Is Amplitude Down Right Now? Check if there is a current outage ongoing.

Monitoring GCP incidence for recovery

Identified: Amplitude bigquery export is failing due to ongoing GCP incidence https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1SsW starting from Jun/12 11:08AM PSP

Our data processing systems are delayed . This incident started at 18:40. US customers will see delayed events. We are investigating the issue and will post an update in an hour or earlier if we identify the issue.

We wanted to inform you that we've identified an unexpected issue where a "[Segment]" prefix has been added to some event types starting earlier this morning. Our team is actively investigating this and working closely with the Segment team to resolve the issue as quickly as possible. We understand the importance of consistent event naming for your workflows and analytics, and we're treating this with high priority. We'll keep you updated as we make progress, and will notify you as soon as the issue is resolved. If you have any questions or need assistance in the meantime, feel free to reach out. Thank you for your patience and understanding.

From 2025-05-24 08:10 to 2025-05-24 08:33, we experienced issues with our query capabilities and users may see their charts/dashboards slow to load or unavailable. From 2025-05-24 08:34, we mitigated the issue by skipping realtime system, so query was working except missing recent realtime data(~ last 2 hours' data). Data is not missing and recovered once the realtime system is back. At 08:44, the system was fully recovered. No action is needed from customers at this time.

From 2025-05-24 08:10 to 2025-05-24 08:33, we experienced issues with our query capabilities and users may see their charts/dashboards slow to load or unavailable.From 2025-05-24 08:34, we mitigated the issue by skipping realtime system, so query was working except missing recent realtime data(~ last 2 hours' data). Data is not missing and recovered once the realtime system is back. At 08:44, the system was fully recovered. No action is needed from customers at this time.

All warehouse exports have caught up since 2:00PM PDT. This incident has been resolved. The issue was caused by an unusually large export request from one customer, which created a bottleneck in our data transformation jobs. This led to delays across both the Export API and Data Warehouse Exports. While the system has since been scaled to resolve the lag, we're working on improving monitoring and safeguards to better handle similar situations in the future.

Our Export API lag has fully recovered. We’re continuing to monitor the Data Warehouse Exports. We’ll provide another update once those jobs have fully recovered as well.

Starting at 1:40 AM PDT, our data export processing began experiencing significant lag. This has impacted both the Export API and Data Warehouse Exports, with delays potentially reaching up to 10 hours. We've identified the underlying bottleneck and have scaled up the affected system. As a result, we expect the lag to decrease significantly within the next 30 minutes. Our team is closely monitoring the recovery and will continue to provide updates as the situation improves. No data is lost.

Starting at 1:40 AM PDT, our data export processing began experiencing significant lag. This has impacted both the Export API and Data Warehouse Exports, with delays potentially reaching up to 10 hours. We've identified the underlying bottleneck and have scaled up the affected system. As a result, we expect the lag to decrease significantly within the next 30 minutes. Our team is closely monitoring the recovery and will continue to provide updates as the situation improves. No data is lost.

We will be undergoing scheduled maintenance during May 23rd 12:00 AM UTC

On May 23rd, 12:00 AM UTC, we will be performing maintenance on components of the core ingestion infrastructure in our US data center.During this time, customers may see data processing delays, and some temporary errors in the data collection endpoints could happen but are not expected. No data loss is expected during the maintenance.The EU data center won't be affected.We will post an update when the maintenance is completed.

Scheduled maintenance is currently in progress. We will provide updates as necessary.

On May 20th, between 1:07 and 1:17 AM UTC, Amplitude customers sending data to the US data center data collection endpoints experienced a significant spike in HTTP 500 response codes. This was quickly resolved after addressing an issue within our internal systems. For requests that received an HTTP 200 response, no data has been lost.

On May 20th, between 1:07 and 1:17 AM UTC, Amplitude customers sending data to the US data center data collection endpoints experienced a significant spike in HTTP 500 response codes. This was quickly resolved after addressing an issue within our internal systems. For requests that received an HTTP 200 response, no data has been lost.

This incident has been resolved.

A fix has been implemented and we are monitoring the results.

Since 2025-05-06 10:00 PT (17:00 UTC) Amplitude Web Experiments have been sending Impression events to Amplitude for the control variant of experiments on pages that are not targeted by the experiment.

Since 2025-05-06 10:00 PT (17:00 UTC) Amplitude Web Experiments have been sending Impression events to Amplitude for the control variant of experiments on pages that are not targeted by the experiment.

Realtime ingestion has fully caught up. No data was lost, and no further action is required from customers.

A fix has been implemented and we are monitoring the results.

Our realtime ingestion is delayed. This incident started at 5:40PM PST. Current status: a) Our ingestion systems are working as expected and no data is lost. Our processing systems are not processing newly ingested data. b) Most of our customers are impacted. c) Impacted customers will see delayed metrics for the last 40 to 60 minutes. We are investigating the issue and will post an update in an hour or earlier.

Our realtime ingestion is delayed. This incident started at 5:40PM PST. Current status: a) Our ingestion systems are working as expected and no data is lost. Our processing systems are not processing newly ingested data. b) Most of our customers are impacted.c) Impacted customers will see delayed metrics for the last 40 to 60 minutes. We are investigating the issue and will post an update in an hour or earlier.

This incident has been resolved.

System is back to healthy, we are monitoring

From 2025-04-24 10:17 to 2025-04-24 10:28, we experienced issues with our query capabilities and users may see their charts/dashboards slow to load or unavailable. From 2025-04-24 10:28, we mitigated the issue by skipping realtime system, so query should be good now, except missing recent realtime data(~ last 2 hours' data). Data is not missing, will be recovered once the realtime system is back Our team is working on a fix and no data is lost during this incident. No action is needed from customers at this time. We will post updates as we recover the systems.

From 2025-04-24 10:17 to 2025-04-24 10:28, we experienced issues with our query capabilities and users may see their charts/dashboards slow to load or unavailable.From 2025-04-24 10:28, we mitigated the issue by skipping realtime system, so query should be good now, except missing recent realtime data(~ last 2 hours' data). Data is not missing, will be recovered once the realtime system is back Our team is working on a fix and no data is lost during this incident. No action is needed from customers at this time.We will post updates as we recover the systems.

From 19:00 to 21:00 April 12th UTC, we will be performing maintenance on components of the core ingestion infrastructure in our US data center.During this time, customers may see data processing delays, and some temporary errors in the data collection endpoints could happen but are not expected. No data loss is expected during the maintenance.The EU data center won't be affected.We will post an update when the maintenance is completed.

Scheduled maintenance is currently in progress. We will provide updates as necessary.

The scheduled maintenance has been completed.

Verification is currently underway for the maintenance items.

Scheduled maintenance is currently in progress. We will provide updates as necessary.

On Saturday March 22nd 2025 from 9AM PST to 12PM PST, we will be performing maintenance on some of our critical production databases. During this time, the following impacts are expected:Our product may be temporarily unavailable.Data ingestion may experience delays; however, no data loss is expected.

This incident has been resolved.

A fix has been implemented and we are monitoring the results.

We have identified the root cause of the issue and have developed a fix. Our team is currently rolling out the fix to all affected customers.

Our team is actively investigating the issue to identify the root cause and restore full access as soon as possible. We will provide further updates as we learn more. Thank you for your patience.

The fix has been implemented and the issue has been resolved.

The issue has been identified and a fix is being implemented.

From 17:30 March 2nd, 2025 UTC, our EU data processing component is slower than usual and all customers are seeing a delay in recent data in their charts, streaming exports, or experiments. As of now, we are not expecting any data loss and further actions are not necessary from our customers. We are actively working on this issue and will post another update in the next 1 hour. No impact to the US customers.

This incident has been resolved.

We've identified a regression in our session chart that went out to production around 9:30am PST. A subset of session charts were throwing 500s. We rolled back the change at 11:05am. We are monitoring the errors right now.

This incident has been resolved.

A fix has been implemented and we are monitoring the results.

The issue has been identified and a fix is being implemented.

We are experiencing a delay in data export. We are actively working on fixing it. We expect no data loss and a later export will catch up. No further action is required from our customers.

For some users, some web pages were partially down. The issue started around 12:55 PST and was resolved at 1:39 PST. a) Our ingestion systems worked as expected, and no data was lost. c) Impacted customers might have experienced errors when loading Amplitude web pages.

The incident has been resolved.

We have implemented a fix. Real-time data may be delayed by up to 3 hours. We are expected to fully recover by 2 PM PT.

We have implemented and monitoring the aforementioned fix. We are looking to get customers access to their complete real-time data in queries soon.

We have identified the root cause and implemented mitigation measures. Some customers might see partial real-time data in charts and analytics queries.

Most of our Analytics charts are not loading and failing with an error. Our ingestion systems are working as expected and we are not losing any data. We have identified the issue and are working on resolving it right now. We will share an update in another 30 mins.

We have resolved the intermittent loading issues with the in-product help widget.

Some users may be experiencing intermittent issues with loading the in-product help widget. We are actively investigating. In the meantime, users can still access our documentation at amplitude.com/docs and get help at support.amplitude.com. Our web applications and data ingestion continue to be functional.

We have updated the Google login flow to use callbacks which has resolved this issue. Google sign in can again be used to create new orgs and log in to existing orgs. No data was lost and customers continue to have all their objects inside Amplitude as expected. We sincerely apologize for all inconvenience caused by this incident.

Our web applications are functional and nominal, and we have added a note to our sign-in page about the current issue with Google login. Users can still login via email and a magic link, so we have marked our web applications as operational. We are continuing to investigate a fix for the Google login issue.

We are continuing to investigate this issue.

We're currently experiencing intermittent issues with Google Sign-In. Our team is actively working to resolve this as soon as possible. In the meantime, you can log in directly using your email address to retrieve a magic link. Thank you for your patience and understanding!

We have identified the issue and are working towards resolution.

Logging into Amplitude via Google is currently not functional. We are investigating.

This incident has been resolved.

The incident happens in the US data center. And there is no data loss.

We are experiencing internal latency when importing customers' data since 2024-12-03 8:30 am UTC. For customers using Data Warehouse Import and Cloud Storage Import, export time to Data Warehouse and Cloud Storage integrated services is elevated (up to 19 hours delay). The root cause is identified and the remediation is applied. We expect to drain unfinished jobs in 2 hour.

On November 11th, 2024, from 2:15 PM to 2:24 PM PDT, customers experienced issues with our query capabilities. Users may have seen their Event Segmentation charts unavailable. No data is lost, and the problem has been identified and resolved.

At November 8th, 2024 4:37 PM to 4:46 PM PDT, customers experienced an increase in HTTP 5XX responses to requests to our data reception endpoints in the US data center. The problem has been identified and resolved. All requests that saw a 200 response have been successfully received for processing.

Incident Summary On 10/30 , the Management API service experienced an unexpected outage from 12:35PM PST to 1:57PM PST, impacting users' ability to access and manage data through this service. Issue Identified Our engineering team identified the root cause of the incident as misconfiguration. After detecting the issue, we took immediate steps to restore service. Resolution We rolled back the recent deployment, and as of 1:57PM PST, the Management API is fully operational. We are monitoring the service closely to ensure ongoing stability. Status Resolved. No further impact expected.

From 16:30 till 17:52 Oct 16th, 2024 PDT, our US data processing component was slower that usual and all US customers experience up to 87 minutes of latency in their incoming data in their charts, streaming exports, or experiments. There was no data loss during this incident and further actions are not necessary from our customers.

Our realtime pipeline has caught up. We are monitoring the /batch pipeline to catch up for full recovery. ETA is about one hour from now.

We have identified the root cause and deployed a fix. We are now monitoring the system to clear the data backlog. We do not expect to have any data loss during this incident and further actions from our customers are not necessary. Next update will be in about 2 hours

We are continuing to investigate this issue.

From 16:30 Oct 16th, 2024 PDT, our US data processing component is slower that usual and all customers are missing the last one hour of their incoming data in their charts, straaming exports, or experiments. As of now, we are not expecting any data loss and further actions are not necessary from our customers. We are actively working on this issue and will post another update in the next 1 hour.

All the issues should have been resolved now.

The deployment is completed. We are cleaning up and verifying.

We are testing the fix now. The success rate of the endpoint has been above 99% since 2:00pm PDT.

Our batch endpoint started to reject some of the events being sent to Amplitude. The issue started around 13:18 PDT on Sep 17th. All our customers with the US endpoint are impacted. We have identified the issue and fixing it now. We recommend retrying failed events until you receive a 200 response from Amplitude. Amplitude SDKs already take care of the retry logic.

This incident has been resolved.

We've restored querying on realtime data as well since 8/26 12:03AM PT. We are continuing to monitor the situation. No data is lost and no action is required by our customers.

Since 8/25 10:55PM PT, we started experiencing issues with our query capabilities and users may have seen their charts/dashboards slow to load or unavailable. Our team already identified the issue and is working on a fix. Since 11:15PM, we have partially restored query capabilities so that only data ingested after 8/24 is still missing in the chart results. No data is lost during this incident. No action is needed from customers at this time. We will post updates after 30 minutes or earlier as we recover the systems.

Our application website was down between 10:34am PDT and 10:50am PDT on Aug 23rd. The issue has been identified and resolved. No data has been lost and no customer action is needed at this time.

This incident has been resolved.

We've identified the issue and applied a fix to production at 1:34PM. Our website has been up again since then. We are continuing to monitor the situation. No data is lost and no action is required by our customers.

Our website is down. The issue started around 1PM PDT/PST today. a) Our ingestion systems are working as expected and we are not losing any data. b) All of our customers are impacted. c) Impacted customers will experience a loading bar or errors when loading Amplitude web-pages. We are investigating the issue and will post an update after 30 minutes or earlier if we identify the issue.

This incident has been resolved and the marketing and documentation websites are back online.

At 12:20pm PT our marketing website and documentation websites (amplitude.com) went out due to a vendor outage. The vendor has identified the root cause and is working towards a fix. We expect the site to return to a full functioning state soon.

On Aug 6th 2024 at 23:18:40 UTC Amplitude Experiment's Evaluation and Flag Configuration APIs in the US data center experienced an outage which caused increased latency and error rate for 5 minutes. The issue fully resolved at 23:24:20.

This incident has been resolved.

We've identified the root cause and applied a fix at 02:25PM PT. Query/chart latency should be back to normal now. Most charts are missing real time data from the last 45 minutes because we temporarily paused realtime processing as a remediation. Real time data is catching up now. No data is lost. We will post updates as we recover the systems.

Since 01:50PM PT, we started experiencing issues with our query capabilities and users may see their charts/dashboards slow to load or unavailable. Our team is working on a fix and no data is lost during this incident. No action is needed from customers at this time. We will post updates as we recover the systems.

Our data processing system for Session Replay was delayed due to a bug. The issue started at 3:30PM and was resolved at 7:30PM, at which point the system has fully recovered and processing has fully caught up. During this time frame, replays can experience a longer than usual lag (up to 20mins) before becoming available in the UI. Users may hit "Session Replay Not Found" error while the replays are being processed. There was no data loss during the incident. Amplitude Event ingestion was not impacted.

This incident has been resolved. No data was lost and no further action is required on the customers' end.

We've deployed a fix and are monitoring the results.

We've identified a production bug for team spaces which resulted in all team spaces failing to load. We are working on a fix right now. No data is lost.

On May 2, 2024 from 6:20am PT to 7:25am PT, users were unable to load Amplitude Data (app.amplitude.com/data) on our US Data Centers. The root cause was traced to issues with a database used by the web application that caused new connections to fail. We have since restored access to Amplitude Data. There was not any downtime with data ingestion or data loss.

Realtime lag is fully catch up. All systems should be recovered now.

Realtime lag is coming down to less than 1 hour now, we expect it fully recover in ~20 mins

We identified the root cause and had a fix, the system should start to recovering now. Realtime data is delaying for almost 2 hours, it's catching up now.

We are still investigating the root cause, did some mitigation, so some customers could see recovery, but realtime data is still missing from query result, no data lost. During the investigation, we disabled cohort sync from 6:45 am to 7:20am PT to test a theory.

Since 2024-04-30 05:20 PT (12:20 UTC), we are currently experiencing issues with our query capabilities and users may see their charts/dashboards slow to load or unavailable. Our team is working on a fix and no data is lost during this incident. No action is needed from customers at this time. We will post updates as we recover the systems.

The incident has been resolved. The query cluster's latency has returned to normal since 11:18AM. No data was lost and no further action is required from customers.

We have applied the fix and so far the query cluster has returned back to normal. We will keep monitoring the situation.

Starting at 9:45am PT, one of our query clusters started experiencing extended latency, affecting queries from all customers assigned to the cluster. We've identified the issue and is applying a fix.

At 4:49pm PDT on April 2 2024, Amplitude’s US data center experienced a service interruption after a series of metadata tables were accidentally deleted. Customers on the US data center were unable to access the Amplitude platform—including Analytics, CDP, Experiment, and Session Replay. Our EMEA data center was not impacted. The service came back online at ~11.30pm PDT the same day, and we began processing the data received—but not ingested—during the outage. As a result, there was a lag in performance and/or limited availability of a small amount of metadata for some users as we worked to fully restore the service. As of 12:15am PDT on April 4 2024, Amplitude is running at full capacity and we are conducting a full root cause analysis to ensure this doesn’t happen again.

A small percentage of the metadata, including event types or property names introduced between 12:30pm and 5:00pm PDT on April 2, may still be temporarily absent from charts and dashboards. We expect this issue to be resolved by Thursday, 4/4. We will provide another update by 9am PT.

A small percentage of the metadata, including event types or property names, introduced between 7:10am and 5:00pm PDT on April 2 may still be temporarily absent from charts and dashboards. We are continuing to work on this issue and will provide another update at 6:00pm PDT.

We are done processing the data collected during the outage, and platform performance should be back to normal. A small percentage of the metadata, including event types or property names, introduced between 7:10am and 5:00pm PDT on April 2 may still be temporarily absent from charts and dashboards. We are working on this issue and will provide another update at 4:00pm PDT.

We are still processing the data collected during the outage. We expect this to be complete within 2 hours. Our next update will be at 2:00pm PDT.

We are still processing the data collected during the outage. We expect this to be complete within 1-2 hours. Our next update will be at 12:00pm PDT.

We are still processing the data collected during the outage. We expect this to be complete within 1-2 hours. Our next update will be at 12:00pm PDT.

We are still processing the data collected during the outage. We expect this to be complete within 1-2 hours. Our next update will be at 12:00pm PDT.

We are still processing the collected data during the outage. We estimate the data processing will catch up before 10 AM PDT on Apr 3rd. Next update will be on 10:00 AM PDT.

We are now processing the collected data during the outage. We estimate the data processing will catch up before 7AM PDT on Apr 3rd. Next update will be on 7:30 AM PDT.

We are continuing to monitor for any further issues.

We have successfully restored our database and are working on resuming services. Most our services should become available momentarily. Given that several hours' worth of data was received during the outage but not processed, there may still be a minor lag in performance and/or availability for some users as we work to fully restore the service. Small percentage of data between 7:10 AM PDT and 5 PM on Apr 2nd might not show in charts and dashboards temporarily. We are still working on data processing and recovery and we will provide the next update in 2 hours.

We are still in the process of restoring the data. Next update will be posted here in 2 hours.

At 4:49 PM PDT, a series of metadata tables were deleted that caused a service interruption to Amplitude. At this time, we believe we have a fix and we will continue to update the Status Page with new details as they are available. Note that we are still receiving data and it will be available when we come back online. We apologize for the disruption to your service experience. Next update will be posted here in 2 hours.

We are continuing to work on a fix for this issue.

On 16:49 PM PDT Apr 2nd, all our customers experienced a wide service outage across our analytics, experiment, CDP and session replay. We have identified the root cause and working on the remediation actions. Our EU customers are not affected. Next update will be in the next 2 hours.

From 2024-04-01 16:38 PT(23:38 UTC) to 2024-04-01 19:35 PT, some customers were missing April 1st's data in our data processing component. The data were not showing up in charts/dashboards and were not being streamed or exported to downstream destinations. Our team already identified and fixed the issue and no data is lost during this incident. If you still experience missing data, please refresh your charts.

This incident has been resolved.

A fix has been implemented and we are monitoring the results.

The issue has been identified and a fix is being implemented.

From 3:25 PM on Mar 27th, 2024 PDT, we are experiencing delay in data processing, event streaming and scheduled cohort syncs. We are working to address the issue right now. We expect no data loss during this incident, and no further actions are required by our customers.

A deployment to Experiment's evaluation service on March 22 2024, 21:48 UTC caused on of the endpoints for serving flag configurations to local evaluation SDKs to return an empty empty array of configurations. Certain versions of SDK which use this endpoint to run local evaluations would experience null results for all flags. The issue was identified and rolled back at 23:19 UTC.

From 7:02 AM till 12:40 PM PDT, March 15th 2024, all our customers using our US data center have experienced a data processing and event streaming latency of up to 91 minutes. This was due to some malformed data sent by some customers. All our components are now caught up and work normally. We have not had any data loss during this incident.

We are continuing to monitor for any further issues.

We have identified the root cause and the data processing latency is recovering. Next update will be before 11:30 AM PT

From 7:02 AM March 15th 2024, all our customers are experiencing a growing latency of up to 60 minutes in our data processing component. We do not expect any data loss. We are investigating the issue and will provide an update before 10 AM PT

This incident has been resolved.

From 14:30 to 15:36, especially from 15: 25 to 15:45, you may experience slowness in queries. We already identified and fixed the issue, there is no data lost. If you refresh your charts, it should go back to normal.

Access to Analytics and Experiment Web Apps is now restored. There was NO identified data loss.

We are currently investigating the issue where customers in the EU cannot log-in to the Amplitude Web Application. We have identified that event ingestion is not impacted, there is NO identified data loss at this time.

This incident has been resolved.

EU event collection endpoint should be full recovered now. We are continuing to investigate more about the root cause and how to avoid in the future.

EU event collection endpoint should be mostly recovered now. We are continuing to investigate the root cause.

5xxs should be improving right now. We are continuing to investigate the root cause.

We are continuing to investigate this issue.

Our event servers started to return 5xxs and we are rejecting most of the events being sent to Amplitude. The issue started around 12:12pm PST on March 4th. All our customers with the EU endpoint are impacted. We are investigating the issue and will post an update in an hour or earlier if we identify the issue. We recommend retrying failed events until you receive a 200 response from Amplitude. Amplitude SDKs already take care of the retry logic.

This incident has been resolved.

A fix has been implemented and we are monitoring the results.

Our system is having some slowness, we have identified the problem and are working on the mitigation now.

This incident has been resolved.

The system should be back now, we will keep monitoring for a while to make sure it works well.

We deployed a bad bug causing query being slow or not stable. We are in the progress of rolling back, will update soon.

System should be fully recovered.

Rollback finished and we also re-enabled realtime data.

We disabled realtime data temporarily during the fix, you will see "incomplete realtime data" as a warning in charts

We did a deployment at 6:30 am and it has a bug causing chart loading much slower. We are rolling back right now, it will take 20 mins to finish.

This incident has been resolved.

Website is back at 23:27, we are still monitoring it

A lot of unknown traffic come into our system suddenly, we are trying to block the traffic and also scale up our system. You can still experience down time transiently.

https://analytics.amplitude.com is currently down and showing a 502 bad gateway error, we are investigating now. Will update soon.

This incident has been resolved. Starting at 4PM PT on Friday 1/12, customers may have experienced missing events or properties in dropdowns on our legacy chart controls and user lookup user interfaces. This incident is now fully resolved as of 1/15 at 2:58PM PT. No data loss occurred during this incident.

A fix has been implemented and we are currently monitoring to verify recovery.

We've identified the root cause of the issue and are working to deploy a fix.

We are experiencing an issue with our chart and user-lookup user interface that is causing some events or properties to become hidden. We are actively investigating the issue and will provide an update once the root cause is identified. No data loss is expected.

This incident has been resolved.

We have deployed a fix and are seeing resolution. Continuing to monitor until we see a full resolution.

We are continuing to work on a fix for this issue.

We are currently experiencing a web application outage. We have identified the issue and are deploying a fix. We expect the issue to be resolved soon. No data loss is expected.

This incident has been resolved.

We are continuing to monitor for any further issues.

From 6:45 PM on Dec 9th, 2023, most of our services are performing in a degraded state. This includes, but is not limited to, up to 80 minutes of data processing and event streaming lag and slower than usual chart performance. This is due to an AWS DynamoDB outage reported by AWS. We expect no data loss during this incident, and no further actions are required by our customers. We are monitoring the health of AWS DynamoDB. The next update will no later than 10PM.

This incident has been resolved.

We identified an issue that make "View user streams" from microscope doesn't load. The issue started from 1:40 PM PST. We rolled back the change at 8:35 PM and it should work now for all customers.

We deployed the fix at 11:25PM PT and the regression has been resolved. No further action is required from the customers.

We introduced a regression in our Microscope code path at 3:30PM PT, which caused User Streams launched from Microscope to load indefinitely. We are deploying a fix and will provide an update in the next 30 minutes.

Event streaming have recovered.

A fix has been implemented and we are monitoring the results.

From 4:25AM PT, 2023-11-01, an issue with our event streaming pipeline caused delays reaching ~40 minutes in sending events and users. All streaming customers are impacted. We have identified the issue and are already recovering. We expect no data loss during this window and further actions are not required from our customers.

From 16:30 PDT to 20:30 PDT on Oct 20, 2023, all our data warehouse customers has experienced a delay of up to 4 hours in their data warehouse import and export jobs. The root cause of this incident was a bottleneck in one of our databases. As of now, all the backlogs are processed and our systems are working normally. We do not expect any data loss during this incident and further actions are not required from our customers.

We have remediated the database issue and our systems are back to their normal state. We are processing our backlog of imports and exports and our system will be current in the next 1.5 hours. We will post the next update when the backlogs are all processed.

From 16:30 PDR Oct 20, 2023, all our data warehouse customers are experiencing no data import and export into/out of their source or destinations. This is due to a high load on one of our databases and we working on a remediation. We do not expect any data loss and further actions are not required from our customers. We post the next update in 1 hour.