Is Dixa Down?

## Summary Between 20 and 30 October 2026, some Instagram direct messages sent to our customers never reached Dixa. Meta, the company behind Instagram, had a fault that stopped it from passing these messages on to us. Meta confirmed the fault and fixed it on 30 October. Almost all affected accounts are receiving messages again. ## Who was affected The problem affected Instagram direct messages only. All other channels, including Facebook Messenger, worked normally. A small number of Instagram accounts stopped receiving messages. Some others received only part of their messages. When a message did not arrive, no conversation was created in Dixa, so your agents could not see it or reply to it. No messages were sent to the wrong place, and no data was exposed. Some affected accounts also had trouble sending replies on Instagram. This was a separate issue on Meta's side. ## Timeline All times are Copenhagen time. | Date | What happened | | --- | --- | | 20 October | The first Instagram account stopped receiving messages. Over the next week, more accounts followed. | | 27 October | Customers reported missing Instagram messages to our Support team. Our engineers started investigating the same day. | | 28 October | We conclued that Instagram was not sending us any messages for the affected accounts, and contacted Meta. | | 29 October | Meta confirmed the fault was on their side. | | 30 October, 07:24 | Messages started arriving again. | | 30 October, 15:12 | Meta confirmed the fault was fixed. | | 3 November | Most accounts are receiving messages again. We are working with Meta on the remaining accounts. | ## Why it happened When someone sends a direct message to your Instagram account, Instagram passes it on to Dixa. A fault on Meta's side stopped this for some Instagram accounts. Instagram simply stopped sending us the messages, so Dixa had nothing to receive. We carried out a thorough review of our own systems and found no issues. Dixa remained ready to receive messages throughout the incident, and no changes had been made on our side. Meta confirmed this on 29 October. It took a week to spot. Our monitoring looked for errors, and messages that never arrive do not cause errors. Many of the affected Instagram accounts also receive few messages, so a quiet inbox did not stand out as unusual. ## What we did * Checked all of our own systems to rule out a problem on Dixa's side. * Added extra tracking so we could see exactly which accounts were not receiving messages. * Reported the problem to Meta with detailed evidence and pushed for a fix. * Advised affected customers not to reconnect their Instagram accounts, on Meta's advice, as it would not help. As the fault originated within Meta’s systems, it was not something we were able to resolve directly. Meta implemented a fix on 30 October. ## What we are changing While we cannot prevent faults within Meta’s systems, we can significantly reduce the time it takes to detect them. To that end, we are introducing the following improvements: * **Automatic alerts.** We will be alerted when an Instagram or Facebook account that usually gets messages suddenly stops getting them. This means we can act before you notice. * **Better records.** We now keep a record of every message notification Meta sends us. This lets us show Meta exactly what is missing and get faster answers. * **Faster escalation.** We have found the quickest way to reach Meta's engineers and will use it first next time. ## What you should do now * **Check Instagram directly.** Messages sent to you between 20 and 30 October may be missing from Dixa. Open your Instagram inbox to find and answer them. * **No need to reconnect.** Your Instagram connection to Dixa does not need to be changed. * **Contact us if something looks wrong.** If your Instagram messages still are not arriving in Dixa, please contact Dixa Support. A small number of accounts are still affected, and we are working with Meta and the customers involved. We are sorry for the disruption this caused you and your customers. Thank you for your patience while we worked with Meta to fix it. --- [2026-10-30T17:24:57.409+01:00] (resolved) All known issues to this incident have been resolved. We thank you for your patience and cooperation. Post mortem about this incident will be posted within 5 business days. --- [2026-10-30T16:24:31.539+01:00] (monitoring) Meta has informed us that the issue affecting Instagram message delivery for some connected accounts has been fixed, and our own testing confirms that incoming and outgoing Instagram messages are flowing through Dixa again. We’re keeping this incident open while we monitor to make sure the fix holds. If you were affected, please check that new Instagram messages are arriving in Dixa as expected. Messages received while the issue was ongoing (since 28 October) may not have reached Dixa, so we recommend reviewing your Instagram inbox directly in Meta for any conversations from that period. If you’re still seeing issues, please contact Dixa Support. --- [2026-10-30T11:37:45.980+01:00] (identified) We’ve identified the cause of the intermittent issues affecting incoming and outgoing Instagram messages in Dixa. The issue is on Meta’s side and affects Instagram message delivery for some third-party connected accounts. Meta’s engineering team is actively working on a fix, and we’re in close contact with them. We’ll share an update as soon as Meta confirms a resolution. --- [2026-10-28T11:48:22.788+01:00] (investigating) We’re investigating intermittent issues affecting incoming and outgoing Instagram messages in Dixa. Only some Instagram connections are affected, and Facebook Messenger is working as expected. We recommend checking whether your Instagram messages are coming through as usual. If they aren’t, please manage those conversations directly in your Meta account for now. Reconnecting your Instagram page to Dixa won’t resolve the issue at this time. We’ll share an update as soon as we have more information.

## Summary On October 23, 2026, at 11:15 AM CEST, all requests to the Dixa Public API \(dev.dixa.io\) began failing with an error. The outage was caused by a faulty deployment of the component that authenticates incoming API requests. Our team detected the issue at 11:16 AM CEST, identified the cause, and deployed a fix. Full service was restored at 11:31 AM CEST. ## Impact Between 11:15 AM and 11:31 AM CEST, the Dixa Public API was unavailable for all customers. Every API request in this window, across all API versions, received an HTTP 500 error, and no request succeeded. Requests were rejected before they reached our systems, so no request was partially processed, and no data was lost or changed. Integrations that retry failed requests automatically will have recovered on their own. Integrations that do not retry may need to resend any calls made during this window. ## Root Cause The outage was caused by a manual deployment of the component that authenticates every request to the Dixa Public API. The new version of this component was packaged incorrectly, so it could not start. Without a working authentication step, every incoming API request failed. The code itself was correct; the problem was only in how it was packaged for deployment. ## Resolution Our team detected the issue at 11:16 AM CEST and began investigating the affected services. By 11:26 AM CEST, the root cause had been traced to the faulty deployment. A correctly packaged version of the same code was deployed at 11:30 AM CEST, and API requests began succeeding again immediately. Error rates were back to normal levels by 11:32 AM CEST. ## Preventive Measures To help prevent similar incidents going forward, we are: * Replacing manual deployments of this component with an automated deployment pipeline * Testing every change in a staging environment before it reaches production * Improving versioned releases, so we can roll back to a known-good version immediately * Adding further automated alerts on API authentication errors and error rates to detect issues faster ## Closing Note We understand how important reliable access to the Dixa Public API is to your integrations and workflows. We apologize for any inconvenience caused during this period and are committed to strengthening the resilience of this service. --- [2026-10-23T12:05:24.698+02:00] (resolved) On October 23, Dev API was unavailable from 11:15 to 11:31 CEST (16 minutes). The outage was caused by an error during a deployment step for one of our components. A postmortem will be published within 5 business days.