View the status
The data in the HBase table is synchronized to the index in the Search service by (Big Data Service) BDS. You can view the status of the synchronization in the BDS console.
- On the details page of the BDS instance that is associated with the HBase cluster, find the web console entrance.
- In the BDS console, click
Lindorm Search
and selectReal time replication to Solr
. - Click the name of a channel to view the synchronization details. In most cases, you need to only pay attention to the synchronization status and latency between the HBase table and the Search index.
Monitoring and alerting
You can view the latency indicators of the data synchronization on the Monitoring and Alerting page in the BDS console and subscribe to alerts by using Cloud Monitor.
- Click Cloud Monitor in the BDS console to go to the Cloud Monitor console.
- Select BDS Metrics to view the current latency status in the task delay max(ms) section. Click the bell icon to configure alerts.
Troubleshoot common issues
If the synchronization latency in the BDS console does not meet the requirements for your business, you can troubleshoot the issue.
- The synchronization progress is stuck at a specific offset
- If the synchronization is stuck at a specific offset in the BDS console, it generally
indicates that invalid data is written to the HBase table, which causes the synchronization
to fail.
You can click View Details on the right side to find the error message on the details page.
If you have issues that you are unable to resolve, you can contact customer service in theApsaraDB for HBase Q&A
DingTalk group or submit a ticket.
- Slow synchronization
- If the synchronization is still in progress, and the synchronization speed shown in
the
Write Details
andRead Details
columns is slow, the write performance of the Search service may have reached the upper limit. You can scale out the service or upgrade the configuration. Before you scale out the service or upgrade the configuration, you can check the server load of the Search instance in the Cloud Monitor console. If you have questions, you can contact customer service in theApsaraDB for HBase Q&A
DingTalk group or submit a ticket for support.
- A large number of accumulated logs exist in the task queue.
- If the synchronization speed in the
Write Details
andRead Details
columns is normal, but the number of the accumulated logs that are shown in theTask Queue
column is increasing, HBase may write data too fast. As a result, an increasing amount of data is waiting to be synchronized to the Search index by BDS. In this case, you can scale out the BDS cluster. This issue may also be caused by the limited write performance of the Search service. In this case, you can scale out the service or upgrade the configuration. Before you scale out the service or upgrade the configuration, you can view the details of the BDS and the Search instances in the Cloud Monitor console. If you have questions, you can contact customer service in theApsaraDB for HBase Q&A
DingTalk group or submit a ticket for support.