How do I follow my instance's resource usage?

Is this instance cramped, or is it sleeping on a machine far too big for it? Hodiploy's "Resource usage" screen answers with measurements rather than impressions, over the last hour as well as the last month.

Where it happens

1. Log in to Hodiploy from your Client Area
2. On the instance's row, open the "Actions" menu, then "Resource usage"
3. At the top of the window, pick the period: 1 h, 24 h, 7 d or 30 d

What you see

The curves separate the instance's two containers: Odoo in purple and the database in amber. A good many questions resolve themselves just by seeing which of the two is climbing.

Processor

The share of a core actually used, where 100% is one core fully busy. The legend reminds you what the instance is allowed: an instance you gave 2 cores can therefore reach 200%, and a curve that keeps hitting that ceiling at peak hours is the clearest sign that it needs more CPU, or a different worker count.

Memory

Memory in use, against the instance's limit. This is the curve to check after an unexplained restart: a container brushing its limit just before vanishing was stopped for exceeding it, not because Odoo crashed.

Network and disk

Traffic in and out and disk writes, for the Odoo container, per second. Useful for recognising a bulk import, a synchronisation running away with itself, or a scheduled job writing far more than you expected.

Storage

Space used on the server's disks, with what is left free. Read it for what it is: these are the machine's disks, shared with the other instances it hosts, not this instance's exact share. It is still the right warning before a disk fills up.

Where the numbers come from

This is the point that explains all the rest: your server measures itself, every five minutes, and keeps its own history. Opening this screen only reads what it has already written; closing it leaves nothing running.

That choice has two pleasant consequences. Nothing connects permanently to your machine to watch it. And the history has no gaps when our platform restarts, when the network hiccups or during a deployment, which would be exactly the moment you'd miss the graphs most.

The history keeps one measurement every five minutes over the last 48 hours, then one hourly average over the past month. Long periods are therefore smoothed: over 30 d, a ten-minute spike shows up inside an hourly average, not as an isolated peak. For a specific incident, look at 1 h or 24 h.

"No history yet"

That message means there is nothing to show for now, not that something is broken: the instance has never started, so it has consumed nothing yet. Once it runs, the graphs start filling within a few minutes, and you naturally have to let 24 hours pass before the 24 h view is complete.

What it's for, in practice

The most direct use is sizing your instances properly: seeing before deciding, rather than adding CPU as a precaution. An instance whose processor tops out every day at 9 am needs resources; a test instance sitting at 3% can give some back, and that capacity returns to the other instances on the same server.

It is also how you answer "Odoo is slow" with something other than a hunch: a saturated CPU, a database doing heavy work and a module waiting on an outside API do not look remotely like the same curves.

A question about your graphs?

If a curve surprises you or you're not sure what to conclude from it, contact us: our team reads them all the time and will tell you what they see.

Updated on: 24/08/2026