In a multi-node xMatters deployment, one or more nodes failed to start after an upgrade. What can I do to troubleshoot this issue?
After upgrading xMatters:
- One or more nodes fail to start in a multi-node environment;
- The main AlarmPoint.txt log file (located at <xMHOME>/logs) is either empty or has not been created; and,
- The Java process has stopped.
Typically, one of the following upgrade mistakes causes this problem:
- The node was not actually upgraded prior to restart. For the entire deployment to function properly, each xMatters component must be upgraded. Check the Version column in the table on the “Nodes and Device Engines” page for mismatched component version.
- Although the node was upgraded, the incorrect existing node name was selected during installation, resulting in an improperly configured node. Check whether the AlarmPoint.err file (located at <xMHOME>/node) includes stack traces as well as the following error (line breaks added for clarity):
[AlarmPoint Node-main] ERROR - The messaging server for this node failed to start on URL http://192.123.456.78:50000. This could be because an instance of this node is already running, or another service is using the required port.
The failing node must be reinstalled and upgraded properly. The user must ensure that they select the correct node name when using the ‘Use Existing Node’ option in the installer, and that all component versions match.
JDN-1188 Originally created by Don Clark