Several trading infrastructure providers and exchanges reported technical failures between November 2 and 4, 2026, affecting market-data delivery, order execution, liquidation-price displays and customer withdrawals. In today’s trading infrastructure roundup, Databento disclosed permanent gaps in its Eurex and EEX historical data following a Deutsche Börse network reconfiguration, Kraken corrected misleading liquidation-price estimates, and CMC Markets resolved a UK data-center execution problem. Crypto.com also experienced a predictions trading interruption, while Bitso reported separate incidents affecting account access, trading and Mexican peso withdrawals.
Databento confirms permanent Eurex and EEX data gaps
Market-data provider Databento has confirmed that portions of its November 2 and November 3 historical data for Eurex and European Energy Exchange feeds are permanently unavailable following a network infrastructure failure at Deutsche Börse’s Frankfurt FR2 data center.
The incident began at 04:47 UTC on November 2, disrupting Databento’s XEUR.EOBI and XEEE.EOBI feeds. Both its Live API and Historical API were affected, with the historical service initially described as experiencing degraded availability.
At 17:48 UTC, Databento reported that it had identified an unexpected network reconfiguration within Deutsche Börse’s FR2 facility. According to the provider, the change caused a complete loss of its feeds at that location.
Deutsche Börse subsequently scheduled a reversal of the configuration change following the end of Eurex’s extended trading hours. The following morning, Databento confirmed that traffic had not returned as of 02:00 UTC and that the exchange operator was reviewing the network configuration against its internal records.
Live data access was restored at 11:20 UTC on November 3, approximately 30 hours and 33 minutes after the initial incident notification.
But the recovery came with a permanent limitation. Databento stated at 11:23 UTC that historical data for November 2 and the affected portion of November 3 had been permanently lost.
The service came back, but those missing records, they were still missing.
At 14:12 UTC, Databento confirmed that live and intraday historical services were available again for both affected datasets. It subsequently marked the incident fully resolved at 20:54 UTC on November 4, although the earlier disclosure of permanent historical gaps remained unchanged.
The data loss concerns Databento’s historical coverage of the affected feeds. Its incident report does not establish that Deutsche Börse or other market-data vendors permanently lost their own corresponding exchange records.
For trading firms using Databento’s data to reconstruct order-book conditions, validate execution algorithms or backtest strategies, the missing sessions create an archival limitation. Eurex EOBI data provides order-book information used in market microstructure research and automated trading systems, where accurate sequencing of market events is often essential.
Databento has not disclosed the number of subscribers affected, whether alternative archives can support a partial reconstuction of the missing sessions or whether customers will receive service credits.
Kraken fixes incorrect liquidation prices on perpetual futures
Kraken reported a separate trading-platform incident on November 3 after discovering that some customers were seeing incorrect estimated liquidation prices for perpetual futures positions on Kraken Pro.
The exchange acknowledged the problem at 18:40 UTC, specifying that the displayed estimates could be incorrect on its web and mobile trading interfaces.
Kraken identified the problem at 19:17 UTC and implemented a fix eight minutes later. The exchange continued monitoring the results before confirming complete resolution at 21:16 UTC, approximately two hours and 36 minutes after the initial notice.
The number on screen was wrong, Kraken says, but that is about as far as the technical disclosure goes.
Estimated liquidation prices are an important input for traders managing leveraged positions. They indicate the approximate market level at which an account’s margin could become insufficient under the exchange’s liquidation methodology.
As explained in DaveFinances’ analysis of perpetual futures and liquidation risks, actual liquidation thresholds can depend on maintenance margin, collateral, mark prices, funding costs and position-specific calculations.
Kraken’s incident report describes incorrect displayed estimates, not confirmed errors in the exchange’s underlying liquidation engine.
The company did not report customers being liquidated at incorrect prices or disclose evidence that actual margin requirements were miscalculated. It also provided no figures for affected positions, contract types or any financial losses.
Those missing details limit the assessment of customer impact. A displayed estimate could have affected trading decisions even if the underlying liquidation engine continued operating correctly, but no specific losses attributable to that scenario have been established.
CMC Markets suffers another execution-related disruption
CMC Markets reported intermittent trading difficulties on November 3 involving orders routed through its UK7 data center.
The London-listed broker published its first notice at 11:59 UTC, warning customers that technical difficulties could affect their ability to trade. The incident was classified under the CMC CFD Global trading service.
CMC marked the problem resolved at 12:15 UTC, giving the recorded incident a duration of approximately 16 minutes.
Sixteen minutes, and the notice offers little more than that.
CMC did not identify the underlying technical fault or disclose whether affected orders were delayed, rejected or executed incorrectly. The company also did not specify which instruments, account types or customer regions were affected.
The latest problem follows an October 1 order-processing disruption at CMC Markets that lasted approximately 65 minutes. During that earlier incident, the broker warned that technical problems could affect customer orders but did not establish the exact execution consequences.
CMC also experienced a separate pricing disruption affecting its 24/5 US securities offering on October 21, attributed to an external pricing provider.
More recently, a November 1–2 incident caused incorrect client positions to appear on its web and mobile platforms for approximately seven hours and 37 minutes. Another technical problem reported on November 2 affected the display of payment information in the platform’s History section.
The UK7 incident differs from those display-related problems because the latest notice specifically identifies trade execution for orders routed through a particular data center.
CMC reported group-wide platform uptime of 99.98% for its financial year ended March 2026, compared with 99.93% in the preceding year. The November incidents fall in the subsequent financial year and cannot be directly incorporated into those previously reported availabilty figures.
CMC has not published findings connecting the recent incidents to a common infrastructure issue. Its latest notice confirms restoration but leaves the execution impact and number of affected orders undisclosed.
Crypto.com predictions and Bitso payments hit by outages
Crypto.com experienced an interruption affecting predictions trading, including strike options, on November 3, temporarily disrupting access to the products through its main application.
The company acknowledged the problem at 07:01 Hong Kong time, equivalent to 23:01 UTC on November 2, and said its technical team was investigating.
At 07:22 HKT, Crypto.com confirmed that investigations were continuing. The incident was marked resolved at 09:19 HKT, approximately two hours and 18 minutes after the first notification.
The timing is awkward here, some of these contracts expire far sooner than the outage lasted.
Crypto.com’s strike options include contracts linked to whether cryptocurrency prices exceed specified levels at expiration. Some supported contracts have durations as short as five or 20 minutes, leaving limited time for traders to change positions before settlement.
The November incident follows an October 2 prediction-order outage at Crypto.com, when a backend failure prevented customers from placing prediction orders for approximately 43 minutes.
The latest disruption lasted more than three times as long as the October incident. Crypto.com has not confirmed whether the same infrastructure component was responsible.
Its November 3 notice did not specify whether customers were prevented from opening positions, closing existing contracts or both. The exchange also did not disclose affected contract volumes, missed settlements or compensation arrangements.
Separately, Mexican cryptocurrency exchange Bitso reported two service incidents within approximately two days.
The first affected login and account access for retail and business customers using Bitso’s website and application on November 3. The company acknowledged an operational problem at 15:24 UTC and confirmed that services had recovered by 18:42 UTC, approximately three hours and 18 minutes later.
Customers still experiencing access problems were advised to log out and log back into their accounts.
A second incident occurred between 00:00 and 00:48 UTC on November 4, affecting trading performance and Mexican peso withdrawals through the SPEI payment network.
Bitso reported that some customers experienced slower trading and delayed MXN withdrawals during that 48-minute period. Its technical team identified the problem and confirmed that affected services were restored.
The exchange also stated that delayed withdrawals were automatically retried and processed once normal operations resumed.
Two incidents in quick succession, but no confirmed connection between them.
SPEI is Mexico’s interbank electronic payment infrastructure, administered by Banco de México. According to the central bank’s official SPEI documentation, the system operates continuously and processes domestic interbank transfers through a real-time settlement framework.
Bitso’s disclosure does not establish a nationwide SPEI malfunction. The confirmed disruption involved the exchange’s withdrawal processing, and no evidence has been published showing that the broader interbank network experienced the same problem.
Bitso said it would share more information following its review. The exchange has not disclosed the amount of delayed withdrawals, the number of affected customers or whether the two incidents originated from related systems.
All five providers have reported restoration of the affected live services. Databento’s permanently missing historical records remain the clearest lasting operational consequence identified in the disclosures, while the other incidents have left unanswered questions about trading decisions, execution records and the extent of customer disruption. None of the providers has published transaction-level evidence sufficient to establish aggregate customer losses arising from these events.
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape. You can reach out to him via his social media accounts:
Linkedin: https://www.linkedin.com/in/johan-shamshad-742851262/
X: https://x.com/Yasmine_FX
Investing: https://www.investing.com/members/contributors/279781574