Before this change, Client had 2 copies of the config object: config and configCopy. There was no guidance around which to use where (other than configCopy's comment to pass it to alloc runners), both are shared among goroutines and mutated in data racy ways. At least at one point I think the idea was to have `config` be mutable and then grab a lock to overwrite `configCopy`'s pointer atomically. This would have allowed alloc runners to read their config copies in data race safe ways, but this isn't how the current implementation worked.
This change takes the following approach to safely handling configs in the client:
1. `Client.config` is the only copy of the config and all access must go through the `Client.configLock` mutex
2. Since the mutex *only protects the config pointer itself and not fields inside the Config struct:* all config mutation must be done on a *copy* of the config, and then Client's config pointer is overwritten while the mutex is acquired. Alloc runners and other goroutines with the old config pointer will not see config updates.
3. Deep copying is implemented on the Config struct to satisfy the previous approach. The TLS Keyloader is an exception because it has its own internal locking to support mutating in place. An unfortunate complication but one I couldn't find a way to untangle in a timely fashion.
4. To facilitate deep copying I made an *internally backward incompatible API change:* our `helper/funcs` used to turn containers (slices and maps) with 0 elements into nils. This probably saves a few memory allocations but makes it very easy to cause panics. Since my new config handling approach uses more copying, it became very difficult to ensure all code that used containers on configs could handle nils properly. Since this code has caused panics in the past, I fixed it: nil containers are copied as nil, but 0-element containers properly return a new 0-element container. No more "downgrading to nil!"
Return 429 response on HTTP max connection limit. Instead of silently closing
the connection, return a `429 Too Many Requests` HTTP response with a helpful
error message to aid debugging when the connection limit is unintentionally
reached.
Set a 10-millisecond write timeout and rate limiter for connection-limit 429
response to prevent writing the HTTP response from consuming too many server
resources.
Add `nomad.agent.http.exceeded metric` counting the number of HTTP connections
exceeding concurrency limit.
We expect every Nomad API client to use a single connection to any
given agent, so take advantage of keep-alive by switching the default
transport to `DefaultPooledClient`. Provide a facility to close idle
connections for testing purposes.
Restores the previously reverted #12409
Co-authored-by: Ben Buzbee <bbuzbee@cloudflare.com>
Add new namespace ACL requirement for the /v1/jobs/parse endpoint and
return early if HCLv2 parsing fails.
The endpoint now requires the new `parse-job` ACL capability or
`submit-job`.
The `TestHTTPServer_Limits_Error` test never starts the agent so it
had an incomplete configuration, which caused panics in the test. Fix
the configuration.
The PR #11555 had a branch name like `f-ui-*` which caused CI to skip
the unit tests over the HTTP handler setup, so this wasn't caught in
PR review.
This change modifies the Nomad job register and deregister RPCs to
accept an updated option set which includes eval priority. This
param is optional and override the use of the job priority to set
the eval priority.
In order to ensure all evaluations as a result of the request use
the same eval priority, the priority is shared to the
allocReconciler and deploymentWatcher. This creates a new
distinction between eval priority and job priority.
The Nomad agent HTTP API has been modified to allow setting the
eval priority on job update and delete. To keep consistency with
the current v1 API, job update accepts this as a payload param;
job delete accepts this as a query param.
Any user supplied value is validated within the agent HTTP handler
removing the need to pass invalid requests to the server.
The register and deregister opts functions now all for setting
the eval priority on requests.
The change includes a small change to the DeregisterOpts function
which handles nil opts. This brings the function inline with the
RegisterOpts.
Glint pulled in an updated version of mitchellh/go-testing-interface
which broke some existing tests because the update added a Parallel()
method to testing.T. This switches to the standard library testing.TB
which doesn't have a Parallel() method.
Adding '-verbose' will print out the allocation information for the
deployment. This also changes the job run command so that it now blocks
until deployment is complete and adds timestamps to the output so that
it's more in line with the output of node drain.
This uses glint to print in place in running in a tty. Because glint
doesn't yet support cmd/powershell, Windows workflows use a different
library to print in place, which results in slightly different
formatting: 1) different margins, and 2) no spinner indicating
deployment in progress.
The OTT feature relies on having a query parameter for a one-time token which
gets handled by the UI. We need to make sure that query param is preserved
when redirecting from the root URL to the `/ui/` URI.
https://github.com/hashicorp/nomad/pull/9608 introduced the use of the
built-in HTTP 429 response handler provided by go-connlimit. There is
concern though around plausible DOS attacks that need to be addressed,
so this PR reverts that functionality.
It keeps a fix in the tests around the use of an HTTPS enabled client
for when the server is listening on HTTPS. Previously, the tests would
fail deterministically with io.EOF because that's how the TLS server
terminates invalid connections.
Now, the result is much less deterministic. The state of the client
connection and the server socket depends on when the connection is
closed and how far along the handshake was.
Fixes#9017
The ?resources=true query parameter includes resources in the object
stub listings. Specifically:
- For `/v1/nodes?resources=true` both the `NodeResources` and
`ReservedResources` field are included.
- For `/v1/allocations?resources=true` the `AllocatedResources` field is
included.
The ?task_states=false query parameter removes TaskStates from
/v1/allocations responses. (By default TaskStates are included.)
Failed requests due to API client errors are to be marked as DEBUG.
The Error log level should be reserved to signal problems with the
cluster and are actionable for nomad system operators. Logs due to
misbehaving API clients don't represent a system level problem and seem
spurius to nomad maintainers at best. These log messages can also be
attack vectors for deniel of service attacks by filling servers disk
space with spurious log messages.
Introduce limits to prevent unauthorized users from exhausting all
ephemeral ports on agents:
* `{https,rpc}_handshake_timeout`
* `{http,rpc}_max_conns_per_client`
The handshake timeout closes connections that have not completed the TLS
handshake by the deadline (5s by default). For RPC connections this
timeout also separately applies to first byte being read so RPC
connections with TLS enabled have `rpc_handshake_time * 2` as their
deadline.
The connection limit per client prevents a single remote TCP peer from
exhausting all ephemeral ports. The default is 100, but can be lowered
to a minimum of 26. Since streaming RPC connections create a new TCP
connection (until MultiplexV2 is used), 20 connections are reserved for
Raft and non-streaming RPCs to prevent connection exhaustion due to
streaming RPCs.
All limits are configurable and may be disabled by setting them to `0`.
This also includes a fix that closes connections that attempt to create
TLS RPC connections recursively. While only users with valid mTLS
certificates could perform such an operation, it was added as a
safeguard to prevent programming errors before they could cause resource
exhaustion.
Fixes#3697
The existing code and test case only covered the leader behavior. When
querying against non-leaders the error has an "rpc error: " prefix.
To provide consistency in HTTP error response I also strip the "rpc
error: " prefix for 403 responses as they offer no beneficial additional
information (and in theory disclose a tiny bit of data to unauthorized
users, but it would be a pretty weird bit of data to use in a malicious
way).
* Allow server TLS configuration to be reloaded via SIGHUP
* dynamic tls reloading for nomad agents
* code cleanup and refactoring
* ensure keyloader is initialized, add comments
* allow downgrading from TLS
* initalize keyloader if necessary
* integration test for tls reload
* fix up test to assert success on reloaded TLS configuration
* failure in loading a new TLS config should remain at current
Reload only the config if agent is already using TLS
* reload agent configuration before specific server/client
lock keyloader before loading/caching a new certificate
* introduce a get-or-set method for keyloader
* fixups from code review
* fix up linting errors
* fixups from code review
* add lock for config updates; improve copy of tls config
* GetCertificate only reloads certificates dynamically for the server
* config updates/copies should be on agent
* improve http integration test
* simplify agent reloading storing a local copy of config
* reuse the same keyloader when reloading
* Test that server and client get reloaded but keep keyloader
* Keyloader exposes GetClientCertificate as well for outgoing connections
* Fix spelling
* correct changelog style
