When applying a config override to the cluster's URL, the ClusterForm’s status info does not reference the override and shows the error message “No alive nodes found in your cluster”. This occurs even when the rest of the website connects successfully.
This is because the form’s clusterFormInfo() method only references the database's stored settings as opposed to the website’s loaded config for that cluster.
I tend to specify the cluster’s port in a .env file so other website developers can use their own preferred local configuration of Elasticsearch. I have a config override in settings.php to set the cluster URL rather than using what’s actually stored in the cluster’s entity.
E.g. in settings.php: $config['elasticsearch_connector.cluster.docker']['url'] = "http://localhost:{$_ENV['ELASTICSEARCH_PORT']}";
May I suggest that the edit cluster form tests the live configuration for the entity rather than strictly what’s in the database?
Admittedly this practice can lead to some confusion because Drupal doesn’t indicate config overrides in the form UI ( see core issue 2408549 🐛 Display status message on configuration forms when there are overridden values Fixed ) but that’s the norm for this kind of website setup and changing the current behaviour would prevent a disconcerting error message, which can be very misleading when debugging other issues.
Active
7.0
Elasticsearch Connector
Not all content is available!
It's likely this issue predates Contrib.social: some issue and comment data are missing.
Solved this exact issue with es_connector version 7.0-alpha4 within Drupal 10.2.3