This fixes few cases where driver eventor goroutines are leaked during
normal operations, but especially so in tests.
This change makes few modifications:
First, it switches drivers to use `Context`s to manage shutdown events.
Previously, it relied on callers invoking `.Shutdown()` function that is
specific to internal drivers only and require casting. Using `Contexts`
provide a consistent idiomatic way to manage lifecycle for both internal
and external drivers.
Also, I discovered few places where we don't clean up a temporary driver
instance in the plugin catalog code, where we dispense a driver to
inspect and validate the schema config without properly cleaning it up.
This updates Xterm.js to 4.6.0, which includes support for reverse-wraparound
mode, so we no longer need to use a vendored dependency, which closes#7461.
The interface for accessing the buffer that’s used for test assertions changed.
With the dependency now accessed conventionally, we can have it load only when
it’s needed by an exec popup window, which closes#7516. That saves us
≈60kb compressed in the dependency bundle!
Ensure that nomad steps down (and terminate leader goroutines) on
shutdown, when the server is the leader.
Without this change, `monitorLeadership` may handle `shutdownCh` event
and exit early before handling the raft `leaderCh` event and end up
leaking leadership goroutines.
We have been using fatih/hclfmt which is long abandoned. Instead, switch
to HashiCorp's own hclfmt implementation. There are some trivial changes in
behavior around whitespace.
Adding this settled makes this test pass now that Ember Data is using
fetch instead of jquery. The test was presumably always incorrect but
never flaked.
Going off of the error message being "Forbidden" was brittle to begin
with and no longer works with Fetch due to the error message coming from
jquery underpinnings that were unobserved by Ember Data's attempted
recreation.
This runs ember-test-audit for UI PRs to compare how long
the tests take for the base vs the PR. It posts or updates
a comparison of times and test count.
It’s somewhat slow to report back as it runs the test suite
three times to even out variability in a shared environment.
If we end up being unhappy with that slowness, we could try
running the repetitions in parallel as well, but that would
involve more changes to ember-test-audit.
When an allocation runs for a task driver that can't support volume mounts,
the mounting will fail in a way that can be hard to understand. With host
volumes this usually means failing silently, whereas with CSI the operator
gets inscrutable internals exposed in the `nomad alloc status`.
This changeset adds a MountConfig field to the task driver Capabilities
response. We validate this when the `csi_hook` or `volume_hook` fires and
return a user-friendly error.
Note that we don't currently have a way to get driver capabilities up to the
server, except through attributes. Validating this when the user initially
submits the jobspec would be even better than what we're doing here (and could
be useful for all our other capabilities), but that's out of scope for this
changeset.
Also note that the MountConfig enum starts with "supports all" in order to
support community plugins in a backwards compatible way, rather than cutting
them off from volume mounting unexpectedly.
Following the new volumewatcher in #7794 and performance improvements
to it that landed afterwards, there's no particular reason we should
be threading claim releases through the GC eval rather than writing an
empty `CSIVolumeClaimRequest` with the mode set to
`CSIVolumeClaimRelease`, just as the GC evaluation would do.
Also, by batching up these raft messages, we can reduce the amount of
raft writes by 1 and cross-server RPCs by 1 per volume we release
claims on.
The `stats_hook` writes an Error log every time an allocation becomes
terminal. This is a normal condition, not an error. A real error
condition like a failure to collect the stats is logged later. It just
creates log noise, and this is a particularly bad operator experience
for heavy batch workloads.
The tasklet passes the timeout for the script check into the task
driver's `Exec`, and its up to the task driver to enforce that via a
golang `context.WithDeadline`. In practice, this deadline is started
before the task driver starts setting up the execution
environment (because we need it to do things like timeout Docker API
calls).
Under even moderate load, the time it takes to set up the execution
context for the script check regularly exceeds a full second or
two. This can cause script checks to unexpected timeout or even never
execute if the context expires before the task driver ever gets a
chance to `execve`.
This changeset adds a notice to operators about setting script check
timeouts with plenty of padding and what to monitor for problems.
Some of the CSI RPC endpoints were missing validation that the ID or
the Volume definition was present. This could result in nonsense
`CSIVolume` structs being written to raft during registration. This
changeset corrects that bug and adds validation checks to present
nicer error messages to operators in some other cases.
Typically these filterable list views don't have titles beyond the
breadcrumbs, but since this page has no search bar, the title really
helps balance it out.
