Notable SaaS vendor changes, page 7

Rippling reported a major outage in which the Rippling App loaded intermittently or not at all for employees and customers. It started at 2026-09-18 16:18 UTC and was resolved at 2026-09-18 23:00 UTC, lasting 6 h 41 min. A fix was implemented at 17:21 UTC; no cause was stated.

Make reported a major degradation on us1.make.celonis.com affecting scenario executions, webhook processing, and platform access. It started at 2026-09-18 13:36 UTC and was resolved at 2026-09-18 17:32 UTC, lasting 3 h 56 min; corrective actions stabilized the platform, but no specific cause was disclosed.

Cursor reported a major Grok Bot service degradation starting 2026-09-18 11:37 UTC, with degraded performance or bots not responding. The incident lasted 2 h 56 min and was resolved at 2026-09-18 14:33 UTC; no cause or fix was stated.

Sentry reported major delays in EU spans ingestion starting 2026-09-18 10:17 UTC, listing EU Attachment Ingestion, EU Error Ingestion and EU Span Ingestion as affected components. The incident lasted 41 min and was marked resolved at 2026-09-18 10:58 UTC; the final update still described an upstream provider networking issue with mitigations underway.

Sentry reported critical EU ingestion delays affecting new errors, with EU alerting, Dashboard, and attachment, error and span ingestion listed as affected. The incident ran from 2026-09-18 08:36 UTC to 2026-09-18 11:57 UTC (3 h 20 min). Sentry identified an upstream-provider networking issue, implemented mitigation, and restored real-time processing for all data types.

OneSignal reported major intermittent errors creating notifications through /api/v1/notifications, affecting some customers from 2026-09-18 00:48 UTC until resolution at 2026-09-18 01:23 UTC (34 min). Heavy load on a notification service caused the errors; OneSignal scaled up resources and confirmed the endpoint was stable for all customers.

Rippling MCP was down in a critical-impact incident affecting Identity & App Management - Third-party integrations and Third-party integrations. It started at 2026-09-17 20:13 UTC and was resolved at 2026-09-17 20:47 UTC, lasting 34 min; Rippling reported implementing a fix but did not specify the cause.

Socket Website loaded intermittently for some users in an incident rated critical, starting 2026-09-17 19:52 UTC and resolved at 2026-09-17 22:01 UTC; the reported duration was 2 h 8 min. Socket implemented a fix at 20:35 UTC but did not state a cause.

Connection attempts failed for multiple CockroachDB Cloud (Basic) tenants in GCP South America (São Paulo / southamerica-east1), preventing affected customers from connecting to clusters. CockroachDB rated the incident major; it started at 2026-09-17 19:08 UTC and was resolved at 2026-09-17 20:39 UTC, lasting 1 h 30 min. Connectivity was fully restored, with all affected clusters operating as expected.

ChatGPT Work mode experienced a major incident causing failures for some users starting or running Work tasks. It began at 2026-09-17 19:03 UTC and lasted 1 h 21 min, resolving at 2026-09-17 20:24 UTC; OpenAI reported implementing mitigation and confirmed full recovery.

ChatGPT Work mode experienced elevated errors, with some users unable to start or run Work tasks. The vendor rated the impact major; the incident began at 2026-09-17 19:03 UTC and lasted 1 h 21 min, with full recovery reported at 2026-09-17 20:24 UTC after mitigation work.

Discord reported a major incident causing issues or delays in establishing call connections from 2026-09-17 18:18 UTC to 2026-09-17 20:39 UTC, lasting 2 h 21 min. A fix was implemented by 20:17 UTC, and the incident was subsequently resolved.

Astronomer reported a major outage from 2026-09-17 17:54 UTC to 2026-09-17 21:50 UTC, lasting 3 h 56 min. A regression from a change rolled out that day silently dropped custom-role permissions, causing reported deploy failures for tokens using those roles; re-adding deployment.deploys.create mitigated failures. All custom roles were restored, and the incident was resolved.

Elastic Cloud reported a major AutoOps degradation: node-level CPU, memory, disk and thread pool metrics were missing in some regions from 2026-09-17 15:50 UTC to 2026-09-17 16:31 UTC (41 min). A validated fix was rolled out; cluster health, shard and deployment data were unaffected, with no data loss.

Grok Bot experienced a major service degradation starting 2026-09-17 12:30 UTC, with degraded performance or bots not responding. The incident was resolved at 2026-09-17 14:34 UTC after 2 h 3 min; no cause or fix was stated.

Some Maia runners failed to start in the EU and US regions, affecting Agents; Matillion rated the impact major. The incident started at 2026-09-17 10:09 UTC and was resolved at 2026-09-17 10:20 UTC, with a reported duration of 10 min. Matillion identified the issue and reviewed a fix but did not state a cause.

Monday.com reported a major outage affecting automations, integrations, and workflows in EU, which could fail, stop running, or not load. It started at 2026-09-17 10:07 UTC and was resolved at 2026-09-17 10:39 UTC, lasting 32 min. A fix was implemented; customers were asked to refresh their browser to resume regular service.

Pinecone reported a major incident causing readpath 5xx errors for Serverless Indexes in AWS us-west-2, from 2026-09-17 05:37 UTC to 2026-09-17 16:44 UTC (11 h 7 min). Pinecone reported mitigation underway, followed by recovery monitoring and resolution; no specific cause or fix was stated.

RunPod reported a critical network connectivity outage in EU-NL-1 starting at 2026-09-17 03:41 UTC, affecting connectivity in that data center. It lasted 1 h 59 min and was resolved at 2026-09-17 05:40 UTC, when RunPod reported the network was operating normally.

Discord reported a major outage affecting voice call establishment, starting at 2026-09-16 21:32 UTC and resolved at 2026-09-16 22:35 UTC, with a reported duration of 1 h 2 min. A fix was implemented by 2026-09-16 22:15 UTC; no cause was stated.

AssemblyAI recorded a critical US Async API outage starting 2026-09-16 20:22 UTC, lasting 1 h 14 min. At peak, roughly 50% of US async transcription requests failed; some successful requests were delayed, while EU and real-time/streaming services were unaffected. The vendor attributed the failure to AWS SQS and reported resolution after implementing a fix.

Modal reported a critical outage affecting its customer dashboard and sandboxes from 2026-09-16 18:29 UTC to 2026-09-16 18:43 UTC, lasting 14 min. Modal said it identified and remediated the issue at 18:39 UTC, then observed recovery and marked the incident resolved at 18:43 UTC; no specific cause was disclosed.

Discord reported a major outage preventing Android app purchases, including Nitro subscriptions and Shop items, caused by an ongoing Google Play outage. It started at 2026-09-16 16:15 UTC and was resolved at 2026-09-16 16:43 UTC, lasting 27 min. Purchases on iOS, web, and desktop were unaffected, and existing subscriptions remained active.

Cursor reported a major Grok Bot degradation: users could experience degraded performance or bots that did not respond. It started at 2026-09-16 14:27 UTC and was resolved at 2026-09-16 20:57 UTC, lasting 6 h 29 min. Cursor said it identified the root cause and was implementing a fix, but provided no specifics.

Some customers could not access the Elastic Support Portal during a major-impact incident starting 2026-09-16 13:17 UTC, resolved at 2026-09-16 15:38 UTC after 2 h 21 min. Elastic investigated with its provider and directed customers to support@elastic.co; access recovered and remained stable.

Gorgias reported a major incident affecting customer sidebar search in Helpdesk - Web App and Web App. It started at 2026-09-16 12:51 UTC and was resolved at 2026-09-16 14:05 UTC, lasting 1 h 14 min; no cause or specific fix was stated.

Salesloft reported a major Salesforce sync degradation starting 2026-09-16 11:59 UTC, lasting 2 h 0 min and resolved at 2026-09-16 14:00 UTC. An underlying Salesforce outage caused activity, task and opportunity sync delays or failures for some customers across Sales Engagement and Forecast/Inspect/Analyze; support responses could also be delayed. Salesforce confirmed recovery, and Salesloft reported normal functionality restored.

Census reported a major incident affecting some customers: LinkedIn syncs failed with status 426, and new and existing LinkedIn Destination tests could fail. It started at 2026-09-16 11:42 UTC and was resolved at 2026-09-16 15:33 UTC, lasting 3 h 50 min. Impacted syncs should recover on their next scheduled or manual run; destination connections and tests should work again.

Smartsheet reported a major degradation affecting the Salesforce Connector and support case creation in the Customer Support Portal from 2026-09-16 10:36 UTC to 2026-09-16 21:02 UTC, lasting 10 h 26 min. The vendor attributed intermittent errors to an upstream CRM provider outage affecting instance stability and API availability across multiple regions; both services were fully restored.

Braze reported a critical Support Services and Partner Portal outage from 2026-09-16 09:17 UTC to 2026-09-16 16:09 UTC, lasting 6 h 51 min. A Salesforce Service Cloud incident disrupted case submissions through the Braze Support Portal and Braze AI Operator and could delay responses; service was restored following remediation of the underlying issue.

BigCommerce’s critical contact center outage started at 2026-09-16 09:15 UTC and lasted 6 h 27 min, resolving at 2026-09-16 15:42 UTC. Phone, Chat, and Web cases were initially unavailable, followed by delays or intermittent errors. BigCommerce attributed the disruption to its service provider, which rolled out a fix region by region.

MongoDB Atlas reported a major Support Portal outage that degraded or limited customers’ access to technical support. The incident started at 2026-09-16 09:06 UTC and was resolved at 2026-09-16 15:31 UTC, lasting 6 h 24 min; MongoDB stated that the platform issues were resolved.

A major Salesforce disruption across all regions prevented some Xero customers from raising cases, contacting support, using in-app help or accessing Xero Central; Xero platform access was unaffected. The incident ran from 2026-09-16 08:49 UTC to 2026-09-16 15:12 UTC, lasting 6 h 23 min, and was resolved after Salesforce implemented a fix.

Lovable reported a major outage in which some users’ previews and builds failed to start; retrying usually worked but could take longer than normal. It started at 2026-09-16 08:42 UTC and lasted 25 min, with full functionality restored at 2026-09-16 09:07 UTC.

GitHub Copilot reported major degraded availability for Gemini 3.8 Flash starting 2026-09-16 07:20 UTC, resolved at 2026-09-16 17:48 UTC (10 h 27 min). Requests failed at an average rate of 6.4% due to a capacity issue; other models were unaffected, and users could switch models or use 'Auto'.

GitHub reported a major Gemini 3.8 Flash degradation in Copilot starting 2026-09-16 07:20 UTC, resolved at 2026-09-16 17:48 UTC (10 h 27 min). Requests failed at an average rate of 6.4% due to a capacity issue; mitigation was reported at 11:45 UTC. Other models were unaffected; users could switch models or use 'Auto'.

Supabase reported a major Auth service outage affecting a subset of projects, starting 2026-09-16 07:17 UTC and resolved at 2026-09-16 09:33 UTC after 2 h 16 min. A fix was implemented at 09:09 UTC; no cause was stated.

Multiple Confluent Cloud services degraded across all regions, with major impact; the vendor reported onset at 21:25 UTC on September 15th 2026. The recorded incident ran from 2026-09-15 23:59 UTC to resolution at 2026-09-16 01:21 UTC (1 h 22 min); mitigation was deployed and services were recovering by 00:20 UTC on September 16th, 2026.

ChatGPT reported a major incident affecting API image generation and editing requests using gpt-image-2.5-flare, with elevated errors starting at 2026-09-15 23:50 UTC and lasting 40 min. A mitigation was applied at 23:58 UTC, and all impacted services fully recovered by 2026-09-16 00:30 UTC.

OpenAI reported a major incident causing elevated errors for API image generation and editing requests using gpt-image-2.5-flare. It started at 2026-09-15 23:50 UTC and lasted 40 min, resolving at 2026-09-16 00:30 UTC. OpenAI applied a mitigation and confirmed full recovery.

Brex reported a major Spend Management degradation: outdated Accounting transaction statuses could place transactions in the wrong tab, though changes were still being saved. It started at 2026-09-15 20:12 UTC and was resolved at 2026-09-15 23:20 UTC, lasting 3 h 7 min.

ClickHouse Cloud reported critical ClickPipes failures on Kinesis in AWS us-east-1, starting 2026-09-15 18:57 UTC and resolved 2026-09-17 00:43 UTC (29 h 45 min). AWS identified a Kinesis IAM authorization defect and began rolling back the change; ClickHouse confirmed Enhanced Fanhout (EFO) Kinesis pipes were fully operational at resolution.

Lovable reported a major outage affecting some users’ sign-in, sign-up and account actions, including project creation/sharing, API key creation, workspace membership changes and SMS two-factor authentication setup; existing sessions and published projects were unaffected. It lasted 1 h 12 min, from 2026-09-15 16:45 UTC to 2026-09-15 17:57 UTC; Lovable resolved it and advised affected users to reload.