Hi,
On a cluster composed by 8 DMAs in 1+1 FO configuration, the SLDMS.txt log of all the agents is flooded with these error messages (see also the screenshot below):
2021/07/07 16:31:40.605|SLDMS.exe 10.0.1947.7652|7656|8472|CRequest::Request|ERR|0|Remote Request for -DMA- on -VT_I4 : 18007- failed. Invalid data. (hr = 0x80040221)
Type 0/148/0
MESSAGE: Invalid data.
VALUE 1: VT_ARRAY|VT_UI4 (8) : 18006;2175;18006;2180;4294967295;4294967295;7;4294967295
The DMAIDs appearing in the messages are relative to all the other 7 DMAs. A similar log when we check the SLDMS.txt from another agent of the cluster.
What does these errors mean? Some request to the other DMA has invalid data but which request and which data? How can we fix it?
Dataminer release is 10.0 CU9
It seems something is going wrong on type 148 (NT_REGISTER_SLA), which gets called when an SLA has changes to the alarms it listens to to calculate its SLA.
The target of this call is SLDataMiner which is probably the one throwing the INVALID_DATA. At first glance the data in VALUE 1 seems to be ok, but there is always the possibility that SLDataminer is expecting something else. There might be more info logged in the SLDataminer.txt logging or in the log of the service (the first 2 numbers in the VALUE 1 line form the DMAID/SID).
Additional info on the values in VALUE1:
– service dmaid
– service eid
– element dmaid
– element eid
– notify alarm pid
– notify element state pid
– element type
– element hosting dma
The alarm pid/element pid is 4294967295 == -1 and appears to indicate that the request is to unsubscribe from alarm and element updates
First look appears to indicate that the “invalid data” error is most likely generated because the subscription was not active before.
If these errors occur frequently, it might be good to open a task for further investigation