Troubleshooting

The connector provides self-healing features through the usage of retry jobs. However, there are some times when the connector may not be able to recover from an error. In these cases the user has other alternatives.

Manually executing the extract jobs

The connector is designed so every extract job can be retried without generating duplicated data or generating errors for processing various times the same item.

This means that if something fails, for example the processing of a seller or an invoice, the user can use the jobs HTTP API to reprocess the item.

Restarting the connector

Sometimes, the connector may be in a state where it’s unresponsive or its internal structures, like a cache, may contain invalid data. In these cases, the user can restart the connector to restore it to a working state.

Removing persistent data

The connector stores in an internal database some auxiliary data to speed up the processing of the data. This data is stored in the data directory and the connector can regenerate it if it’s deleted.

The user can remove this data and restart the server to restore the connector to a working state.

If the connector is running in a container, the user can remove the container destroying it associated data. If the data is stored in a volume, the user can remove the volume.