When running at scale, it's possible that Docker Engine starts
containers successfully but gets wedged in a way where API call fails.
The Docker Engine may remain unavailable for arbitrary long time.
Here, we introduce a periodic reconcilation process that ensures that any
container started by nomad is tracked, and killed if is running
unexpectedly.
Basically, the periodic job inspects any container that isn't tracked in
its handlers. A creation grace period is used to prevent killing newly
created containers that aren't registered yet.
Also, we aim to avoid killing unrelated containters started by host or
through raw_exec drivers. The logic is to pattern against containers
environment variables and mounts to infer if they are an alloc docker
container.
Lastly, the periodic job can be disabled to avoid any interference if
need be.
This commit introduces support for configuring mount propagation when
mounting volumes with the `volume_mount` stanza on Linux targets.
Similar to Kubernetes, we expose 3 options for configuring mount
propagation:
- private, which is equivalent to `rprivate` on Linux, which does not allow the
container to see any new nested mounts after the chroot was created.
- host-to-task, which is equivalent to `rslave` on Linux, which allows new mounts
that have been created _outside of the container_ to be visible
inside the container after the chroot is created.
- bidirectional, which is equivalent to `rshared` on Linux, which allows both
the container to see new mounts created on the host, but
importantly _allows the container to create mounts that are
visible in other containers an don the host_
private and host-to-task are safe, but bidirectional mounts can be
dangerous, as if the code inside a container creates a mount, and does
not clean it up before tearing down the container, it can cause bad
things to happen inside the kernel.
To add a layer of safety here, we require that the user has ReadWrite
permissions on the volume before allowing bidirectional mounts, as a
defense in depth / validation case, although creating mounts should also require
a priviliged execution environment inside the container.
Without passing the network isolation configuration to the executor,
java tasks are not placed in the same network namespace as the other
processes in their task group, which breaks Consul Connect.
The docker creation API calls may fail with http errors (e.g. timeout)
even if container was successfully created.
Here, we force remove container if we got unexpected failure. We
already do this in some error handlers, and this commit updates all
paths.
I stopped short from a more aggressive refactoring, as the code is ripe
for refactoring and would rather do that in another PR.
This handles a bug where we may start a container successfully, yet we
fail due to retries and startContainer not being idempotent call.
Here, we ensure that when starting a container fails with 500 error,
the retry succeeds if container was started successfully.
hclspec.NewLiteral does not quote its values, which caused `3m` to be
parsed as a nonsensical literal which broke the plugin loader during
initialization. By quoting the value here, it starts correctly.
Using the API as provided from the `mounts` package imposes validation
on the `src:dest` which shouldn't be performed at this time. To workaround
that lift the internal code from that library required to only perform
the split.
Currently, nomad "plugin" processes (e.g. executor, logmon, docker_logger) are started as CLI
commands to be handled by command CLI framework. Plugin launchers use
`discover.NomadBinary()` to identify the binary and start it.
This has few downsides: The trivial one is that when running tests, one
must re-compile the nomad binary as the tests need to invoke the nomad
executable to start plugin. This is frequently overlooked, resulting in
puzzlement.
The more significant issue with `executor` in particular is in relation
to external driver:
* Plugin must identify the path of invoking nomad binary, which is not
trivial; `discvoer.NomadBinary()` now returns the path to the plugin
rather than to nomad, preventing external drivers from launching
executors.
* The external driver may get a different version of executor than it
expects (specially if we make a binary incompatible change in future).
This commit addresses both downside by having the plugin invocation
handling through an `init()` call, similar to how libcontainer init
handler is done in [1] and recommened by libcontainer [2]. `init()`
will be invoked and handled properly in tests and external drivers.
For external drivers, this change will cause external drivers to launch
the executor that's compiled against.
There a are a couple of downsides to this approach:
* These specific packages (i.e executor, logmon, and dockerlog) need to
be careful in use of `init()`, package initializers. Must avoid having
command execution rely on any other init in the package. I prefixed
files with `z_` (golang processes files in lexical order), but ensured
we don't depend on order.
* The command handling is spread in multiple packages making it a bit
less obvious how plugin starts are handled.
[1] drivers/shared/executor/libcontainer_nsenter_linux.go
[2] eb4aeed24f/libcontainer (using-libcontainer)
Nomad 0.9 incidentally set effective capabilities that is higher than
what's expected of a `nobody` process, and what's set in 0.8.
This change restores the capabilities to ones used in Nomad 0.9.
Implements streamign exec handling in both executors (i.e. universal and
libcontainer).
For creation of TTY, some incidental complexity leaked in. The universal
executor uses github.com/kr/pty for creation of TTYs.
On the other hand, libcontainer expects a console socket and for libcontainer to
create the underlying console object on process start. The caller can then use
`libcontainer.utils.RecvFd()` to get tty master end.
I chose github.com/kr/pty for managing TTYs here. I tried
`github.com/containerd/console` package (which is already imported), but the
package did not work as expected on macOS.
