MongoDB had 68 outages in the last 2 years totaling 352h 19m of downtime — averaging 2.8 incidents per month.
There were 68 MongoDB outages since August 18, 2025 totaling 352h 19m of downtime. Each is summarised below — incident details, duration, and resolution information.
VoyageAI APIs are experiencing elevated 5xx errors. We are investigating the issue.
A fix has been applied and we are monitoring the system.
The system has returned to normal operation.
We are investigating an issue with cluster configuration updates getting delayed for a subset of users on free and flex tiers. Cluster access is unaffected.
We have identified the root cause and are rolling out a configuration change to remediate the delayed cluster configuration changes.
We have seen errors subsiding and cluster configurations for free and flex customers should be operating normally again.
This incident has been resolved.
We are investigating an issue affecting Atlas Stream Processing in Azure West Europe. Some users may see processors in this region encounter internal errors. We are actively investigating and will share another update as soon as we have more information.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
All VoyageAI models are experiencing elevated errors. We are investigating the issue.
We have fully recovered as of 16:41 PST.
This incident has been resolved.
We are currently investigating an issue with Azure clusters in westus region. Please refer to the azure status post here for more information https://azure.status.microsoft/en-us/status.
Azure has confirmed that there are networking issues in the westus region. Please follow the official azure status post for more detailed updates https://azure.status.microsoft/en-us/status
We are aware of an ongoing Azure networking incident affecting the West US region. Some Atlas clusters hosted in Azure West US may experience intermittent connectivity issues, including failures connecting to clusters, issues with communication between cluster nodes, nodes appearing down, and delays to cluster modifications. Azure is investigating the issue and evaluating mitigation options. We will provide additional updates as more information becomes available.
Azure is rolling out a mitigation for the ongoing networking incident affecting the West US region, and we are seeing signs of recovery. Some Atlas clusters hosted in Azure West US may continue to experience intermittent connectivity issues, including failures connecting to clusters, issues with communication between cluster nodes, nodes appearing down, and delays to cluster modifications. We are monitoring recovery closely and will provide additional updates as more information becomes available.
Azure has rolled out mitigation for the networking incident affecting the West US region and recovery is complete.
Due to an ongoing incident with AWS's billing data, MongoDB Atlas and Atlas for Government will not be updating AWS network or backup consumption data over this weekend, 2026-07-17 through 2026-07-20. The products themselves will remain fully available. When the upstream issue has been resolved, we intend to display and bill for the correct amounts. Charges will have usage dates that correctly express the usage, but will have the billed date of when this is corrected. Cluster and backup health are unaffected. Other cloud providers' billing data is unaffected. You can follow the upstream incident at https://health.aws.amazon.com/health/status
The upstream issue has been corrected. After approximately 0400 UTC on 2026-07-21, most, if not all, of the updated backup and network consumption on AWS will be reflected in MongoDB Atlas and Atlas for Government. Thank you for your patience during this incident.
This incident has been resolved.
We are currently investigating an issue affecting Stream Processing Azure West Europe. User initiated commands & workloads may see intermittent failures.
A fix has been implemented and we are monitoring the region.
The issue has been resolved.
We are currently investigating issues related to certificate issuing failures with Lets Encrypt. Tenant cluster provisioning is currently impacted.
We are no longer seeing failures with cert issuance from Lets Encrypt. We are still investigating the root cause of the original errors and will continue to monitor.
Let's Encrypt service has returned to normal.
Users of Atlas Data Federation and Online Archive using cloud-provider IAM authentication may see pipelines fail with an error when querying backing Atlas clusters. Other authentication methods are unaffected. As a temporary workaround, users can use SCRAM authentication.
We have identified that the issue is caused by a recent configuration change to IAM authentication and are working on a fix.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
We are currently monitoring the status of Lets Encrypt after an issue with their API. Tenant clusters and Atlas for Government clusters will experience delays in provisioning due to Lets Encrypt service degradation.
Let's Encrypt services have returned back to normal. Cluster operations have now returned to normal.
We are currently monitoring the status of Lets Encrypt. Tenant clusters and Atlas for Government clusters will experience delays in provisioning due to Lets Encrypt service degradation.
We've confirmed that Lets Encrypt has returned to normal operations. Cluster operations are now back to normal.
We are currently investigating the issue where AWS cluster backup and restores are failing. This includes exports and manual downloads as well as restoring to another cluster. We are also investigation potential impact to other operations on AWS clusters such as modifying instance size.
A fix has been implemented and we are monitoring the results.
This incident has been resolved. AWS clusters should no longer encounter errors during backup and restore.
From 2026-06-25 06:55 UTC to 2026-06-25 07:10 UTC, some users may have experienced intermittent Atlas UI loading failures. We identified elevated traffic to a number of endpoints that increased latency in the Atlas UI service. This caused internal health checks to fail and temporarily reduced healthy Atlas UI capacity. We restored healthy Atlas UI capacity and applied additional rate limiting to the affected traffic paths. We are continuing follow-up work to further reduce the risk of recurrence.
A fix has been implemented and we are monitoring the results
This incident has been resolved.
We are currently investigating issues connecting and querying to Atlas Data Federation and Online Archive instances in GCP regions
We've identify an issue with networking in GCP regions and are working on fix
Issues connecting to ADF and OA instances in GCP belgium-eu region has been mitigated, and working on other GCP regions.
All ADF and OA connections in GCP are reporting as successful, and we will continue to monitor.
Issue has been mitigated and connections should be working now to ADF and OA instances in GCP regions.
Starting approximately 17:00 UTC, some Atlas for Government users may be experiencing gaps in metrics sent to their DataDog Gov integration. We are investigating this issue.
The issue has been resolved.
We are investigating an issue impacting some clusters that causes search queries to fail with authentication failures.
We have identified the root cause of the search query failures and are working on a mitigation for impacted clusters. Clusters with dedicated search nodes are unaffected. Some clusters with colocated search nodes may see search query failures.
We have resolved the issue for all impacted clusters. We continue to monitor the fleet for any residual impact.
This incident has been resolved.
The data plane node logs (mongod, audit logs, etc.) are unavailable since May 14th 4am UTC for clusters in Azure, GCP, and AWS gov-cloud regions. We have identified the root cause and are actively working on a fix.
We have applied the fix. The data plane node logs will start to be available to customers within the next 30 minutes.
The data plane node log download has recovered.
AWS Availability Zone use1-az4 in US-EAST-1 region is impaired. For some clusters in AWS US-EAST-1 region, cluster nodes in that availability zone may have impaired health. Some Atlas operations, including instance scaling and cluster configuration changes, may also be delayed for clusters in AWS US-EAST-1. Please see https://health.aws.amazon.com/health/status for more information.
AWS Availability Zone use1-az4 in US-EAST-1 region is still impaired and AWS is working to restore the full availability of the availability zone. The Atlas system is taking healing actions on impacted nodes where possible to restore node health. We continue to monitor the situation. Please see https://health.aws.amazon.com/health/status for more information.
AWS continues to make progress to restore the full availability of the availability zone. The Atlas system is taking healing actions on impacted nodes where possible to restore node health. We continue to monitor the situation. Please see https://health.aws.amazon.com/health/status for more information.
AWS continues to work towards full recovery. The Atlas system has been successfully taking healing actions on impacted nodes to restore node health and using healthy availability zones for provisioning new nodes. We continue to monitor the situation. Please see https://health.aws.amazon.com/health/status for more information.
This incident has been resolved.
DNS changes for AWS clusters are currently degraded. We've identified the issue and are pushing out a patch.
DNS changes for AWS clusters are currently degraded. Cluster topology modifications may be delayed or not able to complete successfully. We've identified the issue and are pushing out a patch.
Mitigation is in progress; recovery for DNS operations and downstream cluster topology changes has begun.
We are still seeing cases where new cluster update requests are failing to go through due to ACME related DNS issues. We are actively investigating this issue.
We've identified a bad piece of metadata causing DNS issues. We've since patched this metadata and are seeing recovery in ACME updates for cluster topology changes. We will continue to monitor recovery.
We are continuing to monitor for any further issues.
DNS updates for ACME are confirmed healthy. Cluster updates are now proceeding as expected.
Identified - We have identified an issue causing failures for a subset of Atlas backup export jobs. We are actively working on a fix. Your cluster health and data are unaffected. We will provide another update by May 7 01:30 UTC
We are continuing to work on the fix for this issue. We will have another update by May 7 16:00 UTC.
A fix has been implemented and we are monitoring the results.
This incident has been resolved.
We are observing widespread issues in in Azure West Europe. Customers may see nodes reporting as "down" and delayed cluster modifications for Azure clusters in West Europe. There are three total AZs in Azure West Europe, and primary failover will be handled by MongoDB.
We're starting to see instances recover. We will continue to monitor the region's status in Azure and ensure Atlas takes any available steps to maintain cluster health.
We are no longer observing these issues in Azure West Europe. Atlas Clusters in the region should be healthy and fully operational.
We have identified a bug affecting some Atlas customers that resulted in higher-than-expected backup charges on invoices for the billing date of April 24, 2026. This is a billing-only issue; your backup configuration, cluster health, and data are unaffected. We are actively working to correct all affected invoices. Any overcharges will be automatically refunded — no action is required on your end. Please contact MongoDB Support with any questions. We will post another update by April 24 at 22:00 UTC.
We have resolved the issue that caused the overage and usages should be reflected correctly by April 28th. If you still see incorrect usages after this date, please reach out to us.
We are currently investigating reports in MongoDB Atlas for Government that the Business Intelligence Connector (BI Connector) is failing. Affected users will have issues connecting to and querying their BI Connectors. No workaround has been identified at this time. Underlying cluster health is not affected.
We have identified the issue affecting the Business Intelligence (BI) Connector in MongoDB Atlas for Government. We are working on a fix. While no workaround exists, data remains accessible to MongoDB clients. Affected users will not be able to connect to their clusters' BI Connectors. Cluster health is unaffected.
A fix has been implemented and we are monitoring the results. The launch of the fix should be complete by approximately 13:00 UTC 21 April 2026.
This incident has been resolved.
Some Atlas users may be unable to access Charts and the Visualization page due to repeated login prompts or unauthorized errors when attempting to sign in. Our engineers have identified a recent change as the likely cause, have rolled it back, and are monitoring the service to confirm that normal access has been restored.