Lindorm:Technical support and FAQs

Technical support

For questions about Lindorm products and services, join the DingTalk group (ID: 35977898) for support.

FAQ

If you encounter connection issues when you use Lindorm, refer to the following FAQs.

• Q: I cannot access a Lindorm instance from an ECS instance over a VPC. What are the possible causes?

  A: Connecting to a Lindorm instance from an ECS instance over a VPC requires network connectivity between the two instances. If the connection fails, check the following common causes:

  • The primary private IP address of the ECS instance is not added to the Lindorm whitelist. Add the primary private IP address of the ECS instance to the Lindorm whitelist. For more information, see Configure a whitelist.

  • A security group rule for the ECS instance blocks access to the Lindorm internal endpoint and port. Add a security group rule to allow access. For more information, see Add a security group rule.

  • The ECS instance and the Lindorm instance use different network types. If the network types are different, consider the following scenarios:

    • Instances in the classic network cannot access instances in a VPC.

    • If the instances are in the same VPC but use different vSwitches, add security group rules and network ACL rules. For more information, see Add a security group rule and Add a network ACL rule.

    • By default, instances in different VPCs cannot access each other. Configure Express Connect. For more information, see How to choose private connection products?.

    • To access the instance from an environment outside Alibaba Cloud, such as a corporate intranet, enable public access or connect by using a leased line.

• Q: The client returns the error "Retry exhausted when update config from seedserver". How do I troubleshoot this?

  A: Troubleshoot the issue as follows:

  • If you access the Lindorm instance from an ECS instance, verify that the two instances can connect to each other. For more information, see the network connectivity troubleshooting steps above.

  • The endpoint or port of the Lindorm instance may be incorrect. For example, the port for an HBase-compatible client is 30020.

• Q: The client returns the error "Failed to connect to jdbc:lindorm:table:url=****". How do I troubleshoot this?

  A: Troubleshoot the issue as follows:

  • If you access the Lindorm instance from an ECS instance, ensure that they can connect to each other. Refer to the network connectivity troubleshooting steps above.

  • The endpoint or port of the Lindorm instance may be incorrect. For example, the port for Lindorm wide-table SQL is 30060. To obtain the endpoint and port, see Obtain connection information.

• Q: What causes the error "DoNotRetryIOException: Detect inefficient query: SELECT*FROM xxxx …"?

  A: This error occurs because the Lindorm wide-table engine identifies the SELECT statement as an inefficient query and the server refuses to execute it. To resolve this error, see SELECT.

• Q: How do I use the ping command to test connectivity between an ECS instance and a Lindorm instance?

  A: After you create a Lindorm instance, or if a running instance becomes unreachable, run the ping command on the ECS instance to test connectivity:

  1. View endpoints.

  2. Log on to the ECS instance and run the following command:

     ping <host>

     Example:

     ping ld-bp1mq0tdzbx1m****-proxy-lindorm-pub.lindorm.aliyuncs.com

     Note

     • host specifies the endpoint without the port number.

     • This command is available on both Windows and Linux. On Windows, use ping <host> -t for continuous testing.

  3. View the test results:

     • On Windows, the results are displayed after 4 pings.

     • On Linux, ping requests are sent continuously. Press Ctrl+C to stop the test and view the statistics.

  Analyze the results: If all requests receive a reply, the connection is active. If requests fail, check your whitelist, security group, and network configurations as described above.

• Q: How do I use the telnet command to test the port connectivity of a Lindorm instance?

  A: If the ping test succeeds but you still have connection issues, use telnet to test whether the service port is reachable.

  Prerequisites: Telnet is installed on the ECS instance (Linux), or the Telnet Client feature is enabled (Windows).

  1. View endpoints.

  2. Log on to the ECS instance and run the following command:

     telnet <host> <port>

     Example:

     telnet ld-bp1mq0tdzbx1m****-proxy-lindorm-pub.lindorm.aliyuncs.com 33060

     Note

     • host specifies the endpoint, and port specifies the port number.

     • This command is available on both Windows and Linux.

  3. View the test results.

  Analyze the results: A successful telnet connection indicates the network is working. If the connection succeeds but the service is still inaccessible, the issue likely lies with your client or application code.

• Q: I receive a "connection timeout" error when I use Spark on MaxCompute to access a Lindorm instance. What should I do?

  A: A connection timeout error may occur when you use Spark on MaxCompute to access a Lindorm instance, regardless of whether you connect by using a compatible protocol or the Lindorm client.

  Cause: In the network architecture of Spark on MaxCompute, multiple pods communicate with data services in other VPCs through an elastic network interface (ENI). The ENI allocates ports based on an algorithm. When different destination IP addresses are accessed, the same port may be reused. Lindorm client connections and protocol-compatible connections use Server Load Balancer (SLB) and direct connections. When SLB performs network address translation (NAT) to convert the source address to the ENI address, the load-balancing server may drop packets if it has not retained the NAT information. This can cause a connection timeout or data stream crossover.

  Solution: Use the HBase client built into open-source Spark to access the Lindorm instance.