why the same service has two nodes on the topology?

[gaoyanyu]

backgroud:

I use the skywalking-scala plugin in my scala project, and trace id produced in the process 'serviceA(scala) -> serviceB(python)' is like this:

trace id produced in the process 'serviceB(python) -> serviceC(python)' is like this:

their format is different, is this normal?
And the topolygy is like this:

my question is why serviceB has two nodes on the topology?

[jxnu-liguobin]

Multiple instances caused by service restart?

[gaoyanyu]

Multiple instances caused by service restart?

This may be caused by the incorrect transmission of the trace IDs on the client and server sides of the grpc request. Does the skywalking-scala plugin implement a grpc interceptor?

[jxnu-liguobin]

Only supports zio grpc and cannot share the same plugin directory with Java grpc service. Because zio grpc does not have a correct server interceptor. see scalapb/zio-grpc#501.

If you are not using zio grpc, Scala cannot guarantee the correct passing of traceids and must implement a series of thread pooling operations such as executors.