Services and job logs

Use the services sidebar on Admin > Update to see the state of every service, restart or stop one, and open its logs. Job logs keeps the output of every restart, stop and update for an hour after it ends. For updates, see Update DataMind OS; for the page itself and who can open it, see Update.

Find a service

1. Open Admin > Update. The sidebar lists the services in their groups.
2. Type in Search services… to find a service by name.
3. Click Filter and choose All services, Unhealthy or Needs update. Unhealthy and Needs update show a count when there is something to count. Needs update appears with Check for Platform Updates.
4. To list the setup services that ran once and finished, turn on Completed init containers in the same menu. They are hidden by default, and a group that holds only those reads "Only completed init containers. Show them from Filter."

Click Reset beside Filter to clear it. On the Map tab, click a group to find it in the sidebar.

Read a service's state

Each row shows a state dot, the service name, its marks and, on the right, how long a running service has been up, such as 4h 12m or 2d 6h. A service that is up shows only that uptime, because the dot already says it is up. Any other state is spelled out in its colour, with the uptime after it when there is one, such as Unhealthy · 1h 5m. A setup service that finished cleanly reads Completed, and a service with no container on this host reads Not deployed. While a restart, a stop or an update runs, that takes the place of the uptime. Hover the name to see its container and what the service does.

In this sidebar the Critical, Update available and Config pending marks are icons, to keep the line short; hover one to read its name and what it means. The state is not among the marks: it sits on the right of the row, with the uptime. Hovering the row hides that right-hand side while the logs, restart and stop buttons take its place. Init keeps its name.

While a DataMind OS update runs, a row shows the service's place in it, such as its download percentage.

The list keeps itself current, and refreshes faster while a job runs or a service is still settling. The refresh button in the sidebar header reloads it at once; with Check for Platform Updates, it checks for updates too.

Open a service

Click a service to open it on the right, in place of the tabs. The header shows Back, the service with its badges and how long it has been up, then Open logs and the restart and stop buttons. Below, its facts: Status, Docker, Container, Exit code, Auto-restarts, Image, Installed, Available, Registry, Config and About, as far as they apply.

Open logs, and the logs button on a hovered row, open Observability > Logs > Platform filtered to the service's container. They need View Container Logs. Without it the details show no Open logs button, and the page reads Reading container logs needs the View Container Logs permission. See Platform container logs.

Restart or stop a service

1. Hover the row and click the restart or stop button, labelled Restart <name> or Stop <name>. The same buttons sit in the service's header when it is open.
2. Read the confirmation, Restart <name>? or Stop <name>?:
   • A restart reads "Restarts <name> in place. Services that depend on it keep running and reconnect on their own. Configuration changes aren't applied by a restart."
   • A stop reads "Stops <name>. It stays down until an update starts it again, or someone starts it on the host."
   • For a critical service, a callout starting with Critical: says what happens to DataMind OS, and the button reads Restart anyway or Stop anyway.
3. Confirm. The row reads Restarting… or Stopping…, and a pill over Map shows the running job; click the pill to open it in Job logs.
4. When it ends, a toast reads "<name> restarted." or "<name> stopped.", or "Couldn't restart <name>" or "Couldn't stop <name>" with the reason. See logs in the toast opens the job in Job logs.

The page has no start button: a stopped service stays down until an update or someone on the host starts it again. Stop is greyed out with "Already stopped" while the service is not running.

Follow a job in Job logs

1. Click Job logs in the sidebar header, or the Job logs tab. While a job runs, the button's icon spins.
2. Read the list. Each job is one line, newest first, with its status, its message, how long it ran and when it started. The running job, or else the newest, opens with its output under it; click another line to open it.
3. A running job that has written nothing yet reads "Waiting for output…"; an ended job without output reads "No output was recorded."

A job stays in the list for an hour after it ends. When nothing ran in that hour, the view reads No jobs in the last hour, with "Restart, stop or update a service and its output shows here."

Who can restart and stop

The list shows only the services your role may see, and the counts describe what you can see. An action on a service your role cannot touch is refused: a restart with Access denied. You do not have permission to execute the service and the service name, a stop with Access denied. You do not have permission to delete the service and the name.

Platform Administrator holds all of these; Platform Viewer is read-only. See Roles and permissions and the permission reference.

Limits and edge cases

• One maintenance job runs at a time. While it runs, restart and stop wait with the tooltip "Wait for the running job to finish".
• While the Installer replaces itself, a notice reads Updating the Installer, and service controls are unavailable until it is back, usually in under five minutes.
• A setup service, or a service with no container, has no restart or stop button.
• When the group definition cannot be read, the sidebar reads "Service schema unavailable, so the groups are approximate". When the deployment files on the host are incomplete, it reads "Deployment files are incomplete on the host"; the list still shows what is running.
• If the groups cannot load, a strip reads Service groups unavailable: with the reason and Retry. If a restart or stop cannot start, the toast reads Could not start the restart or Could not start the stop, with the reason. More in Administration troubleshooting.

Questions, answered

What is the difference between Restart and Stop?

A restart brings the container back in place, and services that depend on it keep running and reconnect on their own; a restart does not apply configuration changes. A stop takes the container down, and it stays down until an update or someone on the host starts it again.

What is the number at the end of a service row?

How long the service has been running, such as 4h 12m or 2d 6h. A service that is up shows only this, because its dot already says it is up; any other state is spelled out first, such as Unhealthy · 1h 5m. A service with no container here reads Not deployed, a setup service that finished cleanly reads Completed, and while a restart, a stop or an update runs, that takes the place of the uptime.

What does "Config pending" mean?

The service's configuration changed after it started, so the running container does not use it yet. A restart does not apply it; the next DataMind OS update does. See Update DataMind OS.

Why are the restart and stop buttons missing from a row?

Your role lacks Restart Platform Services or Stop Platform Services, or the service is a setup service or has no container here. Those rows cannot be restarted or stopped.

Why are the restart and stop buttons greyed out?

A maintenance job is running, or the Installer is replacing itself, so new actions wait until it finishes; the tooltip reads "Wait for the running job to finish". Stop is also greyed out with "Already stopped" while the service is not running.

How do I see the setup services that already finished?

Click Filter and turn on Completed init containers. They are hidden by default because they ran once and exited cleanly.

What does the Critical badge mean?

DataMind OS depends on that service: restarting or stopping it takes DataMind OS down for everyone. Its confirmation says so in a callout, and the button reads Restart anyway or Stop anyway.

How long does a job stay in Job logs?

An hour after it ends. The newest job is listed first, and the running job opens with its output.

Why do I see only some of the installation's services?

The list shows the services your role may see. The others are left out, and the counts describe the visible set. Ask an administrator if you need to see more.

Why is every service in one group?

The group definition could not be read, so the sidebar reads "Service schema unavailable, so the groups are approximate". The states still come from the running installation; refresh to try again.

Where do I read a service's logs?

A service's logs open in Observability > Logs > Platform, filtered to its container: use Open logs in its details or the logs button on its row. See Platform container logs.