Is awkward supposed to have a negative connotation here? The author gives no context, so I’m not sure. I thought those all looked like pretty reasonable questions when doing any kind of architecture design.
I think the intention (particularly with the use of “bandwagon” in the title) is to draw out the difficult parts of running separate services when compared to extracting code into libraries.
How do you plan to monitor your microservices?
How do you plan to trace the interactions between different microservices in a production environment?
These two in particular bring out some concerns which get more important as you take your application towards a distributed architecture. Another, which I think sits well with these, is “Does any data need to be kept consistent across services?” The jump from “just throw a transaction round it” to dealing with partial failure is a big one.
Is awkward supposed to have a negative connotation here? The author gives no context, so I’m not sure. I thought those all looked like pretty reasonable questions when doing any kind of architecture design.
I think the intention (particularly with the use of “bandwagon” in the title) is to draw out the difficult parts of running separate services when compared to extracting code into libraries.
These two in particular bring out some concerns which get more important as you take your application towards a distributed architecture. Another, which I think sits well with these, is “Does any data need to be kept consistent across services?” The jump from “just throw a transaction round it” to dealing with partial failure is a big one.