Date: November 28, 2025
Location: Chicago / London / Singapore
Key points
- CME Group halted nearly all Globex futures and options trading late Thursday after a cooling failure at a CyrusOne data center in the Chicago area. [1]
- Benchmarks from S&P 500 and Nasdaq 100 futures to WTI crude, gold, U.S. Treasuries and Nikkei contracts went dark for hours, freezing price discovery across global markets. [2]
- By early afternoon in London, CME had begun partially restoring operations, with currencies resuming trading on its EBS forex platform, while many futures markets remained halted. [3]
A Black Friday nobody expected
Black Friday, typically a quieter session for global derivatives, turned into one of the most disruptive trading days in years after CME Group suffered a major technical outage tied to a third‑party data center cooling failure.
Late on Thursday evening U.S. time, CME reported that its Globex futures and options markets and its EBS foreign‑exchange platform had been halted due to “a cooling issue at CyrusOne data centers,” according to repeated system alerts and a matching post on X (formerly Twitter). [4]
The disruption rippled almost instantly through Asia‑Pacific and European sessions on Friday, freezing some of the world’s most traded contracts and leaving brokers “flying blind” with no live prices for key benchmarks. [5]
How the CME outage unfolded
Timeline of the halt
CME’s own incident log shows that trouble began on the evening of Thursday, November 27 (Chicago time):
- Around 20:40 CT, CME Globex futures and options markets were first flagged as “halted due to a technical issue,” along with Bursa Malaysia Derivatives (BMD) markets and CME’s EBS FX platforms. [6]
- By late evening and into the early hours of November 28, CME updated those alerts to specify a “cooling issue at CyrusOne data centers” as the root cause, repeatedly warning clients that all Globex futures and options remained halted and that pre‑open details would follow “in the near term.” [7]
Public reporting from multiple outlets aligns with that internal timeline. Reuters notes that the outage began in Asian hours and was still ongoing by 11:20 GMT on Friday, making it one of the longest CME disruptions in years. [8]
According to the Economic Times, Globex was effectively shut down after the failure at CyrusOne’s CHI1 facility, with pricing for oil, gold, Treasuries and U.S. equity futures frozen and no updates showing in LSEG data feeds by around 5:30 a.m. ET. [9]
U.S. stock index futures “frozen in time”
A second Reuters report focused on U.S. index futures shows just how abrupt the stop was:
- CME’s S&P 500, Nasdaq 100 and Dow futures all show last trades at 9:44 p.m. ET on November 27, after which quotes stopped updating. [10]
- CME first posted publicly about the outages around 9:40 p.m. ET, citing the cooling issue. [11]
With futures frozen, traders turned instead to ETFs such as SPDR S&P 500 (SPY) and Invesco QQQ Trust (QQQ) to gauge market sentiment ahead of a shortened U.S. cash session, which still opened as scheduled. [12]
Which markets were hit?
The outage was so broad that it knocked out a swath of futures and FX benchmarks across asset classes:
- Equity index futures: S&P 500, Nasdaq 100, Dow, Nikkei 225, Russell 2000 and others saw trading halted or quotes go stale. [13]
- Rates & bonds: U.S. 10‑year Treasury futures and other interest‑rate contracts on CME stopped updating, though CME’s fixed‑income platforms BrokerTec U.S. Actives and BrokerTec EU remained open. [14]
- Commodities:
- FX:
- CME’s EBS spot FX platform, which typically handles close to $60 billion a day in major currency pairs, went dark in euro/dollar and dollar/yen, forcing banks to reroute FX flow to other venues. [17]
The disruption extended beyond CME’s own branded markets. Bursa Malaysia said trading of its derivatives products listed on Globex would not resume at all today, even as contract settlement processes continued. [18]
Root cause: cooling failure at CyrusOne’s CHI1 data center
While CME has described the problem as a “cooling issue at CyrusOne data centers,” the most detailed technical description so far has come from the data center operator itself and specialized infrastructure media.
Data Center Dynamics, citing CyrusOne, reports that:
- The incident occurred at the CHI1 data center in the Chicago area, which supports critical workloads for CME and other clients.
- On November 27, CHI1 suffered a chiller plant failure that knocked out multiple cooling units, forcing protective shutdowns to prevent equipment damage.
- CyrusOne says it has restarted several chillers at limited capacity and deployed temporary cooling equipment while full capacity is restored. [19]
CME’s dependence on CyrusOne dates back to a 2016 sale‑leaseback deal, when CME sold its Aurora, Illinois, data center campus—now three facilities totalling about 450,000 square feet and 109 MW of IT capacity—to CyrusOne while remaining a tenant. [20]
The outage also lands awkwardly against the backdrop of CME’s longer‑term plan to shift Globex infrastructure onto a dedicated Google Cloud region in Chicago, a migration announced in a November 24 Globex notice and expected to begin preview phases in 2026. [21]
CME’s response and the first signs of recovery
From the outset, CME told clients that it expected to resolve the issue “in the near term” but offered no hard timeline. [22]
For most of the Asian and early European sessions, the answer to the question “When will CME reopen?” was simply: nobody knew.
By early afternoon in London, however, there were concrete signs of partial restoration:
- A notice on CME’s site, cited by Investing.com, said that the EBS market—CME’s primary platform for FX trading—reopened at 12:00 p.m. London time, even as other futures markets remained offline. [23]
- LSEG data, reported by Reuters, showed currencies resuming trading on EBS around the same time. [24]
- Bloomberg headlined the development as “CME Partially Restores Operations With Restart of Forex Platform,” noting that the exchange had begun gradually bringing systems back after hours of disruption. [25]
At the time of writing, CME had not yet indicated when equity, rates and commodity futures on Globex would fully reopen, and its own system‑alert page still listed EBS markets as “currently halted” in the most recent updates before the FX restart. [26]
How brokers and traders coped with “flying blind” markets
With core price feeds frozen, brokers and trading firms were forced into improvisation.
Reuters quotes brokers who:
- Suspended trading outright in products linked to CME futures, including some commodity and index derivatives.
- For other instruments, constructed internal synthetic prices based on alternative venues or related assets—an approach that raised risk, given the lack of direct hedging in the underlying futures. [27]
CMC Markets’ regional leadership in Asia and the Middle East described the outage as highly unusual and warned of heightened volatility once markets reopen, especially because many positions could not be adjusted during the freeze. [28]
Some CFD and FX brokers—including LiteFinance and KCMTrade—issued notices to clients pausing trading or widening conditions on instruments tied to CME products such as:
- S&P 500, Nasdaq 100 and Dow futures
- WTI crude
- Agricultural contracts like corn, wheat and soybeans
- Precious metals including platinum and palladium
until normal Globex connectivity and liquidity return. [29]
Why the timing amplified the shock
On one level, CME may have been “lucky” that the failure hit during a holiday‑thinned Black Friday session rather than during a busy mid‑week U.S. open. Several analysts noted that volumes were already expected to be light after the Thanksgiving holiday. [30]
But the outage also landed at a sensitive moment for markets:
- It coincided with month‑end, when many futures contracts are rolled and hedges rebalanced—activities that rely heavily on CME benchmarks. [31]
- Commodity markets were already volatile, with oil and gold responding to geopolitical risks, making the absence of transparent WTI and COMEX gold futures prices particularly problematic. [32]
- In equities, investors were trying to judge whether a difficult November for global stocks was stabilising, and frozen S&P and Nasdaq futures removed a key early signal normally used ahead of the U.S. cash open. [33]
ETF markets, alternative futures venues and OTC pricing helped fill some gaps, but the outage underscored just how central CME’s infrastructure is to global price discovery and risk management.
A wake‑up call on market plumbing
For regulators, risk officers and market structure specialists, Friday’s events are likely to become a case study in technology concentration risk:
- CME is the world’s largest exchange operator by market value, with average daily derivatives volume of 26.3 million contracts in October alone. [34]
- A single cooling failure at one third‑party data center was enough to freeze trading in core benchmarks across equities, bonds, commodities and FX, despite redundancies elsewhere in the system. [35]
Analysts and commentators have already begun drawing parallels with past outages—such as the 2013 Nasdaq “flash freeze” and a 2014 CME technical halt in agricultural futures—but note that this time the trigger was a physical infrastructure failure, not just a software bug. [36]
Several market‑structure think‑pieces published today argue that the incident strengthens the case for:
- More geographically distributed, active‑active data‑center architectures for exchanges
- Tighter resilience and failover standards for critical third‑party providers
- Clearer real‑time communication protocols to help brokers and end‑investors understand what is—and isn’t—tradable when core benchmarks go dark [37]
What happens next?
As of late afternoon on November 28, 2025, the situation is still evolving:
- FX trading on EBS has restarted, but many futures and options contracts remain halted or in limbo pending pre‑open guidance from CME. [38]
- Bursa Malaysia Derivatives has told market participants that its Globex‑hosted contracts will not trade again today, though clearing and settlement are functioning normally. [39]
- Oil markets, equities and rates traders are preparing for potentially sharp moves once full price discovery returns, especially given the backlog of orders and the holiday‑thinned liquidity backdrop. [40]
In the days ahead, expect:
- Detailed post‑mortems from CME and CyrusOne on the exact sequence of failures and why backup systems didn’t prevent a full trading halt. [41]
- Scrutiny from regulators and clearing members on CME’s business continuity plans, particularly as it prepares to migrate significant workloads to Google Cloud in Chicago. [42]
- Debate among market participants about whether critical global benchmarks have become too dependent on a handful of physical sites and vendors—and what redundancy should look like in a “cloud plus colocation” world. [43]
CME Group outage hits U.S. futures trading
References
1. www.cmegroup.com, 2. www.reuters.com, 3. www.investing.com, 4. www.cmegroup.com, 5. www.reuters.com, 6. www.cmegroup.com, 7. www.cmegroup.com, 8. www.reuters.com, 9. m.economictimes.com, 10. www.reuters.com, 11. www.reuters.com, 12. www.reuters.com, 13. www.reuters.com, 14. www.reuters.com, 15. www.reuters.com, 16. www.reuters.com, 17. www.reuters.com, 18. www.bernama.com, 19. www.datacenterdynamics.com, 20. www.datacenterdynamics.com, 21. www.cmegroup.com, 22. www.cmegroup.com, 23. www.investing.com, 24. www.marketscreener.com, 25. www.bloomberg.com, 26. www.cmegroup.com, 27. www.reuters.com, 28. www.reuters.com, 29. www.litefinance.org, 30. www.reuters.com, 31. www.ft.com, 32. www.theedgemarkets.com, 33. www.reuters.com, 34. www.reuters.com, 35. www.datacenterdynamics.com, 36. en.wikipedia.org, 37. www.ainvest.com, 38. www.investing.com, 39. www.bernama.com, 40. www.reuters.com, 41. www.datacenterdynamics.com, 42. www.cmegroup.com, 43. www.ainvest.com
