LangSmith

1. investigating 2026-10-14 19:47 UTC

   Following a caching change sandbox performance is experiencing some degradation. We are investigating.

2. monitoring 2026-10-14 20:08 UTC

   We've identified the issue as a bad host and resolved. Monitoring for any abnormalities.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Orchestration, control plane, service discovery or load balancing failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. investigating 2026-10-12 06:08 UTC

   We are seeing errors fetching runs in APAC.

2. resolved 2026-10-12 06:24 UTC

   An image pull failure temporarily took down our querying services. We swapped to a pullable image resolving the issue.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Login, tokens, certificates or credentials failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

1. investigating 2026-10-12 05:42 UTC

   Some users can't log in to or sign up for LangSmith right now. Login attempts may time out or fail. Users who are already logged in should not be affected. We are actively investigating and will post another update as we learn more.

2. monitoring 2026-10-12 06:10 UTC

   We were having an issue with our captcha provider. For now we have disabled captcha which has resolved the issue

3. resolved 2026-10-14 16:02 UTC

   Underlying captcha issue has been resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A configuration change caused the failure. Root cause: A configuration change. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

1. investigating 2026-10-08 07:08 UTC

   The LLM Gateway ([gateway.smith.langchain.com](http://gateway.smith.langchain.com)) is returning an elevated rate of 502 errors. We are investigating.

2. resolved 2026-10-08 07:43 UTC

   From 07:08 to 07:43 UTC on October 8, 2026, the LLM Gateway ([gateway.smith.langchain.com](http://gateway.smith.langchain.com)) returned high 502 error rates. It was due to an incorrect infrastructure configuration change that stopped traffic from reaching the gateway. We fully...

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Exhausted memory, disk, CPU, queues, connection pools or quotas. Root cause: Not enough capacity for the load. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. investigating 2026-10-08 00:48 UTC

   We are investigating a latency increase in our V1 Run Query APIs. V2 Run Query APIs backed by SmithDB are unaffected.

2. monitoring 2026-10-08 01:39 UTC

   Query API performance is back to normal after capacity adjustment. We are currently monitoring for further problems.

3. resolved 2026-10-08 02:30 UTC

   Capacity remained stable in the last hour. Incident resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

Names an upstream provider: Google Cloud.

1. investigating 2026-10-03 05:10 UTC

   Creation of new LangSmith deployment dedicated instances are failing during database provisioning because of an issue with an upstream cloud provider. Only GCP regions are affected.

2. investigating 2026-10-03 05:55 UTC

   Creating new deployments and new revisions for dedicated instances is failing for LangSmith Deployment in all GCP regions due to an issue with an upstream cloud provider. Existing deployments continue to run and serve traffic.

3. identified 2026-10-03 06:46 UTC

   Issue has been identified, attempted mitigation is being deployed.

4. resolved 2026-10-03 10:13 UTC

   This is now resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).

1. investigating 2026-09-04 17:19 UTC

   This is impacting the creation of new agents with a computer per thread configuration

2. identified 2026-09-04 18:12 UTC

   We have identified the cause and are working on resolving the issue.

3. monitoring 2026-09-04 19:31 UTC

   The fix has been deployed. All systems are operational

4. resolved 2026-09-04 19:38 UTC

   All systems are operational

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A configuration change caused the failure. Root cause: A bug in the vendor's own code. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. investigating 2026-09-02 17:04 UTC

   This affects only newly created agents with a shared computer.

2. identified 2026-09-02 17:59 UTC

   We have identified the root cause of this incident and are actively working on a fix. The issue originates from a recent change to how sandbox slugs are generated in our backend systems, which now derive slugs directly rather than referencing sandbox IDs. This has introduced a...

3. monitoring 2026-09-02 20:32 UTC

   The fix has been deployed. All systems are operational

4. resolved 2026-09-02 20:47 UTC

   All systems are operational

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).

1. investigating 2026-09-01 19:43 UTC

   We are investigating an elevated error rate for bulk exports executions. Bulk exports are currently delayed. New bulk exports can still be created.

2. monitoring 2026-09-01 19:53 UTC

   We have mitigated the issue and have seen recovery. We're continuing to monitor.

3. resolved 2026-09-01 20:17 UTC

   This incident has been resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Exhausted memory, disk, CPU, queues, connection pools or quotas. Root cause: Not enough capacity for the load. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. investigating 2026-08-26 16:16 UTC

   We are currently investigating an issue where some trace loads are timing out

2. investigating 2026-08-26 16:26 UTC

   Currently dealing with some heavy queries causing issues fetching traces. We are scaling out our system which should remediate.

3. investigating 2026-08-26 16:27 UTC

   This is primarily affecting run querying

4. monitoring 2026-08-26 16:42 UTC

   We've identified the query causing problems to our systems and have pushed a mitigation. All systems look recovered but we are monitoring for any other problems.

5. resolved 2026-08-26 16:53 UTC

   A pathological query was impacting other query services. We've added protections against the pathological case and scaled up our query services. All services are now functioning as normal

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

Names an upstream provider: Google Cloud.

Started during Google Cloud: We are investigating an issue where customers may experience timeouts, service degradations, errors, and elevated latencies across multiple products in the us-west1 region..

1. investigating 2026-08-20 15:54 UTC

   Builds for new revisions of LangSmith Deployments are not being processed.

2. monitoring 2026-08-20 16:30 UTC

   Builds are being processed. Validating recovery.

3. identified 2026-08-20 17:05 UTC

   GCP is having an incident in us-west-1. Deployments may experience issues deploying new revisions.

4. monitoring 2026-08-20 17:54 UTC

   New revisions are being processed normally. Validating recovery.

5. resolved 2026-08-20 18:32 UTC

   New revisions are deploying correctly.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Data delayed, stale, lost or corrupted, or needing replay after recovery. Root cause: A bug in the vendor's own code. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. identified 2026-08-06 15:03 UTC

   Traces including large payloads might currently fail to be ingested or can be delayed. We are working on a fix.

2. identified 2026-08-06 15:33 UTC

   Updated affected regions

3. monitoring 2026-08-06 16:43 UTC

   The service is temporarily patched. Large payloads are now ingested normally.

4. resolved 2026-08-07 07:40 UTC

   The full fix took an extraordinarily long time to release due to an outage at our CI provider. In the meantime, the mitigation we had deployed mitigated the problem.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A code release introduced the failure. Root cause: A bug in the vendor's own code. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

Names an upstream provider: Anthropic (Claude).

1. investigating 2026-08-05 15:27 UTC

   We're investigating an issue where agents and workflows using Anthropic Claude models with custom tools may fail to complete runs. Affected requests return a tool input validation error. We'll share more information as we have it.

2. identified 2026-08-05 18:36 UTC

   We're seeing elevated error rates when using Fleet with Claude models and certain MCP tools.

3. monitoring 2026-08-05 20:21 UTC

   We're seeing elevated error rates when using Fleet with Claude models and certain MCP tools.

4. resolved 2026-08-05 21:52 UTC

   Resolved. Fix is now deployed that skips an unsupported tool schema and logs a warning instead of failing the request.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. investigating 2026-08-03 18:15 UTC

   We are aware of an issue causing high latency and intermittent HTTP 500 errors for analytical queries (Stats and Dashboards) in LangSmith. Our team is actively investigating and working on a resolution.

2. investigating 2026-08-03 18:20 UTC

   We are aware of an issue causing high latency and intermittent HTTP 500 errors for analytical queries (Stats and Dashboards) in LangSmith. Our team is actively investigating and working on a resolution.

3. monitoring 2026-08-03 19:49 UTC

   We have identified the root cause, mitigated the issue, and are actively monitoring the service to ensure continued stability.

4. resolved 2026-08-03 21:13 UTC

   The issue has been resolved, and service has returned to normal. We identified the root cause, implemented a mitigation, and have observed stable system performance following the change. We will continue to monitor the service closely to ensure ongoing stability. Thank you for...

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

Names an upstream provider: Google Cloud.

1. investigating 2025-08-06 02:57 UTC

   We are noticing increased latency for multiple endpoints serving trace information and metrics with LangSmith. Run ingestion is not impacted currently.

2. investigating 2025-08-06 03:20 UTC

   We are investigating a potential issue with our cloud provider related to GCS.

3. identified 2025-08-06 03:35 UTC

   We have identified an issue with our cloud service provider and are monitoring remediation. Will post further updates as we get them.

4. monitoring 2025-08-06 04:32 UTC

   LangSmith API and Application endpoint latency issues were affected by an issue with Google Cloud Platform infrastructure in the us-central1 region. The GCP issue appears to be resolved and response times are back to normal. We will continue to monitoring our services to ensure...

5. resolved 2025-08-06 05:27 UTC

   All LangSmith services are operational after the us-central1 region service impact was addressed by Google Cloud Platform.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).

1. identified 2025-03-25 15:30 UTC

   We've identified an infrastructure issue causing run ingest delays and slow queries. Some queries are failing and retryable. We are actively working on a mitigation.

2. identified 2025-03-25 15:30 UTC

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

3. monitoring 2025-03-25 15:42 UTC

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

4. resolved 2025-03-25 16:04 UTC

   Performance has returned to historic norms.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Traffic surge or denial-of-service attack. Root cause: Not enough capacity for the load. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

1. investigating 2025-01-24 20:58 UTC

   We are actively investigating an issue impacting run ingestion, frontend and API access in the LangSmith US instance.

2. monitoring 2025-01-24 23:06 UTC

   An activity spike caused an unexpected period of increased latency and API query timeouts. We have added resources to compensate and are working with our providers to identify the root cause.

3. resolved 2025-01-25 15:25 UTC

   This incident has been resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).

1. investigating 2026-05-22 16:35 UTC

   All sandbox create calls error

2. identified 2026-05-22 17:15 UTC

   Fix is rolling out

3. monitoring 2026-05-22 17:16 UTC

   Creates succeeding again

4. resolved 2026-05-22 17:29 UTC

   Resolved

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).

1. investigating 2025-09-25 18:22 UTC

   We have detected an elevated rate of 5xx on run ingestion and a subset of API endpoints and are actively investigating.

2. identified 2025-09-25 18:35 UTC

   We have identified the issue and have pushed a fix reducing the error rates. We will continue to address the issue

3. monitoring 2025-09-25 18:41 UTC

   We are recovering wrt 5XX rates and are monitoring.

4. resolved 2025-09-26 16:54 UTC

   All error rates have been at baseline level for almost 24 hours.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A code release introduced the failure. Root cause: A bug in the vendor's own code. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

1. investigating 2025-06-18 08:48 UTC

   New revisions are failing for LangGraph Platform deployments. Engineers have identified the issue and pushed out a fix.

2. monitoring 2025-06-18 08:53 UTC

   Engineers have pushed out a fix for the issue and are monitoring our systems to ensure everything is healthy.

3. monitoring 2025-06-18 22:16 UTC

   Control Plane functionalities resolved as of 5:12 AM EST. New revisions were down from 2:22 AM EST to 4:56 AM EST due to a bad push for our base images. Existing deployments were unaffected.

4. resolved 2025-06-20 19:34 UTC

   This incident has been resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Data delayed, stale, lost or corrupted, or needing replay after recovery. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. investigating 2025-01-20 09:14 UTC

   We are experiencing elevated latency in ingesting runs in the US instance of LangSmith. Ingest is fully operational at this time but with a delay >30s between run submission and the run appearing in the LangSmith UI.

2. identified 2025-01-20 11:20 UTC

   We have identified a performance regression in our production database and are working with our vendor to resolve. Run ingest may experiment intermittent periods of increased latency of up to 60s while we work to mitigate the issue.

3. monitoring 2025-01-22 00:00 UTC

   A fix for the intermittent delays in run ingest has been implemented. Ingest latencies have returned to historic norms and we are monitoring to ensure continued performance.

4. resolved 2025-01-23 14:28 UTC

   This incident has been resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A code release introduced the failure. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. resolved 2024-07-03 17:25 UTC

   Between 1652 and 1711 UTC, an issue with a production deployment caused a backlog in our run ingest queues, delaying the display of traces in the LangSmith UI. No runs were blocked from ingest during this time. The backlogged runs have been processed and as of 1715 UTC median...

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: Data delayed, stale, lost or corrupted, or needing replay after recovery. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).

1. identified 2026-06-05 18:38 UTC

   We have identified an issue causing delays in billable usage tracking for LangSmith Deployments.

2. monitoring 2026-06-06 00:23 UTC

   We have a fix in place and are processing the backlog of usage.

3. resolved 2026-06-06 02:08 UTC

   The backlog has completed processing. This incident is resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).

1. investigating 2025-11-14 16:30 UTC

   Our users are unable to access LangSmith Application and Run Ingestion is impacted. We are investing the root cause to remediate.

2. monitoring 2025-11-14 16:50 UTC

   Our services are back online. We are monitoring to ensure normal operations.

3. resolved 2025-11-14 23:46 UTC

   One of our underlying cloud services was impacted triggering the outage. Further details will be shared offline. Appreciate all the patience

Excerpts are shortened. Read the full incident on LangSmith's status page.

Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).

1. investigating 2025-01-16 17:16 UTC

   We are investigating increased read latency in the LangSmith EU instance. Run ingest is not affected at this time.

2. monitoring 2025-01-16 22:53 UTC

   We identified an issue with our production database and have worked with our provider to mitigate the issue. Query performance has returned to historic norms and we are continuing to monitor to ensure continued stability.

3. resolved 2025-01-20 09:10 UTC

   This incident has been resolved.

Excerpts are shortened. Read the full incident on LangSmith's status page.