This message was deleted.
# generals
sparse-intern-71089
02/24/2021, 5:07 AMThis message was deleted.
b
billowy-army-68599
02/24/2021, 5:27 AM
you're using the Go SDK, and it's the way Go is designed
https://blog.golang.org/error-handling-and-go
the error itself is the result of the call to the cloud provider API, it gets stored as an error type. So let's say you try and create a resource, and something goes wrong - you need to check the error explicitly (as per the go design)
c
clever-cartoon-41433
02/24/2021, 5:47 AMI've used go before, my question was more about what actual api calls were happening as my program is run by pulumi
clever-cartoon-41433
02/24/2021, 5:48 AMIt's not creating these resources one by one as I specify them, right? There's a whole bunch of concurrent scheduling going on when I view the actual pulumi up output, that isn't actually enumerated in my program. Which made me think that the point of the program was just to specify a map for infrastructure creation. In which case, I didn't see what errors could arise from adding another node to the map
w
white-balloon-205
02/24/2021, 6:54 AM
You are right that there are many errors that might not be reported synchronously - like failures from the cloud provider trying to create the resource you specify. But there are also errors that might happen synchronously, for example the inability to register the resource or even some classes of early validation of resource inputs. In general though, it is rare to have synchronous errors from NewResource calls.
3 Views