Command: alloc exec

Alias: nomad exec

The alloc exec command runs a command in a running allocation.

Usage

nomad alloc exec [options] <allocation> <command> [<args>...]

The nomad exec command can be use to run commands inside a running task/allocation.

Use cases are for inspecting container state, debugging a failed application without needing ssh access into the node that's running the allocation.

This command executes the command in the given task in the allocation. If the allocation is only running a single task, the task name can be omitted. Optionally, the -job option may be used in which case a random allocation from the given job will be chosen.

When ACLs are enabled, this command requires a token with the alloc-exec, read-job, and list-jobs capabilities for the allocation's namespace. If the task driver does not have file system isolation (as with raw_exec), this command requires the alloc-node-exec, read-job, and list-jobs capabilities for the allocation's namespace.

General Options

-address=<addr>: The address of the Nomad server. Overrides the NOMAD_ADDR environment variable if set. Defaults to http://127.0.0.1:4646.
-region=<region>: The region of the Nomad server to forward commands to. Overrides the NOMAD_REGION environment variable if set. Defaults to the Agent's local region.
-namespace=<namespace>: The target namespace for queries and actions bound to a namespace. Overrides the NOMAD_NAMESPACE environment variable if set. If set to '*', subcommands which support this functionality query all namespaces authorized to user. Defaults to the "default" namespace.
-no-color: Disables colored command output. Alternatively, NOMAD_CLI_NO_COLOR may be set. This option takes precedence over -force-color.
-force-color: Forces colored command output. This can be used in cases where the usual terminal detection fails. Alternatively, NOMAD_CLI_FORCE_COLOR may be set. This option has no effect if -no-color is also used.
-ca-cert=<path>: Path to a PEM encoded CA cert file to use to verify the Nomad server SSL certificate. Overrides the NOMAD_CACERT environment variable if set.
-ca-path=<path>: Path to a directory of PEM encoded CA cert files to verify the Nomad server SSL certificate. If both -ca-cert and -ca-path are specified, -ca-cert is used. Overrides the NOMAD_CAPATH environment variable if set.
-client-cert=<path>: Path to a PEM encoded client certificate for TLS authentication to the Nomad server. Must also specify -client-key. Overrides the NOMAD_CLIENT_CERT environment variable if set.
-client-key=<path>: Path to an unencrypted PEM encoded private key matching the client certificate from -client-cert. Overrides the NOMAD_CLIENT_KEY environment variable if set.
-tls-server-name=<value>: The server name to use as the SNI host when connecting via TLS. Overrides the NOMAD_TLS_SERVER_NAME environment variable if set.
-tls-skip-verify: Do not verify TLS certificate. This is highly not recommended. Verification will also be skipped if NOMAD_SKIP_VERIFY is set.
-token: The SecretID of an ACL token to use to authenticate API requests with. Overrides the NOMAD_TOKEN environment variable if set.

Exec Options

-task: Sets the task to exec command in.
-job: Use a random allocation from the specified job ID.
-i: Pass stdin to the container, defaults to true. Pass -i=false to disable explicitly.
-t: Allocate a pseudo-tty, defaults to true if stdin is detected to be a tty session. Pass -t=false to disable explicitly.
-e <escape_char>: Sets the escape character for sessions with a pty (default: '~'). The escape character is only recognized at the beginning of a line. The escape character followed by a dot ('.') closes the connection. Setting the character to 'none' disables any escapes and makes the session fully transparent.

Examples

To start an interactive debugging session in a particular alloc, invoke exec command with your desired shell available inside the task:

$ nomad alloc exec eb17e557 /bin/bash
root@eb17e557:/data# # now run any debugging commands inside container
root@eb17e557:/data# # ps -ef

To run a command and stream results without starting an interactive shell, you can pass the command and its arguments to exec directly:

$ nomad alloc exec eb17e557 cat /etc/resolv.conf
...

When passing command arguments to be evaluated in task, you may need to ensure that your host shell doesn't interpolate values before invoking exec command. For example, the following command would return the environment variable on operator shell rather than task containers:

$ nomad alloc exec eb17e557 echo $NOMAD_ALLOC_ID # wrong
...

Here, we must start a shell in task to interpolate $NOMAD_ALLOC_ID, and quote command or use the heredoc syntax

$ nomad alloc exec eb17e557 /bin/sh -c 'echo $NOMAD_ALLOC_ID'
eb17e557-443e-4c51-c049-5bba7a9850bc

$ # by using heredoc and passing command in stdin
$ nomad alloc exec eb17e557 /bin/sh <<'EOF'
> echo $NOMAD_ALLOC_ID
> EOF
eb17e557-443e-4c51-c049-5bba7a9850bc

This technique applies when aiming to run a shell pipeline without streaming intermediate command output across the network:

$ nomad alloc exec eb17e557 /bin/sh -c 'cat /output | sort | uniq -c | sort -rn | head -n 5'
...

Using Job ID instead of Allocation ID

Setting the -job flag causes a random allocation of the specified job to be selected.

nomad alloc exec -job <job-id> <command> [<args>...]

Choosing a specific allocation is useful for debugging issues with a specific instance of a service. For other operations using the -job flag may be more convenient than looking up an allocation ID to use.

Disabling remote execution

alloc exec is enabled by default to aid with debugging. Operators can disable the feature by setting disable_remote_exec client config option on all clients, or a subset of clients that run sensitive workloads.

Exec targeting a specific task

When trying to alloc exec for a job that has more than one task associated with it, you may want to target a specific task.

# open a shell session in one of your allocation's tasks
$ nomad alloc exec -i -t -task mytask a1827f93 /bin/bash
a1827f93$