App infrastructure
The Infrastructure tab of an app in Apps mode shows what the app uses and what that costs, by the question a person asks. For the fleet of app databases, see Infrastructure in Lumen mode; this tab is the same idea for one app.
The controls
| Control | Does |
|---|---|
| All · Compute · CPU · Memory · Storage · Network | Narrows the tiles, charts and tables to one question. All shows everything |
| Range | Last 24 hours, Last 7 days or Last 30 days. Longer ranges use coarser steps (hourly, 6-hourly, daily) |
| Refresh | Reads again. The tab also refreshes itself every two minutes |
What each filter shows
| Filter | Tiles | Charts and tables |
|---|---|---|
| CPU | CPU now with the peak, CPU average | CPU used over the range, and the Pods table with each pod's CPU |
| Memory | Memory now with the peak, Memory average | Memory used over the range, and each pod's memory |
| Compute | Services (and how many copies run), Machines | Services and machines: kind, copies, wanted state, last change, and each machine's size, mode and status |
| Storage | Site files, Database (size and state), Disks (total and count) | Data: each disk's size and last backup, a line about the database, and a Database metrics… button that opens the same database charts as the Lumen Infrastructure page |
| Network | Requests, Server errors, Data sent out, Response time p95 (the slowest hour), Reachable, Certificate | One chart each for requests, data sent out, response time, reachability and site storage |
Under All the tab also shows This month, the estimated cost so far, and every tile above.
Metered this month lists, item by item, what hosting counted this month and its estimate, for example CPU-hours, memory GiB-hours, disk GiB-hours, site storage, requests and data sent out. The filter narrows the list to the matching items.
Where the numbers come from
| Numbers | From |
|---|---|
| CPU, memory, pods | Lumen metrics for the app's namespace on the platform |
| Requests, errors, data out, response time, reachability, certificate, site storage | The platform's hosting meters for the app's site |
| Metered this month | App hosting's usage for the project |
| Services, machines, disks, database | What the platform has stored about the app |
When a tile is empty
- The hosting meters only appear where the platform has recorded them for the site in the chosen range. When it has not, the Network tiles show an en dash and a line says "The platform hasn't recorded traffic, probes or storage for this site in this range yet." That is not a zero.
- An app with no running pods shows "No pod metrics for this app's namespace in this range" in place of CPU and memory.
- Metered this month can take a while on a busy project, and says so while it reads. If it fails it says why instead of showing zero.
- Only hosted apps are metered; an app that runs elsewhere shows "Only hosted apps are metered."
Nothing on this tab changes the app. To change how it runs, use the Add menu and the change sheet in Apps mode.
Apps mode
The third mode of Orkestia Cockpit — every app of the organization, how it is hosted and how it is doing, with its own sidebar and a page for each app
Install and sign in
Platforms and installers for Orkestia Cockpit, how updates arrive, the browser sign-in with PKCE, where the session is stored, chats-only mode, signing out and switching organization