Blip Desk [Low performance]
Timeline · 4 updates
- investigating Sep 29, 2026, 12:21 PM UTC
We are currently experiencing low performance on Blip Desk. Impact: Slowness when using Blip Desk. Update: Our support team is actively working on it.
- investigating Sep 29, 2026, 01:01 PM UTC
Issues with support interactions via Blip Desk: messages sent by agents are neither displayed on the screen nor recorded in the history/threads. Additionally, while messages sent by customers are logged, they do not appear to agents or encounter flow errors. High volume of warnings and server processing errors identified (envelope processing failures and "Destination not found" errors). Verification of Starter component configurations and instances, confirming isolated instances thus far.
- identified Sep 29, 2026, 01:43 PM UTC
Issues operating Blip Desk: messages sent by customers are logged but do not appear to the agent; agent responses neither reach customers nor save to the history; and contact data and history older than 90 days are unavailable. Actions taken: Identified spikes in server errors/warnings (approx. 60,000 records/min) related to processing failures and "Destination not found" errors. Mapped and defined the impact on the customer base and Desk workflows. Analyzed infrastructure, WebSocket communication, pod integrity, and the Starter component. Performed a controlled shutdown of Starter instances to isolate potential consumer concurrency issues in Confluent. Restored and normalized message flow for Desk and WhatsApp following the Starter shutdown.
- resolved Sep 29, 2026, 02:04 PM UTC
We identified a service outage affecting Blip Desk operations. This resulted in customer service difficulties: messages sent by customers failed to appear on the agent's screen, while responses sent by agents neither reached the customer nor were saved to the history; additionally, contact details and Desk history older than 90 days were not displayed. Actions taken: Mapping of error/warning spikes on the server (approximately 60,000 records/min) related to processing failures and "Destination not found" errors. Analysis of infrastructure, WebSocket connections, pod integrity, and Starter component configurations. Controlled shutdown of Starter instances to isolate concurrency issues regarding message consumption from Confluent topics. Restoration and validation of normal message flow for Desk and WhatsApp. Restart and final real-time monitoring tests to confirm environment stability.