Restart or rebuild proxy nodes on Tair (Redis OSS-Compatible) cluster and read/write splitting instances to resolve connection imbalance, high latency, or service disruptions.
Restart mode comparison
|
Mode |
Effect |
When to use |
|
In-place Restart |
Restarts the proxy process on the existing node |
Connection imbalance or minor latency spikes |
|
Rebuild Proxy |
Replaces the proxy node entirely |
In-place restart did not resolve the issue |
Start with In-place Restart. If the problem persists, use Rebuild Proxy.
Impact on service availability
This operation causes brief connection interruptions. Perform it during off-peak hours and ensure your application supports automatic reconnection.
Supported instance types
This operation applies to the following instance types:
-
A cluster instance in proxy mode
Restart or rebuild a proxy node
-
Log on to the console and go to the Instances page. Select the region and click the target instance ID.
-
In the left-side navigation pane, click Service Availability.
-
In the Proxy Node section, select the target proxy nodes and click Batch Restart.
-
In the panel that appears, set Restart Mode:
-
In-place Restart: Restarts the proxy nodes.
-
Rebuild Proxy: Replaces the proxy nodes entirely. Use this if in-place restart did not resolve the issue.
-
-
Click OK.
Verify the result
After the operation completes:
-
On the Service Availability page, verify that the instance status changes to Running.
-
Check your application logs to confirm that client connections are re-established.
-
Monitor instance metrics to confirm that latency and connection counts return to normal. View performance monitoring data.