Releases: wirepas/gateway
Releases · wirepas/gateway
Release 1.4.6
Main differences from v1.4.5
Fixes
- Transport: Simplify setStatus mechanism #292
- Status: Optimize the sending of async response with id=0 #295
- LG-71 Respond to SetScratchpadTargetAndActionReq with invalid parameters #294
- Fix boolean parameter reading from dbus message #284
- Fix missing sd_bus_message_unref calls in sink service in #288
- Validate max network address for sink #280
Enhancements
- Graceful shutdown support for sink service #287
- Update base container images and use UPB implementation of protobuf on transport service #285
- Enable clearing of sink scratchpad #276
Full Changelog: v1.4.5...v1.4.6
v1.4.6 release candidate 2
Main differences from v1.4.6-rc.1
Fixes
- Transport: Simplify setStatus mechanism #292
- Status: Optimize the sending of async response with id=0 #295
- LG-71 Respond to SetScratchpadTargetAndActionReq with invalid parameters #294
Full Changelog: v1.4.6-rc.1...v1.4.6-rc.2
v1.4.6 release candidate 1
Main differences from v1.4.5
Fixes
- Fix boolean parameter reading from dbus message #284
- Fix missing sd_bus_message_unref calls in sink service in #288
- Validate max network address for sink #280
Enhancements
- Graceful shutdown support for sink service #287
- Update base container images and use UPB implementation of protobuf on transport service #285
- Enable clearing of sink scratchpad #276
Full Changelog: v1.4.5...v1.4.6rc1
Release 1.4.5
Main difference from 1.4.4
New features
- Black hole: add an option to stop stack in case of black hole instead of increasing sink cost
- Add a mechanism to limit the rate at which gateway is publishing on the broker.
- It is disabled by default and can be enabled with option --mqtt_rate_limit_pps (or with global variable WM_SERVICES_MQTT_RATE_LIMIT_PPS)
- Add a backoff mechanism to randomize the interval to reconnect to broker
- It is useful to avoid all gateways reconnecting at same time in case broker is down
Enhancements
- Upgrade c-mesh-api version for sink-service to support bigger MTU than 102
- Display the error code from transport in places where it was missing
Fixes
- Fix how black hole is enabled
v1.4.5 release candidate 3
Compare to v1.4.5rc2:
- Add a mechanism to limit the rate at which gateway is publishing on the broker.
- It is disabled by default and can be enabled with option --mqtt_rate_limit_pps (or with global variable WM_SERVICES_MQTT_RATE_LIMIT_PPS)
- Add a backoff mechanism to randomize the interval to reconnect to broker
- It is useful to avoid all gateways reconnecting at same time in case broker is down
v1.4.5 release candidate 2
Compare to v1.4.5rc1:
- Upgrade c-mesh-api version for sink-service to include a fix for dynamic mtu
- Display the error code from transport in places where it was missing
- Black hole: add an option to stop stack in case of black hole instead of increasing sink cost
v1.4.5-rc.1
Compare to v1.4.4:
- Upgrade c-mesh-api version for sink-service to support bigger MTU than 102
- Fix how black hole is enabled
Release 1.4.4
Main difference from 1.4.3
New features
- No new features compare to 1.4.3
Enhancements
- Increase default timeout for sink communication from 60s to 120s (to cover long image processing)
Fixes
- Add a cache in transport service for sink configs to avoid sending an uncomplete config in gateway status.
- Fix exception handling in RTC service and add possibility to specify the ntp server to use
- Fix logging of dbus-print-service
v1.4.4 release candidate 2
Changes from 1.4.4 rc 1
- Fix a bug in gateway status generation when connection is back after a disconnect
- Increase default timeout for sink communication from 60s to 120s (to cover long image processing)
v1.4.4 release candidate 1
Compare to v1.4.3:
- Add a cache in transport service for sink configs to avoid sending an uncomplete config in gateway status.
- Fix exception handling in RTC service and add possibility to specify the ntp server to use
- Fix logging of dbus-print-service