-
Type: Bug
-
Status: Resolved (View Workflow)
-
Priority: Medium
-
Resolution: Fixed
-
Affects Version/s: None
-
Fix Version/s: VOLTHA v2.10
-
Component/s: Testing
-
Labels:
-
Story Points:2
-
Epic Link:
-
Sprint:VOLTHA 2.10 - Sprint 3
There are 3 issues: * Set Hw Component Info
TC flow: Get Hw Component Info, then do Set Hw Component Info by updating the component_name. Test expects result as StatusCode.UNIMPLEMENTED
BBSim: Working fine. The test is in-line with the bbsim implementation here. https://github.com/opencord/bbsim/blob/master/internal/bbsim/dmiserver/dmi_hw_mgmt.go#L446
Physical Pod: Failing with statement: {{'{'status': 'ERROR_STATUS', 'reason': 'INVALID_PARAMS', 'reason_detail': 'modifying component names is not supported'}' }}Also, from the adtran-olt-device-manager logs:
default/adtran-olt-device-manager-68f58fb5c8-7qjwq[adtran-olt-device-manager]: IH 2022-02-10 10:53:07.650 ERROR SetHWComponentInfo: modifying component names is not supported <protobuf.go:704>
default/adtran-olt-device-manager-68f58fb5c8-7qjwq[adtran-olt-device-manager]: IH 2022-02-10 10:53:07.650 DEBUG SetHWComponentInfo: req[device_uuid:<uuid:"f6af191f-adc4-5968-aee1-32391dd2b63e" > component_uuid:<uuid:"3d7dc791-179e-5b0d-b69a-02b0cb12b780" > component_name:"cpu 0/1" changes:<name:"new-value" > ] response[status:ERROR_STATUS reason:INVALID_PARAMS reason_detail:"modifying component names is not supported" ] <protobuf.go:706>
Verified manually, setting/modifying other fields like alias is working fine.
Solution: Keep it as is for BBSim, update the test trying set for both component name and some field with the expected output results for physical pod.
- SetLoggingEndpoint
a. TC flow: Do SetLoggingEndpoint, verify this by performing GetLoggingEndpoint and finally reset by again calling SetLoggingEndpoint but this time passing logging_enpoint value as '' (empty quotes). The issue is during resetting with empty quotes. This TC does not run on BBSim
b. On physical-pod the error is: {{Unknown macro: {'status'}}}
default/adtran-olt-device-manager-68f58fb5c8-7qjwq[adtran-olt-device-manager]: CM 2022-02-10 10:53:07.811 ERROR invalid endpoint: error[empty endpoint] <protobuf.go:155>
Solution: set both endpoint and protocol to an empty value.
- GetManagedDevice
a. TC Flow: This issue happens during the Suite Teardown where after calling StopManagingDevice, the system is verified through GetManagedDevice output. Test expect empty response.
b. BBSim: Working fine. In bbsim it checks for uuid and if it is empty, it send empty response. https://github.com/opencord/bbsim/blob/master/internal/bbsim/dmiserver/dmi_hw_mgmt.go#L557
c. Physical Pod: Fails because in the response we getting status field:
jenkins@Server4:~/root/workspace/verify_dt-berlin-pod-gpon-adtran_Default_DT_voltha_master_dmi$ ./grpcurl -v --plaintext $DM_IP dmi.NativeHWManagementService.GetManagedDevices
Resolved method descriptor:
rpc GetManagedDevices ( .google.protobuf.Empty ) returns ( .dmi.ManagedDevicesResponse );
Request metadata to send:
(empty)
Response headers received:
content-type: application/grpc
Response contents:
{
"status": "OK_STATUS"
}
Response trailers received:
(empty)
Sent 0 requests and received 1 response
Solution: all responses (unless the RPC in not implemented) should have a field status.
# | Subject | Branch | Project | Status | CR | V |
---|---|---|---|---|---|---|
30966,4 | [VOL-4615] Fixed DMI tests | master | voltha-system-tests | Status: MERGED | +2 | +1 |
30967,1 | [VOL-4615] Added status field in GetManagedDevices response | master | bbsim | Status: MERGED | +2 | +1 |