On March 18th, 2024, our Australian data center encountered a partial outage affecting a limited subset of data integration tasks. The issue stemmed from a single node within a cluster responsible for load balancing these tasks, which became unresponsive due to its inability to write new logs to disk. Functionality was restored to the affected node, and services resumed normal operation.
We are actively exploring strategies to enhance our monitoring capabilities and proactively maintain node health to mitigate any future instances of data integration failures.