IBM Data Science Experience (DSX) Local automatically notifies you when a node or pod goes down or when you're at risk of overloading a resource.
By default, DSX Local issues alerts when:
- CPU usage on a node goes above 90%
- Memory usage on a node goes above 90%
- Disk usage on a node goes above 90%
- A node in the cluster goes down
- A pod fails
- A pod is not running or is in an unknown state for more than 5 minutes
If you want to change the threshold at which alerts are issued, you can configure them on the Settings page. For more information, see Set up Data Science Experience Local.
DSX Local also issues a follow-up alert once the problem is resolved.
Tip: If DSX Local is configured to connect to your SMTP server, each DSX admin receives alerts through email.
When you have alerts, the alert icon displays the number of unread alerts in your queue:
You can access alerts in either of the following ways:
- If you want a quick peek at your alerts, click the alert icon in the menu bar ().
- If you want to manage your alerts, you can access the Alerts page from the menu icon ().
From the Alerts page you can:
- Filter alerts by type
- Filter alerts based on whether they were read or not
- Filter alerts by status
- Mark alerts as read or unread
- Delete alerts
Tip: When you delete an alert, you can't access it again. Make sure that you don't need it before you delete it.