Ubidots had 20 outages in the last 2 years totaling 3310h 5m of downtime — averaging 0.8 incidents per month.
There were 20 Ubidots outages since September 5, 2024 totaling 3310h 5m of downtime. Each is summarised below — incident details, duration, and resolution information.
We have identified an error with our MQTT publish service and are currently looking into the issue.
A fix has been implemented and we are monitoring the results.
We have identified an error with our HTTP data ingestion service and are currently looking into the issue.
A fix has been implemented and we are monitoring the results.
We are currently experiencing a service disruption affecting Ubidots (industrial.ubidots.com) and associated apps. Users may be seeing 503 Service Unavailable errors. Our team is actively investigating a connectivity issue between our connection pooler and the database cluster, following a scheduled PostgreSQL upgrade operation. We are working to restore full service as quickly as possible. We will provide updates every 15–20 minutes. We apologize for the inconvenience.
The service has been restored and we are monitoring the platform for full stability. The issue was caused by a configuration incompatibility introduced during our PostgreSQL upgrade. The database server was requiring SSL connections, which prevented our connection pooler from establishing a successful connection. We resolved the issue by updating the database configuration and upgrading the connection pooler. If you are still experiencing issues, please reach out to [email protected]. We will confirm full resolution shortly.
This incident has been resolved.
We have identified an error with our MQTT publish service and are currently looking into the issue.
A fix has been implemented and we are monitoring the results.
We are continuing to monitor for any further issues.
This incident has been resolved.
We have identified an error with our MQTT data ingestion service and Events engine are currently looking into the issue.
The issue has been identified and a fix is being implemented.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
We have identified an error with our TCP data ingestion service and are currently looking into the issue.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
Earlier today, we identified increased load on one of our underlying database nodes, which led to higher-than-usual latency across dashboards and Synthetic Variables processing in the US deployment. As a result, synthetic variable calculations and updates are delayed, and some dashboards may have experienced slower response times while corrective maintenance is performed to stabilize the platform. The system is gradually returning to normal operation, and we are closely monitoring performance to ensure stability and prevent further impact. We appreciate your patience and understanding.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
We’ve identified an issue affecting the UbiFunction module. Our DevOps team is currently investigating the root cause.
We have restarted the load balancer for the UbiFunction module. This has restored the operation in the module.
Summary: We identified an issue that limited the number of active connections to one of our servers. The root cause was a kernel-level restriction on the maximum number of open files (file descriptors), which also affects network sockets. Impact: The limitation caused a temporary reduction in the number of simultaneous connections the server could handle, potentially affecting availability for some users. Resolution: The kernel and system-level limits for open files were increased to ensure sufficient capacity for concurrent connections. After applying the changes and restarting the affected services, the system returned to normal operation. Preventive Actions: We’ve updated our configuration management to enforce higher default limits across all environments, preventing similar constraints in the future. Status: This incident has been fully resolved, and all systems are operating normally.
Incident window: September 25, 2025 – 7:00 PM (UTC-5) to September 26, 2025 – 4:20 AM (UTC-5) Impact: During the incident window, MQTT subscriptions experienced disruptions that prevented data from being delivered in real time. Devices attempting to subscribe or maintain subscriptions may have encountered failures, delays, or missing data. Resolution: Our engineering team identified abnormal connection behavior that led to message queue saturation within the MQTT service. To restore functionality, queues were purged, affected MQTT clients were reset, and supporting services were restarted. Service was fully restored by 4:20 AM (UTC-5) on September 26. Current Status: The MQTT service is operating normally, and monitoring has been reinforced to ensure stability.
We are investigating reports of intermittent issues loading our website at ubidots.com, stemming from a third-party provider.
Our third party website provider Webflow has declared an incident causing widespread outages across multiple websites they serve. As a result, our website (https://ubidots.com) is intermittent. This does not affect our application, blog, community, or documentation; only our website. We are closely monitoring this internally and will continue to provide updates as they become available
Webflow has implemented a fix which mean our website (ubidots.com) is working as expected.
We've identified latencies in the platform. We're unsure of the reason at this moment. Out team is already investigating. We will publish an update in the next 30 minutes.
The issue has been identified and a fix is being implemented.
This incident has been resolved.
We have received reports pointing out that the custom widgets built on top of the HTML canvas and Widget plugins module aren't working. The reason is that the CDN serving the built-in Ubidots library, that bridges the widgets with the dashboard controls and allows interaction with the dashboard, stopped working. We are: – Uploading the library to a new CDN – Replacing the URL in the HTML canvas widget and Widget plugin module.
A fix has been implemented and we are monitoring the results.
We have replaced old CDN links, serving libraries used in the HTML canvas widget, Widget plugins and Pages, for new one that do not error.
We've identified latencies in the platform. We're unsure of the reason at this moment. Out team is already investigating. We will publish an update in the next 30 minutes.
The platform is back operational. The latency was raise because of an unexpected spike in request to create variables. Our team quickly caught this situation in order to prevent further spreading of the latency. There are measures being taken to prevent this from happening again, particularly, we will isolate those variable creation request in a dedicated infraestrucutre.
We're experiencing a generalized latency in the platform. Our DevOps team is already working on resolving this issue. We will provide additional information as it becomes available.
This incident has been resolved.
The AU deployment experienced an outage because of a sudden spike in data reception, which overloaded the servers and hence the web app was slow to read and display data. Start time: 10:30 PM UTC-5 End time: 11:15 PM UTC-5 Total time: 45 min During this time: – All data was correctly saved. – Real time update in dashboard was interrupted. – The events engine was taking up to 4 minutes to trigger events.
Our DevOps team ran a maintenance task consisting of updating our servers packages. This is standard procedure to guarantee the instances are up to day with the latest security patches or package versions. This task took about 5 minutes to complete, where reading data was slow. No data was lost during this time window.
We have identified an error with our Events service and are currently looking into the issue.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
We have identified an error with our HTTP data ingestion service and are currently looking into the issue.
A fix has been implemented and we are monitoring the results.
We are continuing to monitor for any further issues.
This incident has been resolved.