Salesforce Global Outage
A global outage at Salesforce, coinciding with its Dreamforce conference, took down login and core CRM services for hours and drew attention to a “legacy” shared login system that hit a resource-exhaustion cascade. Commenters dissect the company’s unusually detailed status page, debate whether its complex, highly customizable platform justifies the operational risk, and contrast the realities of running large-scale multi-tenant SaaS with the appeal of simpler, self-hosted or niche tools. Many criticize Salesforce’s UX, bloat, and dependence on consultants, while others argue its SRE and engineering practices are strong given the scale and long-term backward compatibility it has maintained.
Outage cause and scope
- Multiple commenters link Salesforce’s status/incident pages, noting a global or near-global outage.
- Incident details (from status page) say a “legacy login service” hit a resource-exhaustion cascade; a fix was tested and rolled out gradually.
- Some note the outage lasted many hours, blowing through normal three‑nines uptime for the month.
- A few report that services appear to be back up by the time of later comments.
Status page and UX
- Many criticize the status page as cluttered, with long lists of pod IDs and unclear at first glance.
- Others defend it as unusually detailed and honest: real-time per‑pod and per‑service status, drill-down incident timelines, email alerts, search, and predictable URLs for runbooks.
- Debate centers on whether this level of complexity is justified or user-hostile, but customers familiar with their pod IDs find it useful.
Dreamforce timing and internal processes
- The outage coincides with Dreamforce, prompting speculation about rushed prep, increased demo load, or code freeze dynamics.
- Former employees say there is typically a broad change freeze around Dreamforce and that line-level engineers/SREs are mostly not at the conference.
- Some wonder if change-approval processes during the freeze slowed remediation, but this remains unclear.
Diagnosis, SRE practice, and complexity
- Status updates mention attempted rolling restarts, later abandoned; some see this as “turn it off/on again,” others describe it as a standard early diagnostic step.
- Several defend Salesforce SREs as highly competent given the massive, multi-tenant PaaS scale and customer-written code.
- Discussion broadens into reliability of large distributed systems vs smaller/self‑hosted setups, with disagreement over whether “bigger SaaS” inherently adds risk or necessary resilience.
Salesforce product and ecosystem criticism
- Many dislike Salesforce as bloated “glorified CRUD” or “slop,” with complex, messy schemas and painful integrations.
- Others argue it delivers extensive enterprise features and integrations that simpler, cheaper CRMs cannot match.
- There is recurring frustration with UX, tooling, and the enterprise sales/pricing model, alongside acknowledgment of Salesforce’s long-term platform stability and evolution.