What each cause means
The rule engine names one cause of record for each investigation.
AvailabilityAvailable to everyone with early access to Varqeni.
How it works
- Connector paused: the Debezium connector is paused. The autopilot may resume it if your policy allows.
- Task still failed after an outage: a task stayed failed after the outage that caused it cleared. The autopilot may restart it if your policy allows.
- Task failed: a connector task failed for another reason. A person looks at the error.
- Connect worker down: the Kafka Connect worker is not answering.
- Schema registry down: the schema registry is not answering, so changes cannot be written.
- CDC login or replication lost: the connector cannot log in to PostgreSQL or lost its replication rights.
- Kafka unavailable: the Kafka brokers are not answering. Varqeni recommends, it does not act.
- No fault: source idle and No fault: data flowing: nothing is wrong.
- Not enough evidence: Varqeni could not tell. A person takes over.
- Cleared before diagnosis: the fault went away before the investigation finished.
What you should see: Infrastructure investigations use their own causes, such as a virtual machine down, a disk filling, a node not ready or a certificate expiring.
What it can and cannot do
Can
- Name one cause of record for each investigation, from a fixed list built and tested on real faults.
Cannot
- Invent a cause outside the list. When it cannot tell, it says Not enough evidence.
Related articles
- Confidence: high, medium and lowEach cause comes with a confidence. It decides whether the autopilot may act and how sure you can be.
- The fixes Varqeni can runTwo fixes run on pipelines and four on infrastructure. Starting a stopped virtual machine is built and switched off.
- Read the estate under InfrastructureUnder Infrastructure, the console shows hosts, Kubernetes, data services, Kafka, autoscalers, load balancers, storage pools and certificates, with any findings.
Still stuck? Contact support or email info@fortisdatagroup.com.