Payment systems fail, even the biggest ones. Use this log to plan your testing, failover and customer communication before the next outage, not after it.
Last updated: Saturday, November 21, 2026 · Checked monthly. New major incidents are added once official or reputable reporting confirms the facts.
Filter by cause
Filter by region
Showing all 17 incidents, newest first.
| Date | System or provider | What happened | Impact | Cause | Lesson for banks | Source |
|---|---|---|---|---|---|---|
| Nov 16, 2026 | JPMorgan Chase (also reported at other large US banks, Zelle) United States | Customers reported login, balance and transfer problems at several large US banks, and some Zelle payments were delayed. Chase said a software bug in on-premise hardware systems caused the problem and that it was not a cyberattack. | Thousands of customers; peak Downdetector reports of about 5,000 at Chase. | Software bug | One large bank can look like an industry outage. Monitor peer and network health, and tell customers quickly what is and is not affected. | Associated Press |
| Mar 3, 2026 | FedACH (Federal Reserve) United States | ACH files were delivered about eight hours later than the normal window, and a later settlement window was also delayed. Direct deposits posted late at banks across the country. | Nationwide delay to payroll and other ACH credits and debits. | Not disclosed | Plan for a late central-operator file: customer messages, fee waivers and cut-off changes should be ready before payday. | Federal Reserve Financial Services |
| Apr 21, 2025 | Adyen Europe | A distributed denial-of-service (DDoS) attack hit payment services hosted in Europe in three waves over one evening. Card and online payments were degraded until early the next morning. No data loss was reported. | Merchants across Europe saw failed and slow payments for several hours. | Cyberattack (DDoS) | Attacks come in waves. Agree DDoS protection in advance and keep a second processor or routing option for checkout. | Adyen |
| Apr 12, 2025 | UPI (NPCI) India | A surge of transaction status-check API calls from some banks overloaded the national UPI system. The success rate fell below 50 percent for several hours. It was the third disruption in a few weeks, and NPCI then set limits on status checks. | Millions of failed shop, bill and person-to-person payments. | Traffic surge (API retries) | Retries can take down a national system. Rate-limit status polling and back off on errors. | Outlook Money |
| Feb 27, 2025 | TARGET Services: T2 and T2S (ECB) Euro area | A hardware failure stopped T2 for about 10 hours and T2S for about 8 hours. Interbank wholesale euro payments and securities settlement were not possible during the outage. | All TARGET participants; payments worth trillions of euros delayed. | Hardware failure | Central infrastructure can be down most of a day. Know your cut-off extension and contingency steps before you need them. | De Nederlandsche Bank |
| Jan 31, 2025 | Barclays UK United Kingdom | A software problem in the UK mainframe operating system slowed core banking for about three days, on payday and the self-assessment tax deadline. More than half of online payment attempts failed during the incident. | Estimated compensation of 5 to 7.5 million pounds for this incident. | Software fault | Peak dates need extra care. Run capacity and failover tests before payday and tax deadlines. | UK Parliament, Treasury Committee |
| Jan 15, 2025 | FIS (affecting Capital One and 26 other banks) United States | A local power loss and hardware failure at a data center of vendor FIS delayed deposits, payments and transfers. Capital One customers were affected for several days. | Thousands of customers missed paychecks during the mid-month pay period. | Third-party vendor (power) | A vendor fault becomes your outage. Test vendor failover and own the customer message. | Banking Dive |
| Nov 2, 2024 | EPN ACH (The Clearing House) United States | An error in one ACH file obscured some data in payment entries, so receiving banks could not process them. Direct deposits were delayed into the following week. | Under 1 percent of daily US ACH volume, but many paychecks were late. | Processing error (bad file) | One bad file can delay deposits for days. Validate files before release and have a fast correction process. | Fox Business |
| Jul 19, 2024 | CrowdStrike security update (global) Global | A faulty update to a widely used security product crashed Windows systems worldwide. Banks reported problems with apps, card payments and transfers, including PayID in Australia and Capitec in South Africa. | One of the largest IT outages on record, across many sectors. | Third-party vendor (software update) | One vendor update can reach thousands of firms at once. Stage vendor updates and keep manual fallbacks. | NYSSCPA |
| Nov 8, 2023 | Optus network (Australia) Australia | A nationwide telecom outage of about 12 hours cut mobile and internet services. Many card terminals on the Optus network could not take EFTPOS payments, and bank call centers were affected. | More than 10 million people and 400,000 businesses affected. | Third-party vendor (telecom) | Card terminals depend on telecom links. Give critical terminals a backup network connection. | Banking Day |
| Oct 10, 2023 | Zengin System (Japan) Japan | The national interbank transfer network failed during renewal work, its first major failure since 1973. Transfers at about 11 banks were disrupted for two days. | About 5 million transfers affected. | System change (renewal work) | Upgrades are the riskiest moments. Rehearse cutover and rollback, even on systems that never fail. | The Japan Times |
| Sep 7, 2023 | Square (Block) US, UK and other markets | A DNS problem made Square services unavailable for about 14 hours. Sellers could not take card payments unless they used offline mode. | Small businesses lost a day of sales. | Network configuration (DNS) | Treat infrastructure settings like code, and give merchants a working offline mode. | Square |
| Jul 8, 2022 | Rogers network (affecting Interac) Canada | A nationwide outage at telecom provider Rogers made Interac Debit and Interac e-Transfer unavailable for most of the day. | Shops across Canada were cash or credit only for a day. | Third-party vendor (telecom) | A single carrier can stop national debit. Use more than one network carrier for critical payment rails. | Interac |
| Feb 24, 2021 | Fedwire, FedACH and other Federal Reserve services United States | An automated data center maintenance process was triggered during business hours. Fedwire Funds, FedACH and other Fed services were down for roughly three hours. | All 14 Fed services were disrupted for part of the day. | Operational error (automation) | Automation needs guardrails. Lock maintenance jobs to change windows and approvals. | Banking Dive |
| Jun 1, 2018 | Visa Europe UK and Europe | A rare partial failure of a switch in a data center stopped backup systems taking over cleanly. Card payments were disrupted for about 10 hours. | About 5.2 million transactions failed across Europe. | Hardware failure | Backup that is not tested under real load is not backup. Test failover regularly. | Bank of England |
| Apr 22, 2018 | TSB Bank United Kingdom | A migration of customer accounts to a new IT platform failed. Branch, phone, online and mobile banking were disrupted, and some issues lasted until December 2018. | A significant share of 5.2 million customers affected; 48.65 million pounds in fines and 32.7 million pounds in redress. | System change (migration) | Big-bang migrations are high risk. Phase the move, test end to end, and keep a way back. | Financial Conduct Authority |
| Oct 20, 2014 | Bank of England RTGS (CHAPS) United Kingdom | The high-value settlement system was down for about nine hours. An independent review found the root cause was defects added during earlier system changes. | 142,759 CHAPS payments delayed; all were settled the same day. | System change (defects) | Defects from old changes can surface months later. Regression-test every change to core settlement. | Bank of England |
Patterns: what causes payment outages
What this tells us
- Most outages are self-inflicted. 7 of 17 came from software faults, migrations, upgrades or automation. Test every change, rehearse rollback, and avoid big-bang cutovers. See the Testing Playbook.
- Vendors fail too. Security, core-banking and telecom providers caused 4 of these outages. Map critical vendors, test their failover, and keep a second carrier or processor where it matters.
- Backup must be proven. Hardware failures became long outages when the backup did not take over cleanly. Run disaster recovery tests under real load.
- Retries can do the damage. Status-check traffic overwhelmed a national system. Rate-limit and back off.
- Peak dates multiply the harm. Payday, tax deadlines and month-end turn an outage into missed rent and fees. Freeze risky changes and add monitoring before them.
- Communication is part of recovery. Fast, honest updates, a status page and fee waivers limit the damage when payments are late.
Official status pages
For live status, check the operator directly. These pages are run by the providers themselves.
This tracker is a summary for planning and learning, not legal advice. It covers major incidents that were publicly reported; it is not a live outage feed. Descriptions are written in our own words from the linked sources, which remain the property of their publishers. Causes are as stated by the operator or regulator at the time. Corrections: sp@paymentconsultant.top. See also Payment Rules Tracker, Testing Playbook and Payment Tech News.