Implement the new `nomad job restart` command that allows operators to
restart allocations tasks or reschedule then entire allocation.
Restarts can be batched to target multiple allocations in parallel.
Between each batch the command can stop and hold for a predefined time
or until the user confirms that the process should proceed.
This implements the "Stateless Restarts" alternative from the original
RFC
(https://gist.github.com/schmichael/e0b8b2ec1eb146301175fd87ddd46180).
The original concept is still worth implementing, as it allows this
functionality to be exposed over an API that can be consumed by the
Nomad UI and other clients. But the implementation turned out to be more
complex than we initially expected so we thought it would be better to
release a stateless CLI-based implementation first to gather feedback
and validate the restart behaviour.
Co-authored-by: Shishir Mahajan <smahajan@roblox.com>
* command: fixup job multi-stop test
This PR refactors the StopCommand test that runs 10 jobs and then
passes them all to one invokation of 'job stop'.
* test: swap use of assert for must
* test: cleanup job files we create
* command: fixup job stop failure tests
Now that JobStop works on concurrent jobs, the error messages are
different.
* cleanup: use multiple post scripts
Fix numerous go-getter security issues:
- Add timeouts to http, git, and hg operations to prevent DoS
- Add size limit to http to prevent resource exhaustion
- Disable following symlinks in both artifacts and `job run`
- Stop performing initial HEAD request to avoid file corruption on
retries and DoS opportunities.
**Approach**
Since Nomad has no ability to differentiate a DoS-via-large-artifact vs
a legitimate workload, all of the new limits are configurable at the
client agent level.
The max size of HTTP downloads is also exposed as a node attribute so
that if some workloads have large artifacts they can specify a high
limit in their jobspecs.
In the future all of this plumbing could be extended to enable/disable
specific getters or artifact downloading entirely on a per-node basis.
* cli: add -json flag to support job commands
While the CLI has always supported running JSON jobs, its support has
been via HCLv2's JSON parsing. I have no idea what format it expects the
job to be in, but it's absolutely not in the same format as the API
expects.
So I ignored that and added a new -json flag to explicitly support *API*
style JSON jobspecs.
The jobspecs can even have the wrapping {"Job": {...}} envelope or not!
* docs: fix example for `nomad job validate`
We haven't been able to validate inside driver config stanzas ever since
the move to task driver plugins. 😭
Fixes a bug where if a command flag parsing errors, the resulting error
and help usage messages get interleaved in unexpected and non-user
friendly way.
The reason is that we have flag parsing library effectively writes to
ui.Error in a goroutine. This is problematic: first, we lose the sequencing between help
usage and error message; second, cli.Ui methods are not concurrent safe.
Here, we introduce a custom error writer that buffers result and calls
ui.Error() in the write method and in the same goroutine.
For context, we need to wrap ui.Error because it's line-oriented, while
flags library expects a io.Writer which is bytes oriented.
This change updates the formatTime CLI helper function to return
an ISO_8601 time format which will make CLI time usage more
consistent and easier. Previosuly the time format was in US style
format which was somewhat confusing to non US users.
Closes#3806
This PR fixes our vet script and fixes all the missed vet changes.
It also fixes pointers being printed in `nomad stop <job>` and `nomad
node-status <node>`.
