DEV Community

Michael
Michael

Posted on • Originally published at gbase8.cn

GBase 8a Cluster Expansion Requires All Nodes Online

#gbase #database #数据库 #operations

When scaling out a gbase database cluster, the installer checks the online status and SSH connectivity of every node. If any node is OFFLINE due to an unrecoverable failure, the expansion script will fail with login errors, preventing the operation from proceeding.

Example Environment

A two‑node cluster where node2 (10.0.2.202) is offline and unreachable.

[gbase@localhost gcinstall]$ gcadmin
...
|  node1  |    10.0.2.201     |    OPEN     |   OPEN    |    0     |
|  node2  |    10.0.2.202     | OFFLINE     |          |

[gbase@localhost gcinstall]$ ping 10.0.2.202
... Destination Host Unreachable ...
100% packet loss
Enter fullscreen mode Exit fullscreen mode

Error Messages

The expansion script (gcinstall.py) produces errors such as:

Check loginUser:root password failed,nodes are:10.0.2.202
Check root password failed,nodes are:10.0.2.202
Enter fullscreen mode Exit fullscreen mode

The backend log gcinstall.log reveals the underlying cause:

fail to login root@10.0.2.202,err:ssh: connect to host 10.0.2.202 port 22: No route to host
Enter fullscreen mode Exit fullscreen mode

How to Resolve

  1. Repair the faulty node: Restore the operating system and network connectivity of the failed node so it returns to the cluster before you start the expansion.
  2. Emergency workaround: If the node cannot be recovered, contact GBASE support. They can provide a temporary adjustment that bypasses the login check on the unreachable node, allowing the expansion to proceed.

Keeping all nodes healthy is a prerequisite for any topology change in a gbase database cluster. When expansion is blocked by an offline node, address the node first — the expansion will follow smoothly.

Top comments (0)