Skip to content
upstayHelp Center

System health, jobs and incidents

What it does

Section titled “What it does”

Health (System health) shows the server now: CPU, memory, swap, disk and load average, the containers, whether the worker is running, and how many screens are connected live. Server history draws CPU, memory, disk, swap and load for the Last 24 hours or Last 7 days, sampled every minute and kept 7 days. CPU or memory above 85% for 10 minutes, or disk above 85%, goes to Needs you and emails the operator on call, once per episode.

Jobs & queues (Jobs and queues) lists every job the worker repeats, how often it runs, when it last ran and whether it is late. Run now is offered only for jobs that are safe to repeat.

Providers (Providers and credits) shows SMS, WhatsApp, email, payments, the channel manager, Roomie and ID reading, judged by what Upstay sent them and what came back, with the credit left where the provider lets Upstay read it.

Incidents (Incidents and errors) is where you record that something is broken: who took it, what happened, which hotels felt it and what they were told. Errors are grouped so one bug is one row.

Deploys shows which version is running, recent deploys, the checks after each deploy and the latest database changes. From here you can tell hotels before a deploy.

Who can use it

Section titled “Who can use it”

Upstay operators who can see the System area.

Where to find it

Section titled “Where to find it”

System in the rail: Health, Providers, Jobs & queues, Incidents, Deploys.

How to read the server’s history

Section titled “How to read the server’s history”
  1. Open Health.
  2. Under Server history, pick Last 24 hours or Last 7 days.
  3. Each chart shows its peak and average, with a line at 85%.

How to test every provider

Section titled “How to test every provider”
  1. Open Providers.
  2. Click Test all connections. Each provider’s line says what came back.
  3. For a provider whose balance Upstay cannot read, check the balance in its own console.

How to run a job now

Section titled “How to run a job now”
  1. Open Jobs & queues.
  2. Click Run now on the job. Click Log to see its last runs.

How to open an incident

Section titled “How to open an incident”
  1. Open Incidents and click Open an incident.
  2. Enter What is wrong, and if you like What it affects (optional) and First note (optional).
  3. Pick Minor or Major.
  4. Tick Show every hotel a banner if hotels should know. It follows each update and comes down when closed.
  5. Click Open it.

How to work an incident

Section titled “How to work an incident”
  1. Click Take it so everyone sees you have it.
  2. Write Add an update in words hotels can read, set Status to Open or Watching, and click Post update.
  3. When it is fixed but you are still watching, click Post all clear to hotels.
  4. Click Close incident when it is over.

How to tell hotels before a deploy

Section titled “How to tell hotels before a deploy”
  1. Open Deploys and click Schedule a maintenance banner.
  2. Enter Title, Maintenance starts, Minutes and What hotels read.
  3. Click Schedule it. Every hotel sees it a day ahead; it clears itself when the window ends. You can find it in Announcements.

Tips & good to know

Section titled “Tips & good to know”
  • If every job is late, the worker is not running.
  • A provider that keeps failing shows Looks down and joins Needs you.
  • The operator on call is emailed when a job is late, a provider is down or the server crosses 85%. Set who is on call on Team.
  • Every Run now, incident and update is recorded in Activity.

Troubleshooting

Section titled “Troubleshooting”
  • “Every job is late: the worker is not running.” → The worker needs restarting on the server.
  • No samples in Server history → The worker records one every minute, so the first point appears a minute after it starts.
  • Run now is missing → That job is not safe to repeat, so it only runs on its schedule.
Section titled “Related features”