Operations 10 min read

Top 16 Common TDH Community Edition Installation Issues and How to Fix Them

This guide compiles the most frequent problems encountered when installing the TDH Community Edition—such as hostname configuration, logical volume creation errors, service startup failures, firewall settings, and license issues—and provides clear step‑by‑step solutions to help users avoid and resolve these obstacles.

StarRing Big Data Open Lab
StarRing Big Data Open Lab
StarRing Big Data Open Lab
Top 16 Common TDH Community Edition Installation Issues and How to Fix Them

1. How to change the hostname on CentOS and RedHat

Three methods are available: run hostname xxx for a temporary change, edit /etc/sysconfig/network to make it persistent after reboot, or modify /etc/hosts ensuring all cluster nodes have the correct hostname.

Example /etc/hosts entry (note the last line):

127.0.0.1   localhost localhost.localdomain
::1         localhost localhost.localdomain
172.16.2.25 tw-node2025

Windows hosts file location: C:\Windows\system32\drivers\etc\hosts. Linux hosts file location: /etc/hosts. Avoid changing the hostname after installation because many parameters depend on it.

2. Does TDH require a specific hostname format?

Yes. Hostnames may contain only letters, numbers, or hyphens. Underscores (_) and dots (.) are not supported (e.g., TDH1.cluster.com is illegal).

3. pvcreate fails with "Can't open /dev/vdb exclusively. Mounted filesystem"

This indicates the device is in use. Check whether the filesystem is mounted; if so, unmount it before running pvcreate.

4. pvcreate fails with "Can't initialize physical volume \"/dev/vdb\" of volume group \"docker\" without -ff"

The physical volume already exists, so you can skip recreation and continue with the subsequent steps.

5. Installation stops at hostname check

Log into the manager node, run hostname, and compare the output with the name after the IP in /etc/hosts. If they differ, modify the hostname so that all nodes match.

6. Installation fails at OS version check

Verify that the OS version of the installation node is supported by the community edition. Also ensure the TDH package version matches the OS (e.g., CentOS 6.5 requires a corresponding package; using a CentOS 7 package on CentOS 6.5 will fail).

7. Accessing http://hostname:8180 returns an error

Possible causes include incomplete installation, using the wrong protocol, firewall blocking port 8180, or the Manager service not running. Check the Manager status with /etc/init.d/transwarp-manager status. If not running, view the log at /var/log/transwarp-manager/master/transwarp-manager.log. If the service is running, verify that port 8180 is not occupied by another process using netstat -anp | grep 8180, then restart the Manager with /etc/init.d/transwarp-manager restart.

8. License submission reports username/password error

Enter the username and password used when registering Snowflake, not the email address used for the application, and ensure they are typed correctly.

9. Node disconnection reasons

Potential causes are hardware issues (network cable, NIC, switch) or the host firewall blocking SSH.

10. How to check if the firewall is enabled

Method 1: Run the setup command, navigate to firewall configuration, and choose to disable the firewall. Method 2: Execute service iptables stop to stop the firewall service.

11. Does Transwarp Manager have network latency requirements?

If the client’s ping to the Manager exceeds 2 ms, installation may fail. Ensure latency is low.

12. "No configure setting for key 'repo'" error during service installation

Check that /etc/transwarp-manager/master/repo.conf exists and is not corrupted. If missing or damaged, create it with the following content:

repo{
  remove-exist=yes
  os{
    type=remote
    url="<os_repo_url>"
  }
  transwarp{
    type=local
    path=<transwarp_repo_path>
  }
}

Replace <os_repo_url> with the value from /etc/yum.repo.d/os.repo and <transwarp_repo_path> with the value from /etc/yum.repo.d/transwarp.repo.

13. How to download the TDH Client package via command line

Log into the Manager to obtain a cookie:

curl -v -X POST http://<manager_ip>:8180/api/users/login \
  --data '{"userName":"<manager_admin>","userPassword":"<manager_psw>"}' \
  -b cookies.txt -c cookies.txt

Then download the client:

wget --load-cookies cookies.txt http://<manager_ip>:8180/api/manager/tdh-client

14. Why does the Manager web page crash?

The front‑end relies heavily on JavaScript. Older browsers or unstable JS engines can cause crashes. Use Chrome 41+ or Safari 9+, and disable browser extensions that might interfere.

15. License service not working

In the Manager UI, go to Management → License. If the service shows "DOWN" because more than two nodes are down, click the "Start" button to restart it.

License service status
License service status

16. How to check license expiration date

Navigate to Management → License and view the "Expiration Date" field.

License expiration
License expiration

For a complete installation guide, refer to the official TDH Community Edition documentation on the StarRing website.

operationsLinuxTroubleshootingInstallationTDHTranswarphostname
StarRing Big Data Open Lab
Written by

StarRing Big Data Open Lab

Focused on big data technology research, exploring the Big Data era | [email protected]

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.