Cluster manager board

Learn about the Cluster Manager board.

You can view cluster information from the Admin Console. The System Cluster Liveboard contains several ThoughtSpot answers that display latency over time, snapshot status, installed release, node functions, and logs. To view this Liveboard, navigate to the Admin Console by selecting the Admin tab from the top navigation bar. Select Cluster from the side navigation bar that appears.

Admin Console - Cluster Liveboard

Only the Average Latency Last 2 Weeks (sec) panel contains a standard ThoughtSpot visualization. The other visualizations rely on internal queries that aren’t accessible. You can, however, Present or Copy link on them.

Cluster Details

The fields on this visualization have the following meaning:

Cluster Name

The name of the cluster defined at installation time.

Cluster ID

The ID of the cluster set at installation time.

Last Update Time

Last time the cluster was updated.

Release

Version of the current release.

Zookeeper Servers

IP addresses of the Zookeeper servers.

HDFS Name Nodes

Control nodes for Hadoop Distributed File System (HDFS).

Cluster Logs

The fields on this visualization have the following meaning:

Time

A timestamp indicating when an action occurred.

Type

Type of action.

Release

Identifies the full release number.

You can also use the tscli logs command to review log data from your cluster.

List of Snapshots

This visualization shows the snapshots and the backups taken on the cluster. The fields on this visualization have the following meaning:

Time

A timestamp indicating when a snapshot or backup happened occurred.

Name

Name of the snapshot file. These files are stored in the /usr/local/scaligent/backup directory on your cluster.

Reason

Identifies the reason the snapshot/backup was created. You should see several period snapshots if your cluster is configured properly. You may also see evidence here of manual backups. For example, you should be sure your cluster is backed up before major events such as upgrades. Contact ThoughtSpot Support if you don’t see the evidence that your cluster is periodically creating snapshots.

Size

Size of the backup in gigabytes.

Average Latency Last 2 Weeks (sec)

This visualization relies on the TS: BI Server worksheet to display the average database latency over the last 15 days. The database latency measures how long it takes for a search to return data from ThoughtSpot - this doesn’t include the time taken to send the answer back to the client, it measures internal processing time. You can use the visualization menu to drill down to its underlying query:

average datacache (sec)
average total (sec)
daily
last 15 days
for database latency (us) > 0

Related information


Was this page helpful?