字号·· | 护眼
印度斯坦时报

美国科技故障频发的一周背后:例行补丁与老化技术撼动国家数字骨干Behind the week of US tech outages: Routine patches and ageing tech shake country's digital backbone

点「原文对照」整页切到原文,或双击某段只看那段的原文。

过去一周发生的一系列技术故障扰乱了全美各地的航空出行、银行业、政府服务、云平台和在线法院系统。这次同时发生的故障引发了人们的疑问:这些是否是与人工智能相关的故障,或是针对该国关键数字基础设施的网络攻击。

A series of technology outages over the past week disrupted air travel, banking, government services, cloud platforms and online court systems across the United States. The simultaneous outage has raised questions over these being AI-related outages or a cyberattack on the country's critical digital infrastructure.

然而,联邦官员证实,例行软件补丁和老化硬件引发了大部分此类连环故障。国务卿马可·鲁比奥周二也表示,情报机构发现“零证据”将这些事件与敌对行为者联系起来。

However, federal officials confirm that routine software patches and ageing hardware triggered most of these cascading failures. Secretary of State Marco Rubio also stated on Tuesday that intelligence agencies found "zero evidence" linking the incidents to hostile actors.

尽管唐纳德·特朗普政府作出了安抚,但连环故障的绝对规模已引发国会山两党对简报会的要求。众议院少数党领袖哈基姆·杰弗里斯和参议院共和党人希望了解例行维护是如何同时扰乱重要的国家安全和消费基础设施的。过去一周有哪些系统瘫痪?微软365(11月12日至14日)网络网关服务器中一个长期存在的软件缺陷导致Outlook、Teams和Copilot在70个国家离线约36小时。

Despite reassurances from the Donald Trump administration, the sheer scale of the cascading breakdowns has sparked bipartisan demands for Capitol Hill briefings. House Minority Leader Hakeem Jeffries and Senate Republicans want answers on how routine maintenance disrupted vital national security and consumer infrastructure simultaneously. What went down over the past week? Microsoft 365 (November 12-14)A latent software defect in networking gateway servers knocked Outlook, Teams, and Copilot offline for roughly 36 hours across 70 countries.

受影响的有公司、政府机构、航空公司和金融机构。在整个事件期间,Downdetector记录了超过740万份报告,峰值接近每分钟2万份报告。微软表示,没有迹象表明存在恶意活动,停机是由服务器问题引起的。

Among those impacted were companies, governmental organizations, airlines, and financial institutions. Over 7.4 million reports were registered by Downdetector throughout the incident, with a peak of almost 20,000 reports per minute. According to Microsoft, there was no indication of malicious activity and the downtime was due to server problems.

他们后来查明,一个存在数月的潜在软件漏洞、流量故障转移以及服务器容量减少的组合是导致网络网关不堪重负的根本原因。联邦航空管理局(11月在联邦航空管理局实施了25年来持续时间最长的全国性地面停飞令后,有2,000多个航班被取消。周五,负责维护联邦航空管理局长期电信基础设施的承包商L3Harris,在阿肯色州的一个无线电发射台安装了一个新路由器。

They later identified a combination of a months-old latent software bug, a traffic failover, and reduced server capacity as the root cause that overwhelmed network gateways. Federal Aviation Administration (Nov. 13)More than 2,000 flights were cancelled after the FAA imposed the longest nationwide ground stop in 25 years. On Friday, a new router was installed at a radio transmitter site in Arkansas by L3Harris, the contractor in charge of maintaining the FAA's long-standing telecommunications infrastructure.

根据最新报道,联邦航空管理局下令地面停飞,是因为这个新路由器开始同时发送过多的连接信号,导致网络过载,并干扰了空中交通管制员与空中飞机进行通信的能力。

According to the latest reporting, the FAA issued the ground halt because this new router started sending out too many connection signals at once, overloading the network and interfering with air traffic controllers' capacity to interact with airplanes in the sky.

为了稳定航空交通系统,联邦航空管理局经审查发现阿肯色州和得克萨斯州的路由器是问题所在后,官员们将这些路由器进行了隔离。此后,联邦航空管理局已暂停在其全国数千个电信枢纽安装新路由器。Workday(11月13日至14日)11月13日周五,Workday约11,500家客户组织中的数名员工无法登录。他们没有看到人力资源门户网站,而是看到一个显示“Workday当前不可用”字样的维护页面。

In order to stabilize the air traffic system, officials isolated the routers in Arkansas and Texas after the FAA's review revealed them to be the source of the problem. The FAA has since paused the installation of new routers across its thousands of telecommunications hubs throughout the country. Workday (Nov. 13–14)Employees at several of Workday's approximately 11,500 customer organizations were unable to on Friday, November 13. They instead saw a maintenance page with the message "Workday is currently unavailable" instead of their HR portal.

就在前一天,由于另一次微软365故障,Teams和Outlook在全球大部分地区无法使用,这引发了人们对此次故障与全球故障有关的担忧。

It happened one day after Teams and Outlook were unavailable in most of the world due to a different Microsoft 365 outage raising concerns that it was connected to the global outage.

然而,根据《UC Today》报道的Workday给客户的信息,此次故障发生在其灾难恢复站点的测试期间。在测试期间,“灾难恢复站点内的多租户服务突然链接到了生产数据库。”因此,Workday停止了客户端访问,以便在核实和重新索引信息的同时,保护不一致的数据免受影响。没有迹象表明这是一次网络攻击。这是安全演练内部的一次故障。

However, according to Workday's customer message, which was reported by UC Today, the outage occurred during a test of their disaster recovery (DR) site. During the test, “tenanted services within the DR Site suddenly linked to production databases.”Therefore, Workday stopped client access to safeguard the inconsistent data while it verified and re-indexed the information. There is no indication that this was a cyberattack. It was a failure inside a safety exercise.

摩根大通与美国各大银行(11月16日)周一,一起本地硬件故障导致这家美国最大银行的交易处理延迟。本周,多家金融机构报告了时断时续的网上银行和支付问题。

JPMorgan Chase & US Banks (Nov. 16)An on-premise hardware bug delayed transaction processing at America's largest bank on Monday. Several financial institutions reported intermittent online banking and payment issues during the week.

消费者抱怨转账延迟、账户余额消失、交易不成功以及在使用手机银行服务时遇到麻烦。这是由摩根大通的软件错误导致的。这一瓶颈导致美国银行、富国银行和美国合众银行的客户遭遇了间接转账错误。

Consumers complained about delayed transfers, missing account balances, unsuccessfuls, and trouble using mobile banking services. This occurred as a result of a JPMorgan Chase software error. The bottleneck caused secondary transfer errors for customers at Bank of America, Wells Fargo, and US Bank.

11月17日的报道称,摩根大通公司在周一晚间发现并修复了该技术问题,其系统得以恢复正常的交易处理。联合航空(11月16日)在全美联邦航空管理局实施大规模地面停飞几天后,联合航空遭遇了又一次与航空相关的技术故障,不得不实施自身的地面停飞。

Reports from November 17 state that JPMorgan Chase & Co. found and fixed the technical problem by Monday night, allowing its systems to resume regular transaction processing. United Airlines (Nov. 16)A few days after the big national FAA ground stop, United Airlines experienced another aviation-related technological breakdown and had to make a ground stop of its own.

据称,问题之一是帮助飞行员调整飞机重量和平衡的网络应用程序。

Among the issues were allegedly web-based applications that assisted pilots in adjusting the aircraft's weight and balance.

该航空公司在一份声明中表示,空中飞行的航班以及其联航快运支线网络旗下的航班未受影响。他们还表示,“没有迹象表明这是一个与网络安全相关的问题”。国家安全局(11月17日)周二上午,国家安全局的内部通信系统至少部分瘫痪。国会山的一位消息人士告诉福克斯新闻:“这仅仅是国家安全局对系统的例行修补。”

In a statement the airlines said that flights that were in the air and flights on its United Express regional network did not get affected. They also said there was “no indication that this was a cybersecurity-related issue.”National Security Agency (Nov. 17)On Tuesday morning, the National Security Agency's (NSA) internal communications system was at least partially down. "This is merely ordinary patching of systems" at the NSA, a Capitol Hill source told Fox News.

据哥伦比亚广播公司报道,官员们在米德堡实施了“灰码”错峰到达时间表,以防止过载,同时维持关键任务运营。

Officials enacted "Code Gray" staggered arrival schedules at Fort Meade to prevent overloads while maintaining mission-critical operations, CBS reported.

随后证实,国家安全局在“计划中的例行维护期间遇到了‘临时性能下降’”。苹果服务(11月17日)一次内部系统更新导致iCloud登录、Apple Pay、Apple Music和App Store在周二下线了几个小时。苹果就此次故障提供了最新消息,随后证实问题已得到解决。

Later it was confirmed that the NSA experienced “temporary degradation during scheduled routine maintenance." Apple Services (Nov. 17)An internal system update knocked iCloud sign-in, Apple Pay, Apple Music, and the App Store offline for several hours on Tuesday. Apple provided an update regarding the outage, and later confirmed that the problems have been fixed.

据《纽约时报》报道,该公司在一份声明中表示:“在今早的一次内部更新导致一些故障之后,我们的许多系统已经恢复并重新运行。”

In a statement reported by The New York Times, the corporation said, “Many of our systems are back up and running after an internal update caused some outages this morning. We’re working to get the remaining systems online as soon as possible.”

我们正在努力尽快恢复其余系统的运行。”该公司排除了网络攻击导致中断的可能性。恶意软件攻击(11月17日)然而,苹果公司(Apple)成了网络攻击事件的牺牲品。

The company ruled out cyberattack as cause for the outage. Malware attack (Nov. 17)Apple however was the prey to one cyberattack incident.

据报道,安全研究人员在约200家公司使用的在线工具中发现了恶意软件,这些公司包括Equifax、Zoom和汇丰银行(HSBC)。

Security researchers found malicious malware in online tools utilized by around 200 companies, including Equifax, Zoom, and HSBC, according to reports.

该恶意代码旨在窃取支付卡详细信息、转移加密货币交易、截取Cookie和屏幕截图,并提供iOS漏洞利用工具包以感染iPhone和iPad。

The malicious code aimed to skim payment card details, divert cryptocurrency transactions, capture cookies and screenshots, and deliver iOS exploit kits to infect iPhones and iPads.

帮助企业创建网站的AI平台Optimizely成为了攻击目标。网络安全公司Report URI的研究人员于周一发现了这起黑客攻击事件。

Optimizely, an AI platform that enables businesses to create websites, was the target of the attack. The hack was discovered on Monday by researchers at the cybersecurity company Report URI.

Optimizely周二在其网站上发文称,“一名未授权人员利用了”其实验(Experimentation)工具中的“一个漏洞”。“我们已经修复了该漏洞,并从受影响的项目中删除了注入的代码。”加州法院门户网站中断(11月17日)在供应商Journal Technologies检测到针对其公共门户技术的的可疑活动后,加州各地的法院网站(包括萨克拉门托县、阿拉米达县和克恩县)陷入瘫痪。该公司表示,没有证据表明刑事司法信息遭到泄露。所有这些中断之间是否存在关联?

Optimizely stated that "an unauthorized person exploited a vulnerability" in its Experimentation tool in a post on its website on Tuesday. “We've fixed the vulnerability and removed the injected code from affected projects.”California court portal outage (Nov. 17)Court websites across California, including Sacramento, Alameda and Kern counties, went offline after vendor Journal Technologies detected suspicious activity targeting its public portal technology. The company said there was no evidence that criminal justice information had been exposed. Is there a connection between all the outages?

调查人员强调,尽管发生时间相近,但其中几起事件似乎并无关联。

Investigators have emphasised that several incidents appear unrelated despite their timing.

网络安全专家强调,主要的故障是孤立的技术故障,而不是统一的阴谋。

Cybersecurity experts emphasised that the primary disruptions are isolated technical failures rather than a unified plot.

阿肯色大学小石岩分校(UA Little Rock)网络安全教育与外联主任桑德拉·莱特曼(Sandra Leiterman)对哥伦比亚广播公司(CBS)旗下附属电视台THV11表示:“这可能是网络攻击吗?完全可能,这是潜在的解释之一。但远不止于此,还包括软件故障、配置故障、设备老化、共享基础设施中断。”

Sandra Leiterman, Director of Cybersecurity Education and Outreach at UA Little Rock said to CBS-affiliate THV11, “Could it be a cyber-attack? Absolutely, that's one potential explanation. But there's a lot more than that, software failures, configuration failures, aging equipment, shared infrastructure outages.”

她补充说:“我不认为其中任何一个系统的安全性比另一个低。我觉得这就像是一种普遍存在的现象。”我认为我们的银行业一点也不缺乏安全感。

She added, “I don't think any one of these systems is less secure than the other. I think it's just like something in the air. I don't think our banking industry is insecure at all.”

阅读媒体原文 ↗