What vendors changed

GitLabUsers reporting 403 Forbidden responses when downloading NuGet packages

MongoDB AtlasAtlas Clusters Snapshot Failures

MongoDB Atlas cluster snapshots were failing in a major-impact incident affecting MongoDB Cloud, from 2026-03-10 23:18 UTC to 2026-03-11 02:30 UTC (3 h 11 min). MongoDB implemented a fix and confirmed resolution; customers faced snapshot failures during the incident.

Red Hat (incl. Quay.io)Intermittent issues with RHEL 9 and RHUI downloads

FFirebaseWe are currently experiencing high error rates on the Firebase console. Investigations are ongoing.

TTeleportNew Tenant Onboarding Failures

SSalesforceService Disruption: Core Service

LLlamaIndexElevated Latency

ConfluenceAutomation events delayed for some customers in the APAC region

Confluence reported a major incident delaying automation events for some APAC customers, starting 2026-03-10 00:15 UTC and resolved at 2026-03-10 00:59 UTC (43 min). A hotfix restored automation performance while the backlog was processed; the vendor confirmed normal service for affected customers.

JiraAutomation events delayed for some customers in the APAC region

Some APAC customers experienced delayed automations in Jira, Jira Service Management, Jira Work Management, Jira Product Discovery and Confluence. Jira rated the incident major; it started at 2026-03-10 00:06 UTC and was resolved at 2026-03-10 00:59 UTC, lasting 52 min. A hotfix restored automation performance while the backlog was processed.

RenderCustom domains are unable to be created

Render’s Custom Domains experienced a major outage that prevented custom domain creation from 2026-03-09 20:19 UTC to 2026-03-09 20:39 UTC, lasting 20 min. Render marked the incident resolved but did not state a cause or fix.

DiscordIssues sending and loading messages

Discord reported a major API outage affecting sending and receiving messages, starting 2026-03-09 17:08 UTC and lasting 50 min until resolution at 2026-03-09 17:58 UTC. A fix was implemented at 17:29 UTC; no cause was stated.

GrammarlyIncreased latency across Grammarly product offerings

Grammarly reported a critical degradation causing intermittent or slow features, login trouble, and “Service unavailable” errors from 2026-03-09 16:52 UTC to 2026-03-09 20:07 UTC, lasting 3 h 14 min. A fix implemented at 19:15 UTC restored login and use of Grammarly and Superhuman; the incident was resolved.

GitLabCI/CD pipeline processing delays

BrexOkta sign in widget loading slowly

Brex’s Okta sign in widget loaded slowly, affecting Dashboard access, with vendor-rated critical impact. The incident started at 2026-03-08 14:48 UTC and was resolved at 2026-03-08 15:42 UTC, lasting 53 min; Brex reported identifying the issue but provided no cause or fix details.

SingleStoreGCP us-west1 region may be impacted for some customers

SnowflakeAWS - Europe (Zürich): INC0153183

Snowflake reported a major outage in AWS - Europe (Zürich), potentially preventing sign-ins, queries and data management. It started at 2026-03-07 20:05 UTC and was resolved at 2026-03-07 22:20 UTC; the recorded duration was 2 h 14 min. Snowflake identified a third-party cloud platform issue and confirmed service restoration after coordinating a fix with the provider.

SSalesforceCommerce Cloud Regional Service Disruption Managed Runtime: MANAGEDRUNTIMEADMINAPI

SSalesforceService Disruption: MANAGEDRUNTIMEADMINUI

SSalesforceB2 B2 CCommerce Service Disruption: B2 CCore, Cart Service

LLlamaIndexElevated Latency

MMuxvideo clips audio loss

Mux Video - VOD had major audio loss affecting some clips using "basic" or "premium" video_quality, caused by an encoding-change metadata mismatch. It started 2026-03-06 22:57 UTC and was resolved 2026-03-07 01:57 UTC (3 h 0 min); new clips work, but affected clips require re-creation from original assets to restore audio.

Red Hat (incl. Quay.io)Intermittent issues with RHEL 9 and RHUI downloads

GitLabCI/CD pipelines remain stuck in running status

SocketGitHub app reports are stuck not starting

Socket reported a major outage affecting some GitHub app reports not starting, from 2026-03-06 16:15 UTC to 2026-03-07 07:04 UTC, lasting 14 h 48 min. Socket suspected an upstream GitHub webhook problem and advised closing and reopening PRs with stuck checks to trigger another webhook; the incident was resolved.

UpstashRegion ap-northeast-1 outage on Upstash Global

Upstash reported a critical outage affecting Redis Global in Tokyo, Japan (ap-northeast-1), causing elevated latency and connection errors for some databases. The incident ran from 2026-03-06 13:50 UTC to 2026-03-06 14:49 UTC (59 min). A traffic spike increased network utilization and connection load; Upstash mitigated it by scaling regional capacity and redistributing load across additional nodes.

SSemgrepInvestigating issues with pull request checks and repository scans

Semgrep reported a major outage affecting repository scans and GitHub pull request checks: some customers saw scans not starting or checks not appearing. It started at 2026-03-06 12:04 UTC and was resolved at 2026-03-06 16:18 UTC, lasting 4 h 14 min. A fix was rolled out, and scans returned to normal operation.

GitLabGitLab Main Repo Pull Errors

GitLabIntermittent HTTP 50X errors when attempting to access to GitLab.com

BitbucketDisrupted Bitbucket availability

Bitbucket reported a critical outage affecting Website, API, Git via HTTPS, Git via SSH, and Pipelines, leaving affected users unable to access services. It started at 2026-03-06 02:44 UTC and was resolved at 2026-03-06 05:11 UTC, lasting 2 h 27 min. The postmortem attributed the disruption to a hosting-provider regional provisioning API rate limit blocking application workers from handling website traffic.

GitLabCI/CD pipelines intermittently stuck on running status

CCodeiumand WindsurfWe are investigating reports of completions/Tab not working

Codeium reported a major outage affecting Windsurf Tab completions starting 2026-03-05 16:23 UTC, lasting 1 h 22 min. A fix was implemented by 03-05 16:51 UTC, and the incident was resolved at 2026-03-05 17:46 UTC.

SSigma ComputingIssue with Burst email exports are resolved

Sigma Computing reported a major outage of Burst email exports starting 2026-03-05 10:10 UTC, lasting 4 days 7 h until resolution at 2026-03-09 17:31 UTC; standard exports remained operational. Engineers deployed a fix and re-enabled Burst Email Exports at 03-07 01:27 UTC, with no recurring problems observed during subsequent monitoring.

CConfluentMetrics API was experiencing failures from 03:23 UTC to 03:58 UTC

Confluent reported a major Confluent Cloud Metrics API outage, with unavailability from 2026-03-05 03:23 UTC to 2026-03-05 03:58 UTC. The incident record ran from 2026-03-05 04:40 UTC to resolution at 2026-03-05 07:22 UTC (2 h 42 min). Confluent reported mitigation and confirmed normal operation at resolution.

DiscordGuild Unavailability

Some Discord guilds were unavailable in a major-impact incident that started at 2026-03-05 01:37 UTC and was resolved at 2026-03-05 01:58 UTC, lasting 20 min. Discord identified the issue and reported implementing a fix, but did not state a cause.

PPinecone[Serverless][GCP][us-central1] 5xx errors on some read operations for some indexes

Pinecone reported a major incident causing 5xx errors on some read operations for some Serverless indexes in GCP us-central1. It started at 2026-03-04 21:06 UTC and was resolved at 2026-03-04 22:16 UTC, lasting 1 h 10 min. A fix was implemented; affected customers could encounter failed reads during the incident.

GitLabDuo Agentic Chat intermittently fails to respond

CCodeiumand WindsurfWe are investigating reports of completions/Tab not working

Codeium reported a major incident affecting Windsurf Tab completions, starting at 2026-03-04 20:16 UTC and resolved at 2026-03-04 20:54 UTC after 38 min. A fix was implemented at 20:17 UTC; no cause was stated.

AAshbyDelays in generating meeting transcripts

Ashby's AI Notetaker meeting transcript generation was delayed in a major-impact incident starting 2026-03-04 19:59 UTC and resolved at 2026-03-04 20:44 UTC, lasting 44 min. A fix was implemented at 20:31 UTC; no cause was stated.

SSalesforceMule Soft Service Disruption: Exchange

BrexDashboard and mobile app are not loading

Brex’s dashboard and mobile app were not loading during a critical outage that started at 2026-03-04 19:38 UTC and was resolved at 2026-03-04 19:48 UTC, with a reported duration of 9 min. Brex said it identified the issue and applied a fix; no cause was stated.

UUnbounceUnable to Connect Custom Domains

Unbounce reported a major outage preventing custom domain registration, attributed to a Cloudflare outage. It started at 2026-03-04 18:33 UTC and was resolved at 2026-03-05 00:32 UTC, lasting 5 h 59 min; Unbounce said a fix restored registration functionality.

Convexhigh error rates on customer backends

Convex reported a critical incident causing high backend error rates and downtime for some customers, affecting Free & Starter and Live Traffic database, function and Node.js runtimes. It started 2026-03-04 17:28 UTC and lasted 1 h 0 min. Convex remediated the issue and attributed it to a load spike triggering load-shedding and a caching-library panic.

DDevin (Cognition)Sessions slow to start

Windsurfand CCodeiumNetworking issues effecting tab

Windsurf Tab experienced networking issues with major impact starting 2026-03-04 13:49 UTC. The incident lasted 3 h 32 min and was resolved at 2026-03-04 17:22 UTC; no cause or specific fix was stated.

SSalesforceMarketing Cloud Disruption: Advertising Audiences, Analytics Builder, Audience Builder +13 more

GitLabIntermittent HTTP 50X errors when attempting to access to GitLab.com

BrexDashboard Partial Outage

Brex reported a critical Dashboard partial outage starting 2026-03-03 22:14 UTC, lasting 2 h 20 min and resolved at 2026-03-04 00:35 UTC. A fix was in place by 03-03 23:19 UTC; no cause was stated.