High volume of outbound data on inactive db app.

1

I have a phoenix app that I stood up with a free tier postgres.

The app itself is nothing more than a mix phx.new app with nothing added. Its been 10 days since I stood it up and did nothing with it.

The app its self has only 712 MB of total transfer data used, where as the db app is already up to 5 gigs outbound traffic. I thought that was high for an app that has no users and no real data model.

that got me wondering. so I started to look at the logs and noticed.

2022-03-18T20:25:35.983 app[57273b35] sea [info] keeper | 2022-03-18T20:25:35.983Z INFO cmd/keeper.go:1504 our db requested role is master

2022-03-18T20:25:35.984 app[57273b35] sea [info] keeper | 2022-03-18T20:25:35.983Z INFO cmd/keeper.go:1542 already master

2022-03-18T20:25:36.001 app[57273b35] sea [info] keeper | 2022-03-18T20:25:36.000Z INFO cmd/keeper.go:1675 postgres parameters not changed

This log is repeating every second for what looks like forever. Could this be the cause of the outbound traffic being used up on my postgres app?

2

Postgres clusters are very chatty. We include their traffic in those outbound data calculations, but we do not bill for it.

5GB over 10 days is not actually very much. That’s 6KB/s. But we’re not billing for it!

1 Like
3

And what of the rolling logs about role?

Also this is not a cluster, I don’t think as its just one app.

4

Ah! Even single node Postgres is a cluster in our infrastructure. You can add nodes any time to make it HA (and, coincidentally, start paying us money :slight_smile: ).

The logs are normal, that’s the cluster supervisor checking its status against a shared Consul and making sure it’s still the leader. Since it’s only one node, it’s definitely still the leader.

3 Likes
5

Thank you.

1 Like