Help Center/ Cloud Search Service/ User Guide/ Logstash/ Managing Clusters/ Forcibly Restarting a Cluster
Updated on 2026-04-30 GMT+08:00
View PDF
Share

Forcibly Restarting a Cluster

When a Logstash cluster becomes unavailable due to extended runtime or unknown faults such as memory overflow or process failures, regular configuration updates may not be sufficient to restore it. In such cases, CSS offers a forcible restart option that recovers the cluster by directly restarting its underlying VMs.

Impact

Forcibly restarting a Logstash cluster stops all Logstash processes, interrupts all ongoing data streams, and makes the cluster totally unavailable during the restart.

The keepalive setting determines each pipeline's status after the restart completes. Table 1 describes how pipelines recover after the restart, depending on the keepalive setting.

Table 1 Pipeline recovery

Keepalive

Behavior Upon Restart

Recovery Process

No

Stopped
  1. During the restart, all pipeline tasks that are Running are forcibly Stopped.
  2. After the restart completes, manually start the pipelines.

Yes

Attempt to recover automatically
  1. When the restart begins, the pipeline status is set to Recovering.
  2. The system attempts to restart the Logstash processes.
  3. If the Logstash processes are started within 10 minutes, the pipelines are automatically restored to Working.
  4. Otherwise, the pipeline status changes to Failed, and you need to manually start them.

The total restart time depends on how fast the underlying VMs restart. Usually, it takes longer than how long the Logstash processes restart.

Constraints

Clusters whose Task Status is Frozen or that have ongoing tasks cannot be restarted.

Before You Start

  • Check the target cluster's health status. Check the Task Status column to ensure there are no ongoing tasks.
  • Perform this operation during off-peak hours, or ensure the upstream data source (such as Kafka) has sufficient buffering capacity to hold data until consumption resumes after the restart.

Forcibly Restarting a Cluster

  1. Log in to the CSS management console.
  2. In the navigation pane on the left, choose Clusters > Logstash.
  3. In the cluster list, find the target cluster, and choose More > Force Restart in the Operation column.
  4. In the Force Restart Cluster VMs dialog box, confirm the restart information, type in YES for confirmation, and click OK.
  5. When the cluster status changes from Processing to Available, the cluster has restarted successfully.
Parent topic: Managing Clusters