pingoru pingoru Otera had 14 outages in the last 2 years totaling 58h 13m of downtime, averaging 0.6 incidents per month.
There were 14 Otera outages since August 28, 2025 totaling 58h 13m of downtime. Each is summarised below. Incident details, duration, and resolution information.
We are seeing small increase in agent error rates. We are currently investigating.
A fix has been implemented, we are currently monitoring.
The incident has been resolved.
There is an issue logging into Otera Studio. We are currently investigating. All other systems are working normally
We have identified the issue and working on a fix.
The incident has been resolved. Studio is now accessible. The issue was related to a gateway configuration change routing authorization through a different URL. We have updated the configuration to resolve the issue. This only affected Studio login, and had no impact on automations or agents.
The issue has been identified and fix is being implemented.
A fix has been deployed. We are monitoring.
We are continuing to monitor for any further issues.
The incident has been resolved.
We are seeing sporadic timeout issues with document uploader service affecting some of the document splitting agents tasks. We are investigating.
We have deployed a fix and currently monitoring. All processing is normal since the deployment.
The issue has been resolved.
We are seeing an increase in the document splitting agents fail rates. We are currently investigating. The issue is sporadic and affects a small percentage of requests.
The issue has been resolved. We are planning on further improvement to this agent.
At 10:48 CET, we started seeing issues related to some classification tasks failing. This seems to be related to an LLM returning errors. We are currently investigating. These only affect Classification agents.
The issue is identified. It is related to "confidence scores" in agent. Workaround: Disable the confidence scores in the affected agents. We are continuing to investigate to resolve it.
We have a fix being tested internally.
We are continuing to work on a fix for this issue.
A fix has been deployed, we are monitoring.
The incident has been resolved.
We are seeing issues with access to Otera Studio. Logins may take too long or fail. It is currently being worked on. Automations and execution are not affected.
Systems back to normal, we are currently monitoring.
All systems are fully operational.
We are currently investigating this issue.
This incident has been resolved.
We are currently investigating this issue.
We've increased infrastructure capacity. Executions are processing quickly, but there's a backlog in the queue. Performance will continue to improve over the next few hours as we clear it. We're actively monitoring and will update once fully resolved.
Resolved: Earlier today we experienced a significant increase in execution volume. During the ramp-up of infrastructure to handle the extra load, executions were temporarily slower than usual. Capacity has been scaled and performance is back to normal.
We are currently investigating this issue.
Issue traced to a recent deployment. We've rolled it back and are now monitoring executions to confirm recovery.
We are currently investigating this issue.
We identified that some HTTP clients were not following the redirect from *.deepopinion.ai to *.otera.ai correctly. To resolve this, we've changed our approach—requests to *.deepopinion.ai are now handled internally and will respond under the same domain, eliminating the need for client-side redirects.
The fix has been applied and is working as expected. All systems are now operating normally. We'll continue monitoring to ensure stability.
Issues have been resolved.
This incident has been resolved.
We’re currently experiencing performance issues resulting in slowness across DeepOpinion's Platform, and our team is working to mitigate the impact.
A fix has been applied, tested, monitored, and resolved the issue.
On 28 August 2025, DeepOpinion Studio was inaccessible from 10:30–11:20 CET due to a repository sync issue during a routine key rotation. This only impacted Studio access — customer automations were not affected. The issue is now resolved. We have updated our procedures to prevent similar occurrences.