What vendors changed

GGoogle Workspace**Summary** We are experiencing an issue with Google Keep where users may be unable to access keep.google.com via browser. **Description** The issue began on Friday, 2026-03-20 15:30 US/Pacific. Our engineering team is aware of the situation and is actively implementing mitigations. We do not have any ETA at the moment. We will provide the next update by Friday, 2026-03-20 21:00 US/Pacific. **Customer Symptoms** Impacted users may encounter connection timeouts or error pages (such as 502 Bad Gateway) when attempting to reach keep.google.com via a web browser. The Google Keep mobile applications are currently unaffected and remain operational. **Workaround** Users are able to use Google Keep via the Android or iOS mobile app.

AAmplitudeExperiment Management API Degradation

Amplitude’s Experiment Flags Management API (/api/1/flags) returned incomplete or empty flag listings; UI flags and /api/1/experiments were unaffected. The major incident started 2026-03-20 17:31 UTC, with symptoms reported since 16:00 UTC; it lasted 2 h 28 min and resolved at 2026-03-20 20:00 UTC after a verified fix.

GGorgiasAI Agent and Automation analytics are delayed

Gorgias reported a major incident delaying AI Agent and Automation analytics, affecting AI Agent and Automate - AI Agent. It started at 2026-03-20 15:55 UTC and was resolved at 2026-03-20 19:22 UTC, lasting 3 h 26 min; no cause or fix was stated.

DDevin (Cognition)Issue with sending Devin session responses

Box[Medium] Issues with Logins, Folder and File Preview, and Shared Links

Box reported a major incident affecting Login/SSO, folder and file preview, and shared links; customers could encounter blank pages. It ran from 2026-03-20 03:37 UTC to 2026-03-20 05:08 UTC, lasting 1 h 31 min. Box implemented a fix and confirmed all affected services were restored.

DiscordFailure to start new voice calls

Discord reported a major incident preventing new voice calls from being created, from 2026-03-20 02:47 UTC to 2026-03-20 05:56 UTC, lasting 3 h 9 min. A fix was implemented by 2026-03-20 05:43 UTC, and the incident was resolved.

MailgunElevated 401 errors in US Region

Mailgun reported a major incident affecting API and Email Services - API in the US region, where some requests incorrectly failed with 401 errors. It started at 2026-03-19 18:43 UTC and lasted 10 min, resolving at 2026-03-19 18:53 UTC after a fix was deployed and systems returned to normal.

MongoDB AtlasIssue querying Azure blob storage containers in Atlas Data Federation

MongoDB Atlas reported a major incident affecting Atlas Data Federation reads from customer Azure blob storage containers, starting 2026-03-19 18:11 UTC. A fix was implemented at 19:01 UTC, and the incident was resolved at 2026-03-19 19:19 UTC after 1 h 8 min.

LLaunchDarklyServer-side streaming rejected new connections across all commercial regions

LaunchDarkly reported a major incident affecting new server-side SDK streaming connections across all commercial regions, causing 500/503 errors. Start and resolution were both recorded as 2026-03-19 16:58 UTC, with a computed duration of 0 min. The vendor reported deploying mitigation.

CloudinaryErrors with Account Creation

Cloudinary reported a major Management Console incident affecting account sign-ups from 2026-03-19 16:48 UTC to 2026-03-19 18:38 UTC, lasting 1 h 50 min. A fix was deployed by 03-19 17:13 UTC; Cloudinary confirmed resolution and normal service at the end of the incident.

MongoDB AtlasMongoDB Atlas: Clusters page failing to load

MongoDB Atlas reported a major incident where affected users could see a blank Clusters page; cluster health was unaffected. It started at 2026-03-19 14:42 UTC and was resolved at 2026-03-19 15:28 UTC, lasting 45 min. A fix was implemented at 2026-03-19 15:17 UTC.

WiseService partially unavailable

DatadogIntermittent "No Data" Status for Monitors

Datadog reported a major incident where some metric monitors intermittently showed "No Data," primarily affecting distribution-metric monitors and making alerting potentially unreliable. It started at 2026-03-19 00:54 UTC and lasted 1 h 59 min, resolving at 2026-03-19 02:53 UTC; a fix was implemented.

SStatsigRegional Issue for Statsig Console

Statsig reported a major regional incident limiting Console access for some customers and affecting Console API, starting 2026-03-18 21:25 UTC and resolved at 2026-03-18 22:05 UTC (40 min). The Console was repaired before Console API finished recovering. SDK and API usage remained unaffected; values, exposures, and events continued processing.

CCrusoeKubernetes Cluster Authentication and Provisioning Degradation

HHoneycombHoneycomb UI displaying an error

Honeycomb reported a critical incident affecting ui.honeycomb.io - US1 App Interface, with the UI displaying an error. It started at 2026-03-18 18:07 UTC and was resolved at 2026-03-18 18:18 UTC, lasting 11 min; Honeycomb said it identified the issue and shipped a fix.

Box[Medium] Customers may experience issues previewing files

Box reported a major-impact incident affecting file previews for some users, starting at 2026-03-18 18:00 UTC and marked resolved at the same time (computed duration: 0 min). Box stated that no further impact was observed.

SquarespaceSite Loading Issues

Squarespace reported a major Site Loading outage: errors and delays prevented sites from loading properly from 2026-03-18 15:38 UTC to 2026-03-18 16:56 UTC, lasting 1 h 18 min. A fix was implemented at 03-18 16:28 UTC, and the incident was subsequently resolved.

CockroachDBCloud Console Social Login Auth Unavailable

CockroachDB reported a major Cloud Console outage affecting social login for customers using personal/non-SSO Microsoft accounts. It started at 2026-03-18 15:23 UTC and was resolved at 2026-03-18 16:02 UTC, lasting 39 min; Microsoft social login was restored.

GreenhouseInterview Scheduling Errors

Greenhouse reported a major incident causing interview scheduling errors for some users, starting 2026-03-18 15:19 UTC and resolved 2026-03-18 17:04 UTC; the reported duration was 1 h 44 min. A fix was applied at 03-18 15:44 UTC, followed by monitoring; no cause was stated.

HootsuiteBluesky authorization issues

Hootsuite reported a major incident in its Web Application: users could encounter generic "500" errors when adding or reconnecting Bluesky accounts. It started at 2026-03-18 14:12 UTC and was resolved at 2026-03-18 15:12 UTC, lasting 1 h 0 min; Hootsuite confirmed success rates returned to normal.

SSalesforceTableau Service Disruption: Analytics Cloud

BrexDashboard is down

Brex’s Dashboard was down in a critical-impact incident that started at 2026-03-18 12:24 UTC and was resolved at 2026-03-18 13:40 UTC, lasting 1 h 15 min. Customers could not access the Dashboard; Brex reported resolution but provided no cause or fix details.

JiraAutomations aren't triggering on work item creation, field updates, or comment additions

Jira reported critical delays in event-triggered automations for some Cloud customers starting 2026-03-18 11:29 UTC, lasting 16 h 38 min until resolution at 2026-03-19 04:08 UTC. Scheduled, manual, and incoming-webhook triggers operated normally; historical-event replays remained underway for some high-volume customers at resolution. The postmortem attributed the incident to an internal configuration change.

SStatsigClient SDK Success Rate in Southeast Asia

Statsig reported a major degradation in client SDK initialization success rates affecting Asia Southeast 1 and Config Delivery API - Asia Southeast 1. It started at 2026-03-18 01:50 UTC and was resolved at 2026-03-18 02:53 UTC, lasting 1 h 3 min; the vendor said it identified the root cause but did not disclose it.

BBrazeUS05 Dashboard latency

Braze reported a major US05 Dashboard outage affecting users’ ability to reach the dashboard, starting 2026-03-17 20:49 UTC and resolved at 2026-03-17 21:04 UTC (15 min). Engineers implemented a fix; messaging sending and other services were not impacted.

GitLabDelay in background job processing

WWorkOSElevated API Errors

WorkOS reported a major incident involving elevated API errors and latency affecting AuthKit, Core Services - AuthKit, Core Services - SSO, and SSO. It started at 2026-03-17 14:12 UTC and was resolved at 2026-03-17 14:30 UTC, lasting 18 min; services were reported back to normal at 14:27 UTC.

CCerebrasZ.ai GLM 4.7 Service Unavailable

CCerebrasZ.ai GLM 4.7 Service Unavailable

GitLab500 errors across several REST and GraphQL API endpoints

SSalesforceService Disruption: Core Service

KeeperResolved: KeeperPAM Connection Errors in US Data Center

KeeperPAM customers with connections and tunnels enabled experienced sporadic errors after login in the US Data Center; vault login was unaffected. Keeper rated the incident major, starting 2026-03-16 18:25 UTC and resolving 2026-03-16 18:52 UTC (26 min). The stated cause was a communication issue with the KeeperPAM router endpoint.

WarpIntermittent DNS Issues

DDrataGuest Admin Access

Drata reported a major Guest Admin Access outage affecting Guest Admins’ ability to view clients within "Business Accounts". It started at 2026-03-16 16:55 UTC and was resolved at 2026-03-16 19:16 UTC, lasting 2 h 20 min. Drata released an update and stated that Guest Admins should now be able to view clients as expected.

GitLabDelay in background job processing

CCloseDatabase issues

Close reported a critical database incident affecting the API and Application UI, starting 2026-03-16 16:08 UTC and resolved at 16:55 UTC (47 min). Its postmortem states all users experienced severe disruption from 15:57 to 16:15 UTC, with automated systems restoring functionality by 16:15 UTC; normal performance was confirmed at resolution.

SSalesforceService Disruption: Core Service

SSalesforceField Service Scheduling Feature Disruption: Field Service Scheduling

CodecovUnable to login with Bitbucket

Bitbucket login on Codecov Frontend suffered a major outage from 2026-03-16 11:46 UTC to 2026-03-17 16:07 UTC, lasting 28 h 20 min. Codecov identified a Bitbucket OAuth connection issue and implemented a fix. All Bitbucket users must log in again because authorizations were reset; Codecov cannot communicate with Bitbucket until they do.

SocketGitHub App reports not starting

Socket’s GitHub App reports were not starting, affecting Integrations - Socket for GitHub and Socket for GitHub. The vendor rated the impact critical; the incident started at 2026-03-14 06:16 UTC and was resolved at 2026-03-14 06:54 UTC, lasting 37 min. Preliminary investigation showed high error rates from the upstream GitHub API.

SSalesforceService Disruption: Core Service

SocketDashboard access unavailable due to DNS issue

Socket reported a major outage that made dashboard access unavailable starting 2026-03-13 22:27 UTC. The outage lasted 7 min and was resolved at 2026-03-13 22:34 UTC; Socket attributed it to a DNS migration.

Webflow[RESOLVED] Reports of issues loading the Webflow Canvas and Dashboard

Webflow reported a major outage affecting Webflow Canvas and Dashboard loading, recorded as starting at 2026-03-13 19:59 UTC and resolved at 20:49 UTC (50 min). Its final update stated customers were unable to load both services between 19:28 and 20:20 UTC; a fix was implemented and normal service restored.

SSemgrepElevated errors on managed diff scans

Semgrep reported a major incident affecting Managed Scans: some jobs failed, primarily diff scans used for PR/MR checks. The incident ran from 2026-03-13 18:10 UTC to 2026-03-13 19:48 UTC (1 h 37 min), though errors began rising at 2026-03-13 04:30 UTC. Semgrep increased resources for affected projects; customers could push a new commit to trigger a new scan.

MailgunIntermittent API Errors in US Region

Mailgun reported a major degradation involving intermittent API errors in the US Region, affecting API and Email Services - API. It started at 2026-03-13 11:29 UTC and lasted 1 h 20 min, resolving at 2026-03-13 12:49 UTC after a fix was implemented; no cause was stated.

DatadogWeb Application Not Loading

Datadog reported a critical Web Application outage causing loading errors for some users, starting 2026-03-13 09:26 UTC and resolved at 2026-03-13 10:30 UTC (1 h 3 min). A fix was implemented by 03-13 10:05 UTC; no cause was stated.

RenderDegraded builds and deployments in the Singapore region

Render reported a major degradation of builds and deployments in Singapore from 2026-03-13 08:52 UTC to 2026-03-13 11:09 UTC, lasting 2 h 17 min, with longer completion times. Services using free instance types were disabled at 09:27 UTC and re-enabled at 10:18 UTC; Render reported the incident mitigated.

PPinecone[GCP][europe-west4] Errors for upsert operations on some indexes

Pinecone reported a major incident causing upsert errors on some indexes in GCP europe-west4, affecting Serverless Indexes. It started at 2026-03-13 08:50 UTC and was resolved at 2026-03-13 09:53 UTC, lasting 1 h 3 min; a fix was implemented and monitored before resolution.