WeSpeak - Service disruption — Virtual Assistant & CRM Panel (third-party provider) – Incident details

This incident has been resolved. Our infrastructure provider confirmed at 21:17 UTC that the underlying issue was fixed on their end, and our messaging pipeline has been operating normally since 18:09 UTC (15:09 Argentina time): incoming messages are reaching the platform immediately, the Virtual Assistant is replying, and agents can send replies from the CRM panel. All messages delayed during the two disruption windows (16:16–16:40 UTC and 17:51–18:09 UTC) were processed. We apologize for the inconvenience and thank you for your patience.

Our infrastructure provider reports that a fix has been deployed on their side and that they are monitoring to confirm the system is fully operational. Since 18:09 UTC (15:09 Argentina time) our messaging pipeline has been operating normally, with no further delays: incoming messages are reaching the platform immediately, the Virtual Assistant is replying, and agents can send replies from the CRM panel. We will keep this incident in monitoring until the provider confirms full resolution, and will post a final update then.

Service has recovered again and all messages delayed during the disruption have been processed. Today the messaging pipeline was interrupted during two windows: 16:16–16:40 UTC and 17:51–18:09 UTC (13:16–13:40 and 14:51–15:09 Argentina time). Incoming messages received during those windows reached the platform late, and the Virtual Assistant replied late. Our infrastructure provider has not yet confirmed full resolution on their side, so we are monitoring closely and will post a final update once they do.

The disruption at our infrastructure provider has recurred. Incoming messages are again delayed before reaching the platform and the Virtual Assistant is not replying while the delay lasts. Messages are queued, not lost. We are monitoring the provider's incident and will post an update as soon as service recovers.

Service has recovered: incoming messages are reaching the platform again, the Virtual Assistant is replying, and the messages delayed during the disruption have been processed. Our infrastructure provider has not yet closed its incident, so we continue to monitor closely.