Nodes fail to start due to errors in loading scheduled tasks

Question

In both standalone server and single-machine cluster modes, an error is reported when nodes are started, which is shown as follows:

Failed to unmarshall the job [..]. Can't recognize function zmq::getSubJobStat. Invalid message format

I used the zmq plugin in this scheduled job. But instead of loading it in the main function, I wrapped the code for loading it into a separate function, which is then called by the main function.

The reason for the error may be that the main function cannot resolve the plugin.

How can I solve this problem?

Answer 1

There are two methods.

Method 1:

Modify the configuration file by adding the following parameters in dolphindb.cfg (standalone) or cluster.cfg (cluster).

preloadModules=plugins::zmq,plugins::xxx

Restart the node or cluster after modification. The zmq plugin will be preloaded to the system.

Modify it online to load plugins via loadModule("plugins::zmq").

Method 2:

Remove the jobEditLog.meta file from the data/node/sysmgment path and restart the node.

This method is suitable for tests. After this operarion, all scheduled tasks are removed and need to be resubmitted.