A compositional approach to monitoring distributed systems

Using status messages in the distributed test architecture

“…In this context, in order for a subsystem to report the global state it must know the current state of every other subsystem. It may thus be simpler to have a separate status message S i for each port p i where the message S i can be input at p i and the output produced in response to S i is the current state of M i and this is similar to the notion of compositional monitoring [35]. It is relatively straightforward to adapt the notion of weakly synchronized to this case: we require that for any two consecutive transitions tt ′ in which the input from t ′ is applied at port p i , either tt ′ is synchronisable or t leads to a change in state of the component M i .…”

“…We show that controllability and observability problems can always be overcome when there are such status messages that are know to have been implemented correctly. The notion of a status message relates to the concept of a monitor in debugging and there has been much work on the implementation of monitors with the aim of obtaining the global state of a system and achieving this in a non-invasive manner (see, for example, [1,8,15,24,35]). Here, however, the observation of the state of the SUT is made locally and this corresponds to the second type of status message that we consider, in which the state of the system is output at the port that sent the status message.…”

Controllability Problems in MSC-Based Testing

“…For example, the separate testers may be able to communication through coordination channels and testing might use a specific technology such as monitoring systems [49]. In this paper, we discuss controllability problems based on three notions of observability.…”

Oracles for Distributed Testing

“…In addition, there are approaches in which the testers communicate in order to determine the global trace that occurred (see, for example, [25], [26]). There has also been a significant amount of work on monitoring, in which we wish to determine the global state of the SUT (see, for example, [27], [28], [29], [30], [31]). In contrast to these, we are concerned with black-box testing and we are interested in conformance relations that capture the observational power of potential users.…”