Another update: with around 1.5TB (it’s an 8TB drive) of free space left, the HDD is now busy 100% of the time while the node service is running. It even sometimes fails to properly stop the node service, probably due to the heavy HDD usage.
I have reduced my allocated storage in order to stop any ingress for now and will replace the HDD asap and call it a day (aka sell the SMR drive to someone who only needs archive storage). Lesson learned, feel free to quote me on that whenever someone considers getting an SMR HDD 
/edit: also keep in mind that is a 3.5in enterprise class drive, not some 2.5in consumer external HDD (which might behave even worse)