Frequently Asked Questions¶

HOST¶

Why can't I see my server in the HOST list?

Please troubleshoot according to the following steps:

1. Check the DataKit running status: systemctl status datakit
2. Confirm that the HOST has reported data within the last 10 minutes (the HOST list only displays HOSTs online in the last 10 minutes by default)
3. If a HOST has no data reported for more than 24 hours, it will be automatically removed from the list
4. Check DataKit logs: tail -f /var/log/datakit/log

Why is the HOST CPU usage and memory usage displayed as -?

It indicates that the data for this HOST has been interrupted for more than 10 minutes. Please check if DataKit collection is normal.

What is the function after enabling the real-time performance switch?

After enabling, the system refreshes the three metric fields of CPU usage, memory usage, and CPU load every 10 seconds. It is suitable for scenarios requiring real-time monitoring of HOST performance.

How long does it take for HOST tags to take effect after configuration?

After tag configuration is completed, it usually takes 1-5 minutes to take effect.

How to view historical HOST data?

The HOST Explorer displays data from the last 2 days by default. You can select time ranges such as the last 2 hours, 6 hours, 1 day, or 2 days via the Time Widget in the upper right corner.

What does the red color in the "Integration Status" on the HOST details page mean?

It indicates that the collector has encountered an error. You can click the red label to view the error message. Normally running collectors are displayed in light blue.

CONTAINERS¶

Which Kubernetes resources does the CONTAINERS Explorer support displaying?

It supports displaying core resource types such as Containers, Pods, Services, Deployments, Nodes, Replica Sets, Jobs, Cron Jobs, Daemonset, Statefulset, Persistent Volumes, PVC, etc.

What is the difference between installing DataKit via DaemonSet and installing it on the HOST?

• HOST installation: Enables the container collector, supporting the collection of Containers and Pods data
• DaemonSet installation: Supports the collection of full Kubernetes resource data and automatically builds cluster topology

How to handle uncollected CONTAINERS data?

1. Confirm that DataKit has permission to access the Docker/Kubernetes API
2. Check if the container collector is running normally
3. In Kubernetes scenarios, confirm that DataKit is deployed as a DaemonSet

Process-related¶

Why can't I see certain processes in the process list?

The process Explorer collects processes that have been alive for more than 10 minutes by default. To view all processes, you need to adjust the process collector configuration.

How to view historical process data?

The process list supports viewing process data collected within the last ten minutes. You can view historical process playback by dragging the Time Widget, with a query time range of 5 minutes.

What data can be associated on the process details page?

A: The process details page supports associating and viewing data such as Metrics, LOGs, HOSTs, and NETWORK, enabling contextual linkage from processes to infrastructure.

DATABASE¶

Which database types does DATABASE monitoring support?

It supports mainstream databases such as MySQL, Redis, PostgreSQL, SQL Server, Oracle, etc.

What is DBM monitoring?

DBM (Database Monitoring) is an in-depth monitoring feature for SQLServer, MySQL, and PostgreSQL. It allows viewing Table-level information, including table name, number of columns, primary key, engine, row format, creation time, etc.

How long is DATABASE data retained?

A: The DATABASE Explorer displays the latest metric data from the last 2 days by default.

NETWORK¶

What technology is NETWORK monitoring based on?

A: It is based on eBPF technology to achieve non-intrusive collection of network traffic with zero performance loss.

What metrics can the NETWORK module monitor?

It supports monitoring metrics such as sent bytes, received bytes, TCP latency, TCP jitter, TCP connections, TCP retransmissions, TCP close counts, etc.

What conditions are required to use the NETWORK feature?

1. Register and log in to TrueWatch
2. Install DataKit on the HOST
3. Enable the eBPF collector

Resource Catalog¶

What are required attributes?

Required attributes are fields that data must contain, used for data filtering and interception. If the reported data lacks required fields, that data cannot be reported to the TrueWatch workspace. The object's name field is added by default.

Will deleting a Resource Catalog Explorer template delete the data?

No. Deleting an Explorer template only affects the data display method. If data continues to be reported, the default Explorer template will be applied automatically. To completely delete data, you need to go to Manage > Delete Resource Catalog to perform the operation.

How is Resource Catalog data reported?

By creating custom resources and combining DataKit API and DataFlux Func to report any data to TrueWatch, including cloud vendor cloud resource data, enterprise business data, etc.

How to query Resource Catalog data across workspaces?

In the upper left corner of the Resource Catalog page, select all workspaces associated with the account to achieve cross-workspace querying of resource list data.

General Questions¶

What is the infrastructure data retention policy?

• Object data: Last 2 days
• Process data: Last 2 days
• Metric data: Default 7 days (configurable)

How to configure infrastructure alerts?

Go to Monitoring > Monitors to configure monitors such as infrastructure liveness detection and threshold detection. It supports various notification methods including DingTalk, WeCom, Lark, email, etc.

Does infrastructure support cross-workspace data authorization?

Yes. If the current workspace has configured data authorization and the authorized data types include "Infrastructure", you can switch to the authorized workspace in the upper left corner of the Explorer page to view the relevant data.