Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file
Package Dependencies
System Dependencies
Dependant Packages
Launch files
Messages
Services
Plugins
Recent questions tagged ros2_medkit_opcua at Robotics Stack Exchange
Package Summary
| Version | 0.7.2 |
| License | Apache-2.0 |
| Build type | AMENT_CMAKE |
| Use | RECOMMENDED |
Repository Summary
| Description | |
| Checkout URI | https://github.com/selfpatch/ros2_medkit.git |
| VCS Type | git |
| VCS Version | main |
| Last Updated | 2026-11-22 |
| Dev Status | DEVELOPED |
| Released | RELEASED |
| Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- mfaferek93
Authors
- mfaferek93
ros2_medkit_opcua
Gateway plugin that bridges OPC-UA capable PLCs (OpenPLC, Siemens S7, Beckhoff, Allen-Bradley, etc.) into the SOVD entity tree. Enables unified diagnostics for mixed ROS 2 + industrial PLC deployments through a single REST API, with PLC alarms routed to ros2_medkit_fault_manager and numeric PLC values optionally bridged to ROS 2 std_msgs/Float32 topics.
Follows the same plugin pattern as ros2_medkit_graph_provider: implements GatewayPlugin + IntrospectionProvider against the get_routes() plugin API, loaded at runtime by ros2_medkit_gateway via dlopen.
What it does
- Connects to any OPC-UA capable PLC server over
opc.tcp - Emits SOVD entities (area, component, apps) from a YAML-driven node map, or
discovers them automatically by browsing the live address space
(
auto_browse, see below) - zero node-map config required - Exposes PLC values as the
x-plc-datavendor collection - Ships read-only by default: the shipped binary contains no compiled path that writes a value or acknowledges a condition, and exports nothing from the OPC UA stack (see Read-only and write-capable builds)
- In a write-capable build, allows writing setpoints via
x-plc-operationswith type-aware coercion and range validation - Reports the connection state, poll metrics and the build’s write surface via
x-plc-status - Maps threshold-based PLC alarms to SOVD faults on the owning entity
- Optionally publishes numeric PLC values to ROS 2
std_msgs/Float32topics
Architecture
OPC-UA (TCP :4840)
PLC Runtime <─────────────────────────> OPC-UA Plugin (.so)
IEC 61131-3 program │
Cyclic execution (100ms) │ Polls all configured nodes
Variables exposed as OPC-UA nodes │ Maps to SOVD entity tree
│ Alarm thresholds -> fault reporting
│
▼
ros2_medkit Gateway
REST API :8080
│
┌──────┴──────┐
│ │
SOVD REST Fleet Gateway
(direct) (aggregation)
│ │
Dashboard Multi-device view
The plugin connects to any PLC with an OPC-UA server over TCP. No ROS 2 dependency between the plugin and the PLC - communication is pure OPC-UA. The plugin is loaded by the gateway at runtime via dlopen() and registers vendor REST endpoints for PLC data access and control.
SOVD Entity Model
The plugin creates a hierarchical entity tree from a YAML node map configuration:
Area: plc_systems
└── Component: openplc_runtime
├── App: tank_process
│ Data: tank_level (mm), tank_temperature (C), tank_pressure (bar)
│ Faults: PLC_HIGH_TEMP, PLC_LOW_LEVEL, PLC_OVERPRESSURE
├── App: fill_pump
│ Data: pump_speed (%)
│ Operations: set_pump_speed
└── App: drain_valve
Data: valve_position (%)
Operations: set_valve_position
Asset identity (x-medkit.identity)
The plugin auto-populates the PLC component’s asset-identity nameplate from the
live server and serves it as x-medkit.identity on the component (see
docs/api/rest.rst for the JSON shape). Two sources are read, best-effort:
-
Server/ServerStatus/BuildInfo(present on every compliant server):ManufacturerName,ProductName,SoftwareVersion,BuildNumber(carried as theextra.buildNumberkey). - The OPC UA DI companion nameplate (
http://opcfoundation.org/UA/DI/, only when the server exposes that namespace):Manufacturer,Model,SerialNumber,HardwareRevision,SoftwareRevisionfrom the first device underObjects/DeviceSetthat carries identification properties. DI values are device-specific and win over the server-level BuildInfo per field.
The read happens once per session - on the first introspect after a connect - and is refreshed after every reconnect: the cached nameplate is tied to the OPC UA session, so a PLC reboot, firmware update, or device swap is picked up on the next discovery refresh instead of latching the first read forever.
Trust-gated precedence. How the live nameplate ranks against a
hand-authored manifest identity: block depends on whether the session
authenticates the server:
- Secured channel (
security_mode: Sign/SignAndEncrypt) andreject_untrusted: true: the component is tagged with the protocol sourceopcua, which outranksmanifestin the identity merge - an authenticated live read overrides stale manifest fields. - Anything else (
SecurityPolicy=None, orreject_untrusted: false/ accept-any cert): the nameplate is spoofable by a rogue endpoint, so the component keeps the genericpluginsource, which ranks belowmanifest- the read fills fields the operator left empty but never overrides them.
Per-field _provenance records opcua in both cases, so consumers always see
where a value was read even when it merged with low authority.
File truncated at 100 lines see the full file
Changelog for package ros2_medkit_opcua
0.7.2 (2026-11-19)
- No changes. Version aligned with the rest of the stack
0.7.1 (2026-11-18)
-
Breaking: read-only is a property of the build, not a setting.
The default build and the published images carry no controller write
path: a write the build does not carry is refused with
501,POST /apps/{id}/x-plc-operations/{op}is not registered and answers404, andacknowledge_faultandconfirm_faultare refused.x-plc-statushas a new fieldwrite_capable. A warning is logged only for a write setting someone wrote. A write-capable plugin is built with-DMEDKIT_OPCUA_READ_ONLY=OFF(#657) - Node-map points take
raise_after_samplesandraise_after_ms, so a fault is raised only after the condition has held (#767) - The plugin rescans while it is disconnected and heals
PLC_COMMS_LOSTon connect.discovery.interval_ssets the rescan period: unset means 30 s, and0turns the rescan off. The 0.7.0 README example setinterval_s: 0, so a configuration copied from it never rescans (#660) - The plugin binds to the ApplicationUri of the first server it
reaches and drops a session with any other server, also at the same
address. Without
endpoint_urlthe binding is stored indiscovery.binding_file(default/var/lib/ros2_medkit/opcua/binding, environmentOPCUA_DISCOVERY_BINDING_FILE). To replace a PLC, delete that file and restart the gateway. Withendpoint_urla restart is enough (#660) - In config-less mode the component gets a new id from the device nameplate once a session comes up (#660)
- The plugin's clears keep correlated symptoms:
DELETE /{entity}/faults/{code}on an OPC UA entity and thePLC_COMMS_LOSTclear setskip_correlation_auto_clear. 0.7.0 cleared the symptoms too (#660) - New key
fault_service_timeout_ms, default 5000 (#660) - The session is ended on a closed connection even when the channel still reads open, and it is closed before its event contexts are released (#774). Subscription mode is serviced without event alarms configured (#766)
- The report and the clear of one fault are handed over in the order they were worked out, and the fault-service clients are created, used and destroyed on the plugin's own threads
- Contributors: \@bburda, \@mfaferek93, \@ahmedsleem109
0.7.0 (2026-08-27)
- Config-less discovery. A read-only network scan finds the OPC UA
server instead of requiring its endpoint up front
(#509),
auto_browsewalks the address space recursively and builds the SOVD tree from it (#510), and identity, writability and fault triggers are read from the device itself rather than declared in a node map (#544) - Native OPC UA Alarms and Conditions become medkit faults with no
configuration
(#511). Alarms
are routed to distinct faults by message substring
(#506),
severityis accepted as an alias inevent_alarms(#501), and events that are not conditions are dropped from the alarm path instead of being reported as alarms (#552) - Asset identity is populated from the device nameplate, including
order_code(#492, #499) -
PLC_COMMS_LOSTis raised when the connection to the PLC drops, so a lost link is a reported fault rather than an absence of data (#507) - Security and reconnect hardening: a secure connection profile, replay of alarm state after a reconnect, and correct handling of several simultaneous alarms (#485)
- Fixed a data race between the REST thread and the poll thread on the pending-report buffer that crashed the gateway with a SIGSEGV (#519, #521)
- Fault detection moved onto the shared
ros2_medkit_fault_detectionevaluator, and the node map enforces global fault-code uniqueness so two rules cannot claim the same code (#486) - This package is still excluded from the rosdistro binary release pending vendoring of open62541pp (#366)
- Contributors: \@bburda, \@mfaferek93
File truncated at 100 lines see the full file