policy/frameworks/cluster/controller/config.zeek¶
-
ClusterController
¶
Configuration settings for the cluster controller.
- Namespace
ClusterController
- Imports
Summary¶
Redefinable Options¶
The controller’s connect retry interval. |
|
The fallback listen address if |
|
The fallback listen port if |
|
An optional custom output directory for the controller’s stdout and stderr logs. |
|
The network address the controller listens on. |
|
The network port the controller listens on. |
|
The name of this controller. |
|
The timeout for request state. |
|
|
The role of this process in cluster management. |
The controller’s stderr log name. |
|
The controller’s stdout log name. |
|
The controller’s Broker topic. |
Functions¶
Returns a |
|
Returns a |
Detailed Interface¶
Redefinable Options¶
- ClusterController::connect_retry¶
-
The controller’s connect retry interval. Defaults to a more aggressive value compared to Broker’s 30s.
- ClusterController::default_address¶
-
The fallback listen address if
ClusterController::listen_address
remains empty. Unless redefined, this uses Broker’s own default listen address.
- ClusterController::default_port¶
-
The fallback listen port if
ClusterController::listen_port
remains empty.
- ClusterController::directory¶
-
An optional custom output directory for the controller’s stdout and stderr logs. Agent and controller currently only log locally, not via the data cluster’s logger node. (This might change in the future.) This means that if both write to the same log file, the output gets garbled.
- ClusterController::listen_address¶
-
The network address the controller listens on. By default this uses the value of the ZEEK_CONTROLLER_ADDR environment variable, but you may also redef to a specific value. When empty, the implementation falls back to
ClusterController::default_address
.
- ClusterController::listen_port¶
-
The network port the controller listens on. Counterpart to
ClusterController::listen_address
, defaulting to the ZEEK_CONTROLLER_PORT environment variable.
- ClusterController::name¶
-
The name of this controller. Defaults to the value of the ZEEK_CONTROLLER_NAME environment variable. When that is unset and the user doesn’t redef the value, the implementation defaults to “controller-<hostname>”.
- ClusterController::request_timeout¶
-
The timeout for request state. Such state (see the
ClusterController::Request
module) ties together request and response event pairs. The timeout causes its cleanup in the absence of a timely response. It applies both to state kept for client requests, as well as state in the agents for requests to the supervisor.
- ClusterController::role¶
- Type
- Attributes
- Default
ClusterController::Types::NONE
- Redefinition
from policy/frameworks/cluster/agent/main.zeek
=
:ClusterController::Types::AGENT
- Redefinition
from policy/frameworks/cluster/controller/main.zeek
=
:ClusterController::Types::CONTROLLER
The role of this process in cluster management. Agent and controller both redefine this. Used during logging.
- ClusterController::stderr_file¶
-
The controller’s stderr log name. Like
ClusterController::stdout_file
, but for the stderr stream.
- ClusterController::stdout_file¶
-
The controller’s stdout log name. If the string is non-empty, Zeek will produce a free-form log (i.e., not one governed by Zeek’s logging framework) in Zeek’s working directory. If left empty, no such log results.
Note that the controller also establishes a “proper” Zeek log via the
ClusterController::Log
module.
- ClusterController::topic¶
-
The controller’s Broker topic. Clients send requests to this topic.
Functions¶
- ClusterController::endpoint_info¶
- Type
Returns a
Broker::EndpointInfo
record describing the controller.
- ClusterController::network_info¶
- Type
Returns a
Broker::NetworkInfo
record describing the controller.