Open m-kay opened 3 years ago
I'm experiencing the same issue. That makes error go unnoticed even when monitoring available metrics.
Same issue here! The connector status is wrong when SQL query fail!
With version 10.2.4 the task now fails on SQLNonTransientException
(see #1096), which resolves my issue since a failing task can be detected via the exported metrics.
Can you please explain about the setting that we need to make in the connector config to make the it fail in case of a SQL exception?
With version 10.2.4 and later you don't need to configure anything. The connector is stopping when SQL exceptions are thrown.
I don't see it is failing. Attaching a connector config where I passed an unknown table that really don't exist in the DB. But the connector is still running even though there is a SQL exception. My jdbc connector version is 10.5.1 . Any comments?
I'm running a JDBC source connector and try to monitor its status somehow via the exposed JMX metrics and a prometheus exporter. However the status of the connector and all its tasks are still in the running state when the query fails or db can't be reached.
In earlier versions it seems that no value for
source-record-poll-total
in the source-task-metrics was exported when the query failed and this could be used to detect such a failure. In the versions I use (connect-runtime-6.2.0-ccs, confluentinc-kafka-connect-jdbc-10.2.0, jmx_prometheus_javaagent-0.14.0) even when the query is failing the metric is exported with value 0.0 so there is no chance to detect when something is wrong with the database.