Closed mkardous-silabs closed 8 months ago
may I know how to trigger event?
On the Linux app, i'm not sure how to trigger the event. Test event has a way (TCSWTCH...) On a silabs boards, BTN1 triggers the switch events
So I just tried the following steps to reproduce, after compiling the relevant binaries from SHA 4838490b16f8f36b8d7ae3e62c6e83371f4291d3:
Terminal 1:
% ./out/debug/standalone/chip-all-clusters-app
Terminal 2:
% ./out/debug/standalone/chip-tool pairing code 1 MT:-24J042C00KA0648G00
% ./out/debug/standalone/chip-tool interactive start
>>> any subscribe-event-by-id 0xFFFFFFFF 0xFFFFFFFF 0 10 1 0xFFFF
Terminal 3:
% echo '{ "Name": "SoftwareFault" }' > /tmp/chip_all_clusters_fifo_200
In terminal 1 I see:
[1704929918822] [200:7919384] [EVL] LogEvent event number: 0x0000000000000005 priority: 1, endpoint id: 0x0 cluster id: 0x0000_0034 event id: 0x0 Epoch timestamp: 0x0000018CF5BDFF66
After a short delay in terminal 2 I see:
[1704929924908] [218:7918881] [DMG] ReportDataMessage =
[1704929924908] [218:7918881] [DMG] {
[1704929924908] [218:7918881] [DMG] SubscriptionId = 0xc0936fe8,
[1704929924908] [218:7918881] [DMG] EventReportIBs =
...
[1704929924909] [218:7918881] [TOO] Endpoint: 0 Cluster: 0x0000_0034 Event 0x0000_0000
[1704929924909] [218:7918881] [TOO] Event number: 5
[1704929924909] [218:7918881] [TOO] Priority: Info
[1704929924909] [218:7918881] [TOO] Timestamp: 1704929918822
[1704929924909] [218:7918881] [TOO] SoftwareFault: {
[1704929924909] [218:7918881] [TOO] Id: 200
[1704929924909] [218:7918881] [TOO] Name: 200
[1704929924909] [218:7918881] [TOO] FaultRecording: 576564204A616E2031302031383A33383A33382032303234
[1704929924909] [218:7918881] [TOO] }
locally when running the below against all-cluster-app ./out/linux-x64-chip-tool/chip-tool interactive start pairing onnetwork-long 1 20202021 3840 unittesting subscribe-event test-event 0 20 1 1 unittesting test-emit-test-event-request 1 2 false 1 1
I am able to receive the test event whenever maximum period is reached if trigger test-emit-test-event-request periodically.
I just tried the following also:
% ./out/debug/standalone/chip-tool interactive start
>>> any subscribe-event-by-id 0xFFFFFFFF 0xFFFFFFFF 0 10 1 0xFFFF --is-urgent true
>>> accesscontrol write acl '[{"privilege": 5, "authMode": 2, "subjects": null, "targets": null}]' 1 0
I immediately get subscription reports:
[1704990075471] [54161:8230327] [TOO] Endpoint: 0 Cluster: 0x0000_001F Event 0x0000_0000
[1704990075471] [54161:8230327] [TOO] Event number: 5
[1704990075471] [54161:8230327] [TOO] Priority: Info
[1704990075471] [54161:8230327] [TOO] Timestamp: 209453
[1704990075471] [54161:8230327] [TOO] AccessControlEntryChanged: {
etc.
The reason why the event reports were not being sent out was because the subscription request did not have the --is-urgent
flag which requests the report to be as soon as possible.
Reproduction steps
any subscribe-event-by-id 0xFFFFFFFF 0xFFFFFFFF 0 10 <nodeid> 0xFFFF
Attribute reporting works as expected but not events.
Bug prevalence
Systematically
GitHub hash of the SDK that was being used
6b21935fde
Platform
core
Platform Version(s)
No response
Type
Core SDK Crash
Anything else?
Device Side logs
Chip-Tool logs