Hello,

Those graphs are indeed correct, but we just don't have any way to tell you what your application did that resulted in that, similarly to how your power company can tell you how much power you used, but they couldn't tell you what you did with the power.

But the situation with network traffic is different from energy. It is possible to monitor outgoing requests from the private network to the public network. As for the graph, if I consult it in another location, the information about the day and time is different. So the graph is probably wrong. As I said, there is no way the application can consume this absurd amount of data, especially outgoing data.

The increase in consumption only occurred on the day the billing cycle was due and for a few more days. The application did not have any new deployments or changes to increase traffic.

> It is possible to monitor outgoing requests from the private network to the public network...
It is, but we do not monitor that, what our users do with the internet is none of our business, so the power company analogy stands true in this context.

> As for the graph, if I consult it in another location, the information about the day and time is different.
Those two graphs represent different formats of data, the metrics graph found within the service is historical on how much network is used at a point in time, and the project metrics are cumulative.

> there is no way the application can consume this absurd amount of data.
I'm sorry but it has, I agree this can look strange from a user's perspective, but the metrics are correct, I've double checked with our raw metrics.

> The increase in consumption only occurred on the day the billing cycle was due and for a few more days.
Purely a coincidence, It also looks like you have a fairly substantial memory leak, so I wouldn't say that network usage is out of the question.

Going forward I would recommend implementing some form of tracing so you can get a better understanding of what your code is doing.

Brody, thanks for the information.

Is there any way to monitor outgoing traffic?

If the project metrics graph is cumulative, why do CPU, RAM and Volume have values ​​that increase and decrease over the course of days?

I made some changes to my code to try to monitor outgoing requests

> Is there any way to monitor outgoing traffic?

Platform natively, no.

> If the project metrics graph is cumulative, why do CPU, RAM and Volume have values ​​that increase and decrease over the course of days?

Sorry, I misspoke slightly, I'm referring to the network graph, the network graph is cumulative.

Could you recommend a tool to monitor outgoing traffic?

I'm sorry but I have no experience in that regard.