CA Provider"
Vault allows the namespace to be specified as a prefix in the path of
a PKI definition, but our usage of the Vault API includes calls that
don't support a namespaced key. In particular the sys.* family of
calls simply appends the key, instead of prefixing the namespace in
front of the path.
Unfortunately it is difficult to reliably parse a path with a
namespace; only vault knows what namespaces are present, and the '/'
separator can be inside a key name, as well as separating path
elements. This is in use in the wild; for example
'dc1/intermediate-key' is a relatively common naming schema.
Instead we add two new fields: RootPKINamespace and
IntermediatePKINamespace, which are the absolute namespace paths
'prefixed' in front of the respective PKI Paths.
Signed-off-by: Mark Anderson <manderson@hashicorp.com>
Introduces two new public gRPC endpoints (`Login` and `Logout`) and
includes refactoring of the equivalent net/rpc endpoints to enable the
majority of logic to be reused (i.e. by extracting the `Binder` and
`TokenWriter` types).
This contains the OSS portions of the following enterprise commits:
- 75fcdbfcfa6af21d7128cb2544829ead0b1df603
- bce14b714151af74a7f0110843d640204082630a
- cc508b70fbf58eda144d9af3d71bd0f483985893
Description
Add x-fowarded-client-cert information on trusted incoming connections.
Envoy provides support forwarding and annotating the
x-forwarded-client-cert header via the forward_client_cert_details
set_current_client_cert_details filter fields. It would be helpful for
consul to support this directly in its config. The escape hatches are
a bit cumbersome for this purpose.
This has been implemented on incoming connections to envoy. Outgoing
(from the local service through the sidecar) will not have a
certificate, and so are left alone.
A service on an incoming connection will now get headers something like this:
```
X-Forwarded-Client-Cert:[By=spiffe://efad7282-d9b2-3298-f6d8-38b37fb58df3.consul/ns/default/dc/dc1/svc/counting;Hash=61ad5cbdfcb50f5a3ec0ca60923d61613c149a9d4495010a64175c05a0268ab2;Cert="-----BEGIN%20CERTIFICATE-----%0AMIICHDCCAcOgAwIBAgIBCDAKBggqhkjOPQQDAjAxMS8wLQYDVQQDEyZwcmktMTli%0AYXdyb2YuY29uc3VsLmNhLmVmYWQ3MjgyLmNvbnN1bDAeFw0yMjA0MjkwMzE0NTBa%0AFw0yMjA1MDIwMzE0NTBaMAAwWTATBgcqhkjOPQIBBggqhkjOPQMBBwNCAARVIZ7Y%0AZEXfbOGBfxGa7Vuok1MIng%2FuzLQK2xLVlSTIPDbO5hstTGP%2B%2FGx182PYFP3jYqk5%0Aq6rYWe1wiPNMA30Io4H8MIH5MA4GA1UdDwEB%2FwQEAwIDuDAdBgNVHSUEFjAUBggr%0ABgEFBQcDAgYIKwYBBQUHAwEwDAYDVR0TAQH%2FBAIwADApBgNVHQ4EIgQgrp4q50oX%0AHHghMbxz5Bk8OJFWMdfgH0Upr350WlhyxvkwKwYDVR0jBCQwIoAgUe6uERAIj%2FLM%0AyuFzDc3Wbp9TGAKBJYAwyhF14ToOQCMwYgYDVR0RAQH%2FBFgwVoZUc3BpZmZlOi8v%0AZWZhZDcyODItZDliMi0zMjk4LWY2ZDgtMzhiMzdmYjU4ZGYzLmNvbnN1bC9ucy9k%0AZWZhdWx0L2RjL2RjMS9zdmMvZGFzaGJvYXJkMAoGCCqGSM49BAMCA0cAMEQCIDwb%0AFlchufggNTijnQ5SUcvTZrWlZyq%2FrdVC20nbbmWLAiAVshNNv1xBqJI1NmY2HI9n%0AgRMfb8aEPVSuxEHhqy57eQ%3D%3D%0A-----END%20CERTIFICATE-----%0A";Chain="-----BEGIN%20CERTIFICATE-----%0AMIICHDCCAcOgAwIBAgIBCDAKBggqhkjOPQQDAjAxMS8wLQYDVQQDEyZwcmktMTli%0AYXdyb2YuY29uc3VsLmNhLmVmYWQ3MjgyLmNvbnN1bDAeFw0yMjA0MjkwMzE0NTBa%0AFw0yMjA1MDIwMzE0NTBaMAAwWTATBgcqhkjOPQIBBggqhkjOPQMBBwNCAARVIZ7Y%0AZEXfbOGBfxGa7Vuok1MIng%2FuzLQK2xLVlSTIPDbO5hstTGP%2B%2FGx182PYFP3jYqk5%0Aq6rYWe1wiPNMA30Io4H8MIH5MA4GA1UdDwEB%2FwQEAwIDuDAdBgNVHSUEFjAUBggr%0ABgEFBQcDAgYIKwYBBQUHAwEwDAYDVR0TAQH%2FBAIwADApBgNVHQ4EIgQgrp4q50oX%0AHHghMbxz5Bk8OJFWMdfgH0Upr350WlhyxvkwKwYDVR0jBCQwIoAgUe6uERAIj%2FLM%0AyuFzDc3Wbp9TGAKBJYAwyhF14ToOQCMwYgYDVR0RAQH%2FBFgwVoZUc3BpZmZlOi8v%0AZWZhZDcyODItZDliMi0zMjk4LWY2ZDgtMzhiMzdmYjU4ZGYzLmNvbnN1bC9ucy9k%0AZWZhdWx0L2RjL2RjMS9zdmMvZGFzaGJvYXJkMAoGCCqGSM49BAMCA0cAMEQCIDwb%0AFlchufggNTijnQ5SUcvTZrWlZyq%2FrdVC20nbbmWLAiAVshNNv1xBqJI1NmY2HI9n%0AgRMfb8aEPVSuxEHhqy57eQ%3D%3D%0A-----END%20CERTIFICATE-----%0A";Subject="";URI=spiffe://efad7282-d9b2-3298-f6d8-38b37fb58df3.consul/ns/default/dc/dc1/svc/dashboard]
```
Closes#12852
The primary bug here is in the streaming subsystem that makes the overall v1/health/service/:service request behave incorrectly when servicing a blocking request with a filter provided.
There is a secondary non-streaming bug being fixed here that is much less obvious related to when to update the `reply` variable in a `blockingQuery` evaluation. It is unlikely that it is triggerable in practical environments and I could not actually get the bug to manifest, but I fixed it anyway while investigating the original issue.
Simple reproduction (streaming):
1. Register a service with a tag.
curl -sL --request PUT 'http://localhost:8500/v1/agent/service/register' \
--header 'Content-Type: application/json' \
--data-raw '{ "ID": "ID1", "Name": "test", "Tags":[ "a" ], "EnableTagOverride": true }'
2. Do an initial filter query that matches on the tag.
curl -sLi --get 'http://localhost:8500/v1/health/service/test' --data-urlencode 'filter=a in Service.Tags'
3. Note you get one result. Use the `X-Consul-Index` header to establish
a blocking query in another terminal, this should not return yet.
curl -sLi --get 'http://localhost:8500/v1/health/service/test?index=$INDEX' --data-urlencode 'filter=a in Service.Tags'
4. Re-register that service with a different tag.
curl -sL --request PUT 'http://localhost:8500/v1/agent/service/register' \
--header 'Content-Type: application/json' \
--data-raw '{ "ID": "ID1", "Name": "test", "Tags":[ "b" ], "EnableTagOverride": true }'
5. Your blocking query from (3) should return with a header
`X-Consul-Query-Backend: streaming` and empty results if it works
correctly `[]`.
Attempts to reproduce with non-streaming failed (where you add `&near=_agent` to the read queries and ensure `X-Consul-Query-Backend: blocking-query` shows up in the results).
* update raft to v1.3.7
* add changelog
* fix compilation error
* fix HeartbeatTimeout
* fix ElectionTimeout to reload only if value is valid
* fix default values for `ElectionTimeout` and `HeartbeatTimeout`
* fix test defaults
* bump raft to v1.3.8
Adds a timeout (deadline) to client RPC calls, so that streams will no longer hang indefinitely in unstable network conditions.
Co-authored-by: kisunji <ckim@hashicorp.com>
* Implement the ServerDiscovery.WatchServers gRPC endpoint
* Fix the ConnectCA.Sign gRPC endpoints metadata forwarding.
* Unify public gRPC endpoints around the public.TraceID function for request_id logging
Fixes#12048Fixes#12319
Regression introduced in #11693
Local reproduction steps:
1. `consul agent -dev`
2. `curl -sLiv 'localhost:8500/v1/agent/connect/ca/leaf/web'`
3. make note of the `X-Consul-Index` header returned
4. `curl -sLi 'localhost:8500/v1/agent/connect/ca/leaf/web?index=<VALUE_FROM_STEP_3>'`
5. Kill the above curl when it hangs with Ctrl-C
6. Repeat (2) and it should not hang.
Adds a new gRPC endpoint to get envoy bootstrap params. The new consul-dataplane service will use this
endpoint to generate an envoy bootstrap configuration.
Whenever autopilot updates its state it notifies Consul. That notification will then trigger Consul to extract out the ready server information. If the ready servers have changed, then an event will be published to notify any subscribers of the full set of ready servers.
All these ready server event things are contained within an autopilotevents package instead of the consul package to make importing them into the grpc related packages possible
Introduces a gRPC endpoint for signing Connect leaf certificates. It's also
the first of the public gRPC endpoints to perform leader-forwarding, so
establishes the pattern of forwarding over the multiplexed internal RPC port.
Previously we had 1 EventPublisher per state.Store. When a state store was closed/abandoned such as during a consul snapshot restore, this had the behavior of force closing subscriptions for that topic and evicting event snapshots from the cache.
The intention of this commit is to keep all that behavior. To that end, the shared EventPublisher now supports the ability to refresh a topic. That will perform the force close + eviction. The FSM upon abandoning the previous state.Store will call RefreshTopic for all the topics with events generated by the state.Store.
Just like standard upstreams the order of applicability in descending precedence:
1. caller's `service-defaults` upstream override for destination
2. caller's `service-defaults` upstream defaults
3. destination's `service-resolver` ConnectTimeout
4. system default of 5s
Co-authored-by: mrspanishviking <kcardenas@hashicorp.com>
If a service is automatically registered because it has a critical health check
for longer than deregister_critical_service_after, the error message will now
include:
- mention of the deregister_critical_service_after option
- the value of deregister_critical_service_after for that check
* Fixes a lint warning about t.Errorf not supporting %w
* Enable running autopilot on all servers
On the non-leader servers all they do is update the state and do not attempt any modifications.
* Fix the RPC conn limiting tests
Technically they were relying on racey behavior before. Now they should be reliable.
Adds a new gRPC service and endpoint to return the list of supported
consul dataplane features. The Consul Dataplane will use this API to
customize its interaction with that particular server.
Adds a new gRPC streaming endpoint (WatchRoots) that dataplane clients will
use to fetch the current list of active Connect CA roots and receive new
lists whenever the roots are rotated.
* add config watcher to the config package
* add logging to watcher
* add test and refactor to add WatcherEvent.
* add all API calls and fix a bug with recreated files
* add tests for watcher
* remove the unnecessary use of context
* Add debug log and a test for file rename
* use inode to detect if the file is recreated/replaced and only listen to create events.
* tidy ups (#1535)
* tidy ups
* Add tests for inode reconcile
* fix linux vs windows syscall
* fix linux vs windows syscall
* fix windows compile error
* increase timeout
* use ctime ID
* remove remove/creation test as it's a use case that fail in linux
* fix linux/windows to use Ino/CreationTime
* fix the watcher to only overwrite current file id
* fix linter error
* fix remove/create test
* set reconcile loop to 200 Milliseconds
* fix watcher to not trigger event on remove, add more tests
* on a remove event try to add the file back to the watcher and trigger the handler if success
* fix race condition
* fix flaky test
* fix race conditions
* set level to info
* fix when file is removed and get an event for it after
* fix to trigger handler when we get a remove but re-add fail
* fix error message
* add tests for directory watch and fixes
* detect if a file is a symlink and return an error on Add
* rename Watcher to FileWatcher and remove symlink deref
* add fsnotify@v1.5.1
* fix go mod
* do not reset timer on errors, rename OS specific files
* rename New func
* events trigger on write and rename
* add missing test
* fix flaking tests
* fix flaky test
* check reconcile when removed
* delete invalid file
* fix test to create files with different mod time.
* back date file instead of sleeping
* add watching file in agent command.
* fix watcher call to use new API
* add configuration and stop watcher when server stop
* add certs as watched files
* move FileWatcher to the agent start instead of the command code
* stop watcher before replacing it
* save watched files in agent
* add add and remove interfaces to the file watcher
* fix remove to not return an error
* use `Add` and `Remove` to update certs files
* fix tests
* close events channel on the file watcher even when the context is done
* extract `NotAutoReloadableRuntimeConfig` is a separate struct
* fix linter errors
* add Ca configs and outgoing verify to the not auto reloadable config
* add some logs and fix to use background context
* add tests to auto-config reload
* remove stale test
* add tests to changes to config files
* add check to see if old cert files still trigger updates
* rename `NotAutoReloadableRuntimeConfig` to `StaticRuntimeConfig`
* fix to re add both key and cert file. Add test to cover this case.
* review suggestion
Co-authored-by: R.B. Boyer <4903+rboyer@users.noreply.github.com>
* add check to static runtime config changes
* fix test
* add changelog file
* fix review comments
* Apply suggestions from code review
Co-authored-by: R.B. Boyer <4903+rboyer@users.noreply.github.com>
* update flag description
Co-authored-by: FFMMM <FFMMM@users.noreply.github.com>
* fix compilation error
* add static runtime config support
* fix test
* fix review comments
* fix log test
* Update .changelog/12329.txt
Co-authored-by: Dan Upton <daniel@floppy.co>
* transfer tests to runtime_test.go
* fix filewatcher Replace to not deadlock.
* avoid having lingering locks
Co-authored-by: R.B. Boyer <4903+rboyer@users.noreply.github.com>
* split ReloadConfig func
* fix warning message
Co-authored-by: R.B. Boyer <4903+rboyer@users.noreply.github.com>
* convert `FileWatcher` into an interface
* fix compilation errors
* fix tests
* extract func for adding and removing files
* add a coalesceTimer with a very small timer
* extract coaelsce Timer and add a shim for testing
* add tests to coalesceTimer fix to send remaining events
* set `coalesceTimer` to 1 Second
* support symlink, fix a nil deref.
* fix compile error
* fix compile error
* refactor file watcher rate limiting to be a Watcher implementation
* fix linter issue
* fix runtime config
* fix runtime test
* fix flaky tests
* fix compile error
* Apply suggestions from code review
Co-authored-by: R.B. Boyer <4903+rboyer@users.noreply.github.com>
* fix agent New to return an error if File watcher New return an error
* quit timer loop if ctx is canceled
* Apply suggestions from code review
Co-authored-by: Chris S. Kim <ckim@hashicorp.com>
Co-authored-by: Ashwin Venkatesh <ashwin@hashicorp.com>
Co-authored-by: R.B. Boyer <4903+rboyer@users.noreply.github.com>
Co-authored-by: FFMMM <FFMMM@users.noreply.github.com>
Co-authored-by: Daniel Upton <daniel@floppy.co>
Co-authored-by: Chris S. Kim <ckim@hashicorp.com>
* Avoid doing list of /sys/mounts
From an internal ticket "Support standard "Vault namespace in the path" semantics for Connect Vault CA Provider"
Vault allows the namespace to be specified as a prefix in the path of
a PKI definition, but this doesn't currently work for
```IntermediatePKIPath``` specifications, because we attempt to list
all of the paths to check if ours is already defined. This doesn't
really work in a namespaced world.
This changes the IntermediatePKIPath code to follow the same pattern
as the root key, where we directly get the key rather than listing.
This code is difficult to write automated tests for because it relies
on features of Vault Enterprise, which isn't currently part of our
test framework, so it was tested manually.
Signed-off-by: Mark Anderson <manderson@hashicorp.com>
* add changelog
Signed-off-by: Mark Anderson <manderson@hashicorp.com>
Matt Keeler
R.B. Boyer
Dan Upton
cskh
R.B. Boyer
Freddy
Dhia Ayachi
Evan Culver
Jared Kirschner
Jared Kirschner
FFMMM
Mark Anderson
Eric Haberkorn
Eric
Riddhi Shah
Chris S. Kim
Kyle Havlovitz
Mathew Estafanous
Will Jordan
Blake Covarrubias
DanStough
John Murret
Paul Glass
Karl Cardenas
Natalie Smith