Alibaba Cloud Outage in November 2023 Affects Numerous Services and Regions

Strange incidents happen every year, but this one was especially unprecedented: even Alibaba suffered a complete collapse.

On October 23, Alibaba's knowledge base platform Yuque experienced an unprecedented P0-level incident, rendering client, website, and mobile access unusable for nearly eight hours.

Because it fell on a Monday, many users were affected, leading to widespread complaints on social media.

On November 12, the day Alibaba announced its Double‑11 sales results, the company was hit again. Around 5 p.m., a large number of users reported being unable to use Alibaba products.

Soon, Taobao, Xianyu, DingTalk, and many other services went down, flooding hot searches.

The root cause was traced to Alibaba Cloud; since most Alibaba applications are deployed on its own cloud infrastructure, the failure of this “high‑speed highway” prevented all traffic from flowing.

According to user reports, the outage impacted a wide range of Alibaba services, including Ele.me, Amap, and even the multi‑active data centers of Alipay.

Not only Alibaba's own products were affected; over 300,000 enterprise customers using Alibaba Cloud services also faced potential total business paralysis.

Many third‑party platforms that rely on Alibaba Cloud, such as CSDN and Blog Garden, also suffered.

Users noted that the outage even impacted daily life, with parking gates failing to lift and supermarkets unable to process payments.

Both large‑scale business operations and small devices like dryers and smart speakers were affected.

Some speculated that the timing, coinciding with Double‑11, might have contributed to the failure, while others joked it was a side effect of “screen‑jump ads” and “cost‑cutting measures”.

The affected regions spanned mainland China, Hong Kong, Macau, Taiwan, other Asian regions, Europe, North America, the Middle East, and various financial and government clouds.

Alibaba Cloud listed the impacted products, ranging from distributed application services, message queues, micro‑service engines, tracing, high‑availability services, monitoring, AI recommendation, OpenSearch, data hubs, Elasticsearch, graph compute, Flink, MaxCompute, Hologres, DataWorks, media services, object storage, NAS, relational databases, security services, container services, serverless functions, and many more.

The outage also covered numerous data center locations, including Beijing, Shanghai, Shenzhen, Hong Kong, Mumbai, Silicon Valley, London, Seoul, Tokyo, Dubai, Chengdu, Singapore, Sydney, Kuala Lumpur, and many others.

At 18:14 on the day of the incident, Alibaba Cloud finally issued an official response, acknowledging the abnormality in console access and API calls starting at 17:44 Beijing time and apologizing for the inconvenience.

The recovery process took about three and a half hours, with services largely restored by 21:11.

The incident, likely caused by an OSS failure, left users with no self‑help options, forcing them to wait for restoration.

It exposed the vulnerability of cloud computing services, reminding that even high‑reliability, high‑availability, and elastic services can still experience significant failures.

Questions remain about how future incidents will be mitigated and whether Alibaba's technical safeguards can meet the challenges of an increasingly complex internet environment.

Ultimately, the massive impact of this outage serves as a cautionary tale for all large internet companies to strengthen their technology and service reliability to avoid similar costly disruptions.