Given a deployment that has a `progress_deadline`, if a task group runs
out of reschedule attempts, allow it to fail at this time instead of
waiting until the `progress_deadline` is reached.
Fixes: #17260
This complements the `env` parameter, so that the operator can author
tasks that don't share their Vault token with the workload when using
`image` filesystem isolation. As a result, more powerful tokens can be used
in a job definition, allowing it to use template stanzas to issue all kinds of
secrets (database secrets, Vault tokens with very specific policies, etc.),
without sharing that issuing power with the task itself.
This is accomplished by creating a directory called `private` within
the task's working directory, which shares many properties of
the `secrets` directory (tmpfs where possible, not accessible by
`nomad alloc fs` or Nomad's web UI), but isn't mounted into/bound to the
container.
If the `disable_file` parameter is set to `false` (its default), the Vault token
is also written to the NOMAD_SECRETS_DIR, so the default behavior is
backwards compatible. Even if the operator never changes the default,
they will still benefit from the improved behavior of Nomad never reading
the token back in from that - potentially altered - location.
Add structs and fields to support the Nomad Pools Governance Enterprise
feature of controlling node pool access via namespaces.
Nomad Enterprise allows users to specify a default node pool to be used
by jobs that don't specify one. In order to accomplish this, it's
necessary to distinguish between a job that explicitly uses the
`default` node pool and one that did not specify any.
If the `default` node pool is set during job canonicalization it's
impossible to do this, so this commit allows a job to have an empty node
pool value during registration but sets to `default` at the admission
controller mutator.
In order to guarantee state consistency the state store validates that
the job node pool is set and exists before inserting it.
When registering a node with a new node pool in a non-authoritative
region we can't create the node pool because this new pool will not be
replicated to other regions.
This commit modifies the node registration logic to only allow automatic
node pool creation in the authoritative region.
In non-authoritative regions, the client is registered, but the node
pool is not created. The client is kept in the `initialing` status until
its node pool is created in the authoritative region and replicated to
the client's region.
Implement scheduler support for node pool:
* When a scheduler is invoked, we get a set of the ready nodes in the DCs that
are allowed for that job. Extend the filter to include the node pool.
* Ensure that changes to a job's node pool are picked up as destructive
allocation updates.
* Add `NodesInPool` as a metric to all reporting done by the scheduler.
* Add the node-in-pool the filter to the `Node.Register` RPC so that we don't
generate spurious evals for nodes in the wrong pool.
Move all validations related to task fields to Task.Validate(). Prior to
this, some task validations were being done inside TaskGroup.Validate()
because they required access to some group values.
But similarly to how TaskGroup.Validate() tasks the job as parameter,
it's fair to expect the task to receive its group.
This changeset only adds the `node_pool` field to the jobspec, and ensures that
it gets picked up correctly as a change. Without the rest of the implementation
landed yet, the field will be ignored.
Implementation of the base work for the new node pools feature. It includes a new `NodePool` struct and its corresponding state store table.
Upon start the state store is populated with two built-in node pools that cannot be modified nor deleted:
* `all` is a node pool that always includes all nodes in the cluster.
* `default` is the node pool where nodes that don't specify a node pool in their configuration are placed.
* core: eliminate second index on job_submissions table
This PR refactors the job_submissions state store code to eliminate the
use of a second index formerly used for purging all versions of a given
job. In practice we ended up with duplicate entries on the table. Instead,
use index prefix scanning on the primary index and tidy up any potential
for creating (or removing) duplicates.
* core: pr comments followup
When client nodes are restarted, all allocations that have been scheduled on the
node have their modify index updated, including terminal allocations. There are
several contributing factors:
* The `allocSync` method that updates the servers isn't gated on first contact
with the servers. This means that if a server updates the desired state while
the client is down, the `allocSync` races with the `Node.ClientGetAlloc`
RPC. This will typically result in the client updating the server with "running"
and then immediately thereafter "complete".
* The `allocSync` method unconditionally sends the `Node.UpdateAlloc` RPC even
if it's possible to assert that the server has definitely seen the client
state. The allocrunner may queue-up updates even if we gate sending them. So
then we end up with a race between the allocrunner updating its internal state
to overwrite the previous update and `allocSync` sending the bogus or duplicate
update.
This changeset adds tracking of server-acknowledged state to the
allocrunner. This state gets checked in the `allocSync` before adding the update
to the batch, and updated when `Node.UpdateAlloc` returns successfully. To
implement this we need to be able to equality-check the updates against the last
acknowledged state. We also need to add the last acknowledged state to the
client state DB, otherwise we'd drop unacknowledged updates across restarts.
The client restart test has been expanded to cover a variety of allocation
states, including allocs stopped before shutdown, allocs stopped by the server
while the client is down, and allocs that have been completely GC'd on the
server while the client is down. I've also bench tested scenarios where the task
workload is killed while the client is down, resulting in a failed restore.
Fixes#16381
When the server restarts for the upgrade, it loads the `structs.Job` from the
Raft snapshot/logs. The jobspec has long since been parsed, so none of the
guards around the default value are in play. The empty field value for `Enabled`
is the zero value, which is false.
This doesn't impact any running allocation because we don't replace running
allocations when either the client or server restart. But as soon as any
allocation gets rescheduled (ex. you drain all your clients during upgrades),
it'll be using the `structs.Job` that the server has, which has `Enabled =
false`, and logs will not be collected.
This changeset fixes the bug by adding a new field `Disabled` which defaults to
false (so that the zero value works), and deprecates the old field.
Fixes#17076
Their release notes are here: https://github.com/golang-jwt/jwt/releases
Seemed wise to upgrade before we do even more with JWTs. For example
this upgrade *would* have mattered if we already implemented common JWT
claims such as expiration. Since we didn't rely on any claim
verification this upgrade is a noop...
...except for 1 test that called `Claims.Valid()`! Removing that
assertion *seems* scary, but it didn't actually do anything because we
didn't implement any of the standard claims it validated:
https://github.com/golang-jwt/jwt/blob/v4.5.0/map_claims.go#L120-L151
So functionally this major upgrade is a noop.
Some Nomad users ship application logs out-of-band via syslog. For these users
having `logmon` (and `docker_logger`) running is unnecessary overhead. Allow
disabling the logmon and pointing the task's stdout/stderr to /dev/null.
This changeset is the first of several incremental improvements to log
collection short of full-on logging plugins. The next step will likely be to
extend the internal-only task driver configuration so that cluster
administrators can turn off log collection for the entire driver.
---
Fixes: #11175
Co-authored-by: Thomas Weber <towe75@googlemail.com>
* Upgrade from hashicorp/go-msgpack v1.1.5 to v2.1.0
Fixes#16808
* Update hashicorp/net-rpc-msgpackrpc to v2 to match go-msgpack
* deps: use go-msgpack v2.0.0
go-msgpack v2.1.0 includes some code changes that we will need to
investigate furthere to assess its impact on Nomad, so keeping this
dependency on v2.0.0 for now since it's no-op.
---------
Co-authored-by: Luiz Aoqui <luiz@hashicorp.com>
* func: add namespace support for list deployment
* func: add wildcard to namespace filter for deployments
* Update deployment_endpoint.go
* style: use must instead of require or asseert
* style: rename paginator to avoid clash with import
* style: add changelog entry
* fix: add missing parameter for upsert jobs
* api: enable support for setting original source alongside job
This PR adds support for setting job source material along with
the registration of a job.
This includes a new HTTP endpoint and a new RPC endpoint for
making queries for the original source of a job. The
HTTP endpoint is /v1/job/<id>/submission?version=<version> and
the RPC method is Job.GetJobSubmission.
The job source (if submitted, and doing so is always optional), is
stored in the job_submission memdb table, separately from the
actual job. This way we do not incur overhead of reading the large
string field throughout normal job operations.
The server config now includes job_max_source_size for configuring
the maximum size the job source may be, before the server simply
drops the source material. This should help prevent Bad Things from
happening when huge jobs are submitted. If the value is set to 0,
all job source material will be dropped.
* api: avoid writing var content to disk for parsing
* api: move submission validation into RPC layer
* api: return an error if updating a job submission without namespace or job id
* api: be exact about the job index we associate a submission with (modify)
* api: reword api docs scheduling
* api: prune all but the last 6 job submissions
* api: protect against nil job submission in job validation
* api: set max job source size in test server
* api: fixups from pr
The `ephemeral_disk` block's `migrate` field allows for best-effort migration of
the ephemeral disk data to new nodes. The documentation says the `migrate` field
is only respected if `sticky=true`, but in fact if client ACLs are not set the
data is migrated even if `sticky=false`.
The existing behavior when client ACLs are disabled has existed since the early
implementation, so "fixing" that case now would silently break backwards
compatibility. Additionally, having `migrate` not imply `sticky` seems
nonsensical: it suggests that if we place on a new node we migrate the data but
if we place on the same node, we throw the data away!
Update so that `migrate=true` implies `sticky=true` as follows:
* The failure mode when client ACLs are enabled comes from the server not passing
along a migration token. Update the server so that the server provides a
migration token whenever `migrate=true` and not just when `sticky=true` too.
* Update the scheduler so that `migrate` implies `sticky`.
* Update the client so that we check for `migrate || sticky` where appropriate.
* Refactor the E2E tests to move them off the old framework and make the intention
of the test more clear.
The tests for the system allocs reconciling code path (`diffSystemAllocs`)
include many impossible test environments, such as passing allocs for the wrong
node into the function. This makes the test assertions nonsensible for use in
walking yourself through the correct behavior.
I've pulled this changeset out of PR #16097 so that we can merge these
improvements and revisit the right approach to fix the problem in #16097 with
less urgency now that the PFNR bug fix has been merged. This changeset breaks up
a couple of tests, expands test coverage, and makes test assertions more
clear. It also corrects one bit of production code that behaves fine in
production because of canonicalization, but forces us to remember to set values
in tests to compensate.
Wildcard datacenters introduced a bug where a job with any wildcard datacenters
will always be treated as a destructive update when we check whether a
datacenter has been removed from the jobspec.
Includes updating the helper so that callers don't have to loop over the job's
datacenters.
This change resolves policies for workload identities when calling Client RPCs. Previously only ACL tokens could be used for Client RPCs.
Since the same cache is used for both bearer tokens (ACL and Workload ID), the token cache size was doubled.
---------
Co-authored-by: James Rasell <jrasell@users.noreply.github.com>
This change introduces the Task API: a portable way for tasks to access Nomad's HTTP API. This particular implementation uses a Unix Domain Socket and, unlike the agent's HTTP API, always requires authentication even if ACLs are disabled.
This PR contains the core feature and tests but followup work is required for the following TODO items:
- Docs - might do in a followup since dynamic node metadata / task api / workload id all need to interlink
- Unit tests for auth middleware
- Caching for auth middleware
- Rate limiting on negative lookups for auth middleware
---------
Co-authored-by: Seth Hoenig <shoenig@duck.com>
Service jobs should have unique allocation Names, derived from the
Job.ID. System jobs do not have unique allocation Names because the index is
intended to indicated the instance out of a desired count size. Because system
jobs do not have an explicit count but the results are based on the targeted
nodes, the index is less informative and this was intentionally omitted from the
original design.
Update docs to make it clear that NOMAD_ALLOC_INDEX is always zero for
system/sysbatch jobs
Validate that `volume.per_alloc` is incompatible with system/sysbatch jobs.
System and sysbatch jobs always have a `NOMAD_ALLOC_INDEX` of 0. So
interpolation via `per_alloc` will not work as soon as there's more than one
allocation placed. Validate against this on job submission.
* largely a doc-ification of this commit message:
d47678074bf8ae9ff2da3c91d0729bf03aee8446
this doesn't spell out all the possible failure modes,
but should be a good starting point for folks.
* connect: add doc link to envoy bootstrap error
* add Unwrap() to RecoverableError
mainly for easier testing
Add `identity` jobspec block to expose workload identity tokens to tasks.
---------
Co-authored-by: Anders <mail@anars.dk>
Co-authored-by: Tim Gross <tgross@hashicorp.com>
Co-authored-by: Michael Schurter <mschurter@hashicorp.com>
When a Nomad client that is running an allocation with
`max_client_disconnect` set misses a heartbeat the Nomad server will
update its status to `disconnected`.
Upon reconnecting, the client will make three main RPC calls:
- `Node.UpdateStatus` is used to set the client status to `ready`.
- `Node.UpdateAlloc` is used to update the client-side information about
allocations, such as their `ClientStatus`, task states etc.
- `Node.Register` is used to upsert the entire node information,
including its status.
These calls are made concurrently and are also running in parallel with
the scheduler. Depending on the order they run the scheduler may end up
with incomplete data when reconciling allocations.
For example, a client disconnects and its replacement allocation cannot
be placed anywhere else, so there's a pending eval waiting for
resources.
When this client comes back the order of events may be:
1. Client calls `Node.UpdateStatus` and is now `ready`.
2. Scheduler reconciles allocations and places the replacement alloc to
the client. The client is now assigned two allocations: the original
alloc that is still `unknown` and the replacement that is `pending`.
3. Client calls `Node.UpdateAlloc` and updates the original alloc to
`running`.
4. Scheduler notices too many allocs and stops the replacement.
This creates unnecessary placements or, in a different order of events,
may leave the job without any allocations running until the whole state
is updated and reconciled.
To avoid problems like this clients must update _all_ of its relevant
information before they can be considered `ready` and available for
scheduling.
To achieve this goal the RPC endpoints mentioned above have been
modified to enforce strict steps for nodes reconnecting:
- `Node.Register` does not set the client status anymore.
- `Node.UpdateStatus` sets the reconnecting client to the `initializing`
status until it successfully calls `Node.UpdateAlloc`.
These changes are done server-side to avoid the need of additional
coordination between clients and servers. Clients are kept oblivious of
these changes and will keep making these calls as they normally would.
The verification of whether allocations have been updates is done by
storing and comparing the Raft index of the last time the client missed
a heartbeat and the last time it updated its allocations.
Implement a metric for RPC requests with labels on the identity, so that
administrators can monitor the source of requests within the cluster. This
changeset demonstrates the change with the new `ACL.WhoAmI` RPC, and we'll wire
up the remaining RPCs once we've threaded the new pre-forwarding authentication
through the all.
Note that metrics are measured after we forward but before we return any
authentication error. This ensures that we only emit metrics on the server that
actually serves the request. We'll perform rate limiting at the same place.
Includes telemetry configuration to omit identity labels.
* consul: correctly understand missing consul checks as unhealthy
This PR fixes a bug where Nomad assumed any registered Checks would exist
in the service registration coming back from Consul. In some cases, the
Consul may be slow in processing the check registration, and the response
object would not contain checks. Nomad would then scan the empty response
looking for Checks with failing health status, finding none, and then
marking a task/alloc as healthy.
In reality, we must always use Nomad's view of what checks should exist as
the source of truth, and compare that with the response Consul gives us,
making sure they match, before scanning the Consul response for failing
check statuses.
Fixes#15536
* consul: minor CR refactor using maps not sets
* consul: observe transition from healthy to unhealthy checks
* consul: spell healthy correctly
This changeset covers a sidebar discussion that @schmichael and I had around the
design for pre-forwarding auth. This includes some changes extracted out of
#15513 to make it easier to review both and leave a clean history.
* Remove fast path for NodeID. Previously-connected clients will have a NodeID
set on the context, and because this is a large portion of the RPCs sent we
fast-pathed it at the top of the `Authenticate` method. But the context is
shared for all yamux streams over the same yamux session (and TCP
connection). This lets an authenticated HTTP request to a client use the
NodeID for authentication, which is a privilege escalation. Remove the fast
path and annotate it so that we don't break it again.
* Add context to decisions around AuthenticatedIdentity. The `Authenticate`
method taken on its own looks like it wants to return an `acl.ACL` that folds
over all the various identity types (creating an ephemeral ACL on the fly if
neccessary). But keeping these fields idependent allows RPC handlers to
differentiate between internal and external origins so we most likely want to
avoid this. Leave some docstrings as a warning as to why this is built the way
it is.
* Mutate the request rather than returning. When reviewing #15513 we decided
that forcing the request handler to call `SetIdentity` was repetitive and
error prone. Instead, the `Authenticate` method mutates the request by setting
its `AuthenticatedIdentity`.
This change adds a new table that will store ACL binding rule
objects. The two indexes allow fast lookups by their ID, or by
which auth method they are linked to. Snapshot persist and
restore functionality ensures this table can be saved and
restored from snapshots.
In order to write and delete the object to state, new Raft messages
have been added.
All RPC request and response structs, along with object functions
such as diff and canonicalize have been included within this work
as it is nicely separated from the other areas of work.
Upcoming work to instrument the rate of RPC requests by consumer (and eventually
rate limit) require that we authenticate a RPC request before forwarding. Add a
new top-level `Authenticate` method to the server and have it return an
`AuthenticatedIdentity` struct. RPC handlers will use the relevant fields of
this identity for performing authorization.
This changeset includes:
* The main implementation of `Authenticate`
* Provide a new RPC `ACL.WhoAmI` for debugging authentication. This endpoint
returns the same `AuthenticatedIdentity` that will be used by RPC handlers. At
some point we might want to give this an equivalent HTTP endpoint but I didn't
want to add that to our public API until some of the other Workload Identity
work is solidified, especially if we don't need it yet.
* A full coverage test of the `Authenticate` method. This sets up two server
nodes with mTLS and ACLs, some tokens, and some allocations with workload
identities.
* Wire up an example of using `Authenticate` in the `Namespace.Upsert` RPC and
see how authorization happens after forwarding.
* A new semgrep rule for `Authenticate`, which we'll need to update once we're
ready to wire up more RPC endpoints with authorization steps.
* scheduler: create placements for non-register MRD
For multiregion jobs, the scheduler does not create placements on
registration because the deployment must wait for the other regions.
Once of these regions will then trigger the deployment to run.
Currently, this is done in the scheduler by considering any eval for a
multiregion job as "paused" since it's expected that another region will
eventually unpause it.
This becomes a problem where evals not triggered by a job registration
happen, such as on a node update. These types of regional changes do not
have other regions waiting to progress the deployment, and so they were
never resulting in placements.
The fix is to create a deployment at job registration time. This
additional piece of state allows the scheduler to differentiate between
a multiregion change, where there are other regions engaged in the
deployment so no placements are required, from a regional change, where
the scheduler does need to create placements.
This deployment starts in the new "initializing" status to signal to the
scheduler that it needs to compute the initial deployment state. The
multiregion deployment will wait until this deployment state is
persisted and its starts is set to "pending". Without this state
transition it's possible to hit a race condition where the plan applier
and the deployment watcher may step of each other and overwrite their
changes.
* changelog: add entry for #15325
During unusual outage recovery scenarios on large clusters, a backlog of
millions of evaluations can appear. In these cases, the `eval delete` command can
put excessive load on the cluster by listing large sets of evals to extract the
IDs and then sending larges batches of IDs. Although the command's batch size
was carefully tuned, we still need to be JSON deserialize, re-serialize to
MessagePack, send the log entries through raft, and get the FSM applied.
To improve performance of this recovery case, move the batching process into the
RPC handler and the state store. The design here is a little weird, so let's
look a the failed options first:
* A naive solution here would be to just send the filter as the raft request and
let the FSM apply delete the whole set in a single operation. Benchmarking with
1M evals on a 3 node cluster demonstrated this can block the FSM apply for
several minutes, which puts the cluster at risk if there's a leadership
failover (the barrier write can't be made while this apply is in-flight).
* A less naive but still bad solution would be to have the RPC handler filter
and paginate, and then hand a list of IDs to the existing raft log
entry. Benchmarks showed this blocked the FSM apply for 20-30s at a time and
took roughly an hour to complete.
Instead, we're filtering and paginating in the RPC handler to find a page token,
and then passing both the filter and page token in the raft log. The FSM apply
recreates the paginator using the filter and page token to get roughly the same
page of evaluations, which it then deletes. The pagination process is fairly
cheap (only abut 5% of the total FSM apply time), so counter-intuitively this
rework ends up being much faster. A benchmark of 1M evaluations showed this
blocked the FSM apply for 20-30ms at a time (typical for normal operations) and
completes in less than 4 minutes.
Note that, as with the existing design, this delete is not consistent: a new
evaluation inserted "behind" the cursor of the pagination will fail to be
deleted.
