open-nomad/plugins/drivers/base/proto/driver.proto

syntax = "proto3";
package hashicorp.nomad.plugins.drivers.base.proto;
option go_package = "proto";

import "google/protobuf/duration.proto";
import "google/protobuf/timestamp.proto";

import "github.com/hashicorp/nomad/plugins/shared/hclspec/hcl_spec.proto";

// Driver service defines RPCs used to communicate with a nomad runtime driver.
// Some rpcs may not be implemented by the driver based on it's capabilities.
service Driver {

    // DriverTaskConfigSchema returns the schema for parsing the driver
    // configuration of a task.
    rpc DriverTaskConfigSchema(DriverTaskConfigSchemaRequest) returns (DriverTaskConfigSchemaResponse) {}

    // Capabilities returns a set of features which the driver implements. Some
    // RPCs are not possible to implement on some runtimes, this allows the
    // driver to indicate if it doesn't support these RPCs and features.
    rpc Capabilities(CapabilitiesRequest) returns (CapabilitiesResponse) {}

    // Fingerprint starts a stream which emits information about the driver
    // including whether the driver healthy and able to function in the
    // existing environment.
    //
    // The driver should immediately stream a FingerprintResponse when the RPC
    // is initially called, then send any additional responses if there is a
    // change in the driver's state.
    rpc Fingerprint(FingerprintRequest) returns (stream FingerprintResponse) {}

    // RecoverTask is used when a task has been started but the driver may not
    // know about it. Such is the case if the driver restarts or is upgraded.
    rpc RecoverTask(RecoverTaskRequest) returns (RecoverTaskResponse) {}

    // StartTask starts and tracks the task on the implemented runtime
    rpc StartTask(StartTaskRequest) returns (StartTaskResponse) {}

    // WaitTask blocks until the given task exits, returning the result of the
    // task. It may be called after the task has exited, but before the task is
    // destroyed.
	rpc WaitTask(WaitTaskRequest) returns (WaitTaskResponse) {}

    // StopTask stops a given task by sending the desired signal to the process.
    // If the task does not exit on its own within the given timeout, it will be
    // forcefully killed.
    rpc StopTask(StopTaskRequest) returns (StopTaskResponse) {}

    // DestroyTask removes the task from the driver's internal state and cleans
    // up any additional resources created by the driver. It cannot be called
    // on a running task.
    rpc DestroyTask(DestroyTaskRequest)  returns (DestroyTaskResponse) {}

    // ListTasks returns a list of summary information of all the tasks the
    // driver is tracking.
    rpc ListTasks(ListTasksRequest) returns (ListTasksResponse) {}

    // InspectTask returns detailed information for the given task
    rpc InspectTask(InspectTaskRequest) returns (InspectTaskResponse) {}

    // TaskStats collects and returns runtime metrics for the given task
    rpc TaskStats(TaskStatsRequest) returns (TaskStatsResponse) {}

    // TaskEvents starts a streaming RPC where all task events emitted by the
    // driver are streamed to the caller.
    rpc TaskEvents(TaskEventsRequest) returns (stream DriverTaskEvent) {}

    // The following RPCs are only implemented if the driver sets the
    // corresponding capability.

    // SignalTask sends a signal to the task
    rpc SignalTask(SignalTaskRequest) returns (SignalTaskResponse) {}

    // ExecTask executes a command inside the tasks execution context
    rpc ExecTask(ExecTaskRequest) returns (ExecTaskResponse) {}
}

message DriverTaskConfigSchemaRequest {}

message DriverTaskConfigSchemaResponse {

    // Spec is the configuration schema for the job driver config stanza
    hashicorp.nomad.plugins.shared.hclspec.Spec spec = 1;
}

message CapabilitiesRequest {}

message CapabilitiesResponse {

    // Capabilities provides a way for the driver to denote if it implements
    // non-core RPCs. Some Driver service RPCs expose additional information
    // or functionality outside of the core task management functions. These
    // RPCs are only implemented if the driver sets the corresponding capability.
    DriverCapabilities capabilities = 1;
}

message FingerprintRequest {}

message FingerprintResponse {


    // Attributes are key/value pairs that annotate the nomad client and can be
    // used in scheduling contraints and affinities.
    map<string, string> attributes = 1;

    // Detected signifies if the necessary dependancies of the drive are met for
    // nominal operation.
    bool detected = 2;

    // Healthy is true if the driver is able to make changes to a task's state
    // ex. Start/Stop
    bool healthy = 3;
}

message RecoverTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;

    // Handle is the TaskHandle returned from StartTask
    TaskHandle handle = 2;
}

message RecoverTaskResponse {}

message StartTaskRequest {

    // Task configuration to launch
    TaskConfig task = 1;

    //NetworkOverride is set if the driver sets network settings and the service ip/port
    //needs to be set differently.
    NetworkOverride network_override = 2;
}

message StartTaskResponse {

    enum Result {
        SUCCESS = 0;
        RETRY = 1;
        FATAL = 2;
    }

    // Result is set depending on the type of error that occured while starting
    // a task:
    //
    //   * SUCCESS: No error occured, handle is set
    //   * RETRY: An error occured, but is recoverable and the RPC should be retried
    //   * FATAL: A fatal error occured and is not likely to succeed if retried
    //
    // If Result is not successful, the DriverErrorMsg will be set.
    Result result = 1;

    // Handle is opague to the client, but must be stored in order to potentially
    // recover the task.
    TaskHandle handle = 2;

    // DriverErrorMsg is set if an error occured
    string driver_error_msg = 3;
}

message WaitTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;
}

message WaitTaskResponse {

    // Result is the exit status of the task
    ExitResult result = 1;
    // Err is set if any driver error occured while waiting for the task
    string err = 2;
}

message StopTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;

    // Timeout defines the amount of time to wait before forcefully killing
    // the task. For example, on Unix clients, this means sending a SIGKILL to
    // the process.
    google.protobuf.Duration timeout = 2;

    // Signal can be set to override the Task's configured shutdown signal
    string signal = 3;
}

message StopTaskResponse {}

message DestroyTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;
}

message DestroyTaskResponse {}

message ListTasksRequest {}

message ListTasksResponse {

    // Tasks includes a list of summary information for each task
    repeated TaskStatus tasks = 1;
}

message InspectTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;
}

message InspectTaskResponse {

    // Task details
    TaskStatus task = 1;

    // Driver details for task
    TaskDriverStatus driver = 2;

    // NetworkOverride info if set
    NetworkOverride network_override = 3;
}

message TaskStatsRequest {

    // TaskId is the ID of the target task
    string task_id = 1;
}

message TaskStatsResponse {

    // Stats for the task
    TaskStats stats = 1;
}

message TaskEventsRequest {}

message SignalTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;

    // Signal is the operating system signal to send to the task. Ex: SIGHUP
    string signal = 2;
}

message SignalTaskResponse {}

message ExecTaskRequest {

    // TaskId is the ID of the target task
    string task_id = 1;

    // Command is the command to execute in the task environment
    repeated string command = 2;

    // Timeout is the amount of time to wait for the command to stop.
    // Defaults to 0 (run forever)
    google.protobuf.Duration timeout = 3;
}

message ExecTaskResponse {

    // Stdout from the exec
    bytes stdout = 1;

    // Stderr from the exec
    bytes stderr = 2;

    // Result from the exec
    ExitResult result = 3;
}

message DriverCapabilities {

    // SendSignals indicates that the driver can send process signals (ex. SIGUSR1)
    // to the task.
    bool send_signals = 1;

    // Exec indicates that the driver supports executing arbitrary commands
    // in the task's execution environment.
    bool exec = 2;

    enum FSIsolation {
        NONE = 0;
        CHROOT = 1;
        IMAGE = 2;
    }
    // FsIsolation indicates what kind of filesystem isolation a driver supports.
    FSIsolation fs_isolation = 3;
}

message TaskConfig {

    // Id of the task, recommended to the globally unique, must be unique to the driver.
    string id = 1;

    // Name of the task
    string name = 2;

    // MsgpackDriverConfig is the encoded driver configuation of the task
    bytes msgpack_driver_config = 3;

    // Env is the a set of key/value pairs to be set as environment variables
    map<string, string> env = 4;

    // Resources defines the resources to isolate
    Resources resources = 5;

    // Mounts is a list of targets to bind mount into the task directory
    repeated Mount mounts = 6;

    // Devices is a list of system devices to mount into the task's execution
    // environment.
    repeated Device devices = 7;

    // User defines the operating system user the tasks should run as
    string user = 8;

    // AllocDir is the directory on the host where the allocation directory
    // exists.
    string alloc_dir = 9;
}

message Resources {

    // RawResources are the resources set for the task
    RawResources raw_resources = 1;

    // LinuxResources are the computed values to set for specific Linux features
    LinuxResources linux_resources = 2;
}

message RawResources {
    int64 cpu = 1;
    int64 memory = 2;
    int64 disk = 3;
    int64 iops = 4;
    repeated NetworkResource networks = 5;
}

message NetworkResource {
    string device = 1;
    string cidr = 2;
    string ip = 3;
    int32 mbits = 4;
    repeated NetworkPort reserved_ports = 5;
    repeated NetworkPort dynamic_ports = 6;
}

message NetworkPort {
    string label = 1;
    int32 value = 2;
}

message LinuxResources {

    // CPU CFS (Completely Fair Scheduler) period. Default: 0 (not specified)
    int64 cpu_period = 1;
    // CPU CFS (Completely Fair Scheduler) quota. Default: 0 (not specified)
    int64 cpu_quota = 2;
    // CPU shares (relative weight vs. other containers). Default: 0 (not specified)
    int64 cpu_shares = 3;
    // Memory limit in bytes. Default: 0 (not specified)
    int64 memory_limit_in_bytes = 4;
    // OOMScoreAdj adjusts the oom-killer score. Default: 0 (not specified)
    int64 oom_score_adj = 5;
    // CpusetCpus constrains the allowed set of logical CPUs. Default: "" (not specified)
    string cpuset_cpus = 6;
    // CpusetMems constrains the allowed set of memory nodes. Default: "" (not specified)
    string cpuset_mems = 7;
}

message Mount {

    // TaskPath is the file path within the task directory to mount to
    string task_path = 1;

    // HostPath is the file path on the host to mount from
    string host_path = 2;

    // Readonly if set true, mounts the path in readonly mode
    bool readonly = 3;
}

message Device {

    // TaskPath is the file path within the task to mount the device to
    string task_path = 1;

    // HostPath is the path on the host to the source device
    string host_path = 2;

    // CgroupPermissions defines the Cgroup permissions of the device.
    // One or more of the following options can be set:
    //  * r - allows the task to read from the specified device.
    //  * w - allows the task to write to the specified device.
    //  * m - allows the task to create device files that do not yet exist.
    //
    // Example: "rw"
    string cgroup_permissions = 3;
}

enum TaskState {
    UNKNOWN = 0;
    RUNNING = 1;
    EXITED = 2;
}

// TaskHandle is created when starting a task and is used to recover task
message TaskHandle {

    // Config is the TaskConfig for the task
    TaskConfig config = 1;

    // State is the state of the task's execution
    TaskState state = 2;

    // DriverState is the encoded state for the specific driver
    bytes driver_state = 3;
}

// NetworkOverride contains network settings which the driver may override
// for the task, such as when the driver is setting up the task's network.
message NetworkOverride {

    // PortMap can be set to replace ports with driver-specific mappings
    map<string,int32> port_map = 1;

    // Addr is the IP address for the task created by the driver
    string addr = 2;

    // AutoAdvertise indicates whether the driver thinks services that choose
    // to auto_advertise_addresses should use this IP instead of the host's.
    bool auto_advertise = 3;
}

// ExitResult contains information about the exit status of a task
message ExitResult {

    // ExitCode returned from the task on exit
    int32 exit_code = 1;

    // Signal is set if a signal was sent to the task
    int32 signal = 2;

    // OomKilled is true if the task exited as a result of the OOM Killer
    bool oom_killed = 3;

}

// TaskStatus includes information of a specific task
message TaskStatus {
    string id = 1;
    string name = 2;

    // State is the state of the task's execution
    TaskState state = 3;

    // StartedAt is the timestamp when the task was started
    google.protobuf.Timestamp started_at = 4;

    // CompletedAt is the timestamp when the task exited.
    // If the task is still running, CompletedAt will not be set
    google.protobuf.Timestamp completed_at = 5;

    // Result is set when CompletedAt is set.
    ExitResult result = 6;
}

message TaskDriverStatus {

    // Attributes is a set of string/string key value pairs specific to the
    // implementing driver
    map<string, string> attributes = 1;
}

message TaskStats {

    // Id of the task
    string id = 1;

    // Timestamp for which the stats were collected
    google.protobuf.Timestamp timestamp = 2;

    // AggResourceUsage is the aggreate usage of all processes
    TaskResourceUsage agg_resource_usage = 3;

    // ResourceUsageByPid breaks the usage stats by process
    map<string, TaskResourceUsage> resource_usage_by_pid = 4;
}

message TaskResourceUsage {

    // CPU usage stats
    CPUUsage cpu = 1;

    // Memory usage stats
    MemoryUsage memory = 2;
}

message CPUUsage {

    double system_mode = 1;
    double user_mode = 2;
    double total_ticks = 3;
    uint64 throttled_periods = 4;
    uint64 throttled_time = 5;
    double percent = 6;

    enum Fields {
        SYSTEM_MODE = 0;
        USER_MODE = 1;
        TOTAL_TICKS = 2;
        THROTTLED_PERIODS = 3;
        THROTTLED_TIME = 4;
        PERCENT = 5;
    }
    // MeasuredFields indicates which fields were actually sampled
    repeated Fields measured_fields = 7;
}

message MemoryUsage {
    uint64 rss = 1;
    uint64 cache = 2;
    uint64 max_usage = 3;
    uint64 kernel_usage = 4;
    uint64 kernel_max_usage = 5;

    enum Fields {
        RSS = 0;
        CACHE = 1;
        MAX_UASGE = 2;
        KERNEL_USAGE = 3;
        KERNEL_MAX_USAGE = 4;
    }
    // MeasuredFields indicates which fields were actually sampled
    repeated Fields measured_fields = 6;
}

message DriverTaskEvent {

    // TaskId is the id of the task for the event
    string task_id = 1;

    // Timestamp when the event occured
    google.protobuf.Timestamp timestamp = 2;

    // Message is the body of the event
    string message = 3;

    // Annotations allows for additional key/value data to be sent along with the event
    map<string,string> annotations = 4;
}