The plugin will verify the license, the logged-in user, it will open the XML file and validate it against the schema. If the import passes all these verifications, the process will start. Otherwise, an error message will be shown close to the top of the page.
Once the import process has started, your browser will display a page showing information about the progress of the import.
This page contains a progress bar that shows graphically what part of the import has already been performed. Below, you will see a message with the percentage of the import done so far and the time the import process has been running. Finally, there is another message that shows the last operation performed by the import process (in the example, the importer is processing users and it has already processed 983 users out of a total of 8424 users described in the XML file).
Take the percentage of progress as a rough estimate. It is actually based on expected run times in different parts of a "typical" import, so it may not be an accurate estimation in all cases. For example, if an error aborts the import the percentage will jump abruptly from its previous value to 100%.
The time the import process has been running is measured at the server. Remember that the time perceived by the end user is always longer, as that does not include the time to transfer request and response to/from the server, the time the server needs to process the request and launch the import or the time the browser needs to process and display the response. For a small import, these differences may be noticeable.
All these information items are set to refresh automatically every 3 seconds. If you would like to refresh them more often, you can click on the "Refresh" button. When the import finishes, this page will be automatically redirected to the import results page.
When it finishes, the results page will be shown. Just under the header you will see a message with the result of the load. Below, there is an area that contains the trace of the process. The next image shows the result page after a real load:
Under the title of this page, there is a button to go back to the previous page, in case you want to launch another load. This will happen typically after a successful simulated load, in order to execute the actual configuration changes.
Remember that importing a configuration may alter custom field configurations, resolutions, statuses or priorities. You should reindex the database after the import.
The next image shows the result of a simulated import. The resulting message says this load is simulated. Additionally, in the process trace, you can read a warning at the beginning, that no changes will be applied to the database.
If an error is found during the load process (either simulated or real), the load will stop and the error will be shown within a message just under the page header. This message explains the likely cause of the error.
Below, another message states the location, or object that was being loaded when the error happened. Generally, each element is described by its entity type: "Workflow", followed by a string that identifies it, which is usually its name: "Reduced". In this example, the error happened in workflow "Reduced", while trying to translate a status with name "En el limbo".
The trace of the load process is shown up to the error. In this case, the trace also includes an important piece of information: the stack trace produced by the error.
If this feature is enabled, and an error is found during the import of a dashboard or filter: