As we moved to using `-detach` for registering jobs, we should wait
until allocs and deployments are created before asserting their
properties.
Fixing `TestNodeDrainIgnoreSystem` and `TestRescheduleProgressDeadlineFail` tests as they seem particularly flaky, failing 9 and 7 times (respectively) in the last two weeks.
* api: revert to defaulting to http/1
PR #10778 incidentally changed the api http client to connect with
HTTP/2 first. However, the websocket libraries used in `alloc exec`
features don't handle http/2 well, and don't downgrade to http/1
gracefully.
Given that the switch is incidental, and not requested by users.
Furthermore, api consumers can opt-in to forcing http/2 by setting
custom http clients.
Fixes#10922
Noticed that the private Enterprise repository dependencies drifted a bit. Here, we update the OSS to the dependencies used by Enterprise.
We should update all dependencies as a matter of hygiene, but that's an issue for another time.
Fix a panic in handling one-time auth tokens, used to support `nomad ui
--authenticate`.
If the nomad leader is a 1.1.x with some servers running as 1.0.x, the
pre-1.1.0 servers risk crashing and the cluster may lose quorum. That
can happen when `nomad authenticate -ui` command is issued, or when the
leader scans for expired tokens every 10 minutes.
Fixed#10943 .
Namespaces are set-up in Nomad to be an object that has an id property.
However, namespaces actually don't have that shape. Our search was expecting
a namespace object, but we actually don't have a namespace assigned to jobs
in our config and namespace is set to null. Normally, these namespaces would
be set to default, but that would require us to refactor our Mirage config
if we wanted to assert that namespaces are 'default' and not null. So this is
a bandaid solution.
Support the new post-1.0.0 job spec fields in the HCLv1 parser.
The HCLv1 parser is still the default (or only!) parser in many downstream tools, e.g. [Levant](e48c439f14/template/render.go (L13-L30)), and [terraform-provider-nomad](bce32a7831/nomad/resource_job.go (L735-L743)) .
While we initially intended to deprecate HCLv1 parser in 1.0.0, we never communicated that publicly. We did not fully anticipate the public usage of `jobspec` package (we assumed it's a private package), or the friction that HCLv2 introduced in some cases (e.g. https://github.com/hashicorp/nomad/issues/10777, https://github.com/hashicorp/nomad/issues/9838).
So moving forward we intend to ensure that new job spec fields are honored in both the HCLv1 and HCLv2 parser until we solidify the migration path and communicate it properly.
Use glint to determine if os.Stdout is a terminal.
glint Terminal renderer expects os.Stdout [not only to be a terminal, but also to have non-zero size](b492b545f6/renderer_term.go (L39-L46)). It's unclear how this condition arises, but this additional check causes Nomad to render deployments progress through glint when glint cannot support it.
By using golint to perform the check, we eliminate the risk of mis-judgement.
* ui: add parameterized dispatch interface
This commit adds a new interface for dispatching parameteried jobs, if
the user has the right permissions. The UI can be accessed by viewing a
parameterized job and clicking on the "Dispatch Job" button located in
the "Job Launches" section.
* fix failing lint test
* clean up dispatch and remove meta
This commit cleans up a few things that had typos and
inconsistent naming. In line with this, the custom
`meta` view was removed in favor of using the
included `AttributesTable`.
* ui: encode dispatch job payload and start adding tests
* ui: remove unused test imports
* ui: redesign job dispatch form
* ui: initial acceptance tests for dispatch job
* ui: generate parameterized job children with correct id format
* ui: fix job dispatch breadcrumb link
* ui: refactor job dispatch component into glimmer component and add form validation
* ui: remove unused CSS class
* ui: align job dispatch button
* ui: handle namespace-specific requests on job dispatch
* ui: rename payloadMissing to payloadHasError
* ui: don't re-fetch job spec on dispatch job
* ui: keep overview tab selected on job dispatch page
* ui: fix task and task-group linting
* ui: URL encode job id on dispatch job tests
* ui: fix error when job meta is null
* ui: handle job dispatch from adapter
* ui: add more tests for dispatch job page
* ui: add "job dispatch" capability check
* ui: update job dispatch from code review
Co-authored-by: Luiz Aoqui <luiz@hashicorp.com>
In job versions, if you have an ACL token with a write policy
you should be able to revert a job, however, that was not the
case here. This is because we're using ember-can to check if
the user can run a job. That permission relies on policiesSupportRunning
which uses a function called namespaceIncludesCapability. We're going to
need to refactor any cases that use this function.
When the client launches, use a consistent read to fetch its own allocs,
but allow stale read afterwards as long as reads don't revert into older
state.
This change addresses an edge case affecting restarting client. When a
client restarts, it may fetch a stale data concerning its allocs: allocs
that have completed prior to the client shutdown may still have "run/running"
desired/client status, and have the client attempt to re-run again.
An alternative approach is to track the indices such that the client
set MinQueryIndex on the maximum index the client ever saw, or compare
received allocs against locally restored client state. Garbage
collection complicates this approach (local knowledge is not complete),
and the approach still risks starting "dead" allocations (e.g. the
allocation may have been placed when client just restarted and have
already been reschuled by the time the client started. This approach
here is effective against all kinds of stalness problems with small
overhead.
