X Outage Continues
Tens of thousands impacted as X suffers global outage for second consecutive day.
Key Points
During the morning spike, Downdetector data showed roughly half of X outage reports were about the mobile app, about 30% about the feed or timeline, and around 15% about the desktop website.
The outage peaked at about 27,100 reports by 10 a.m. ET, with parallel outages reported on other major sites such as Reddit and Zoom according to Downdetector.
New Delhi-based users reported an evening disruption with feeds failing to refresh and the mobile app crashing, indicating continued impact beyond the morning regional spike.
Late-evening outage activity spiked again, with Downdetector logging 968 reports by 7:35 p.m. local time, and categories including app, login, server connection, feed and timeline, posting, and website access.
X went down globally on Monday, knocking out service for tens of thousands of users in the United States, United Kingdom, India, and beyond. Complaints on Downdetector peaked at between 27,100 and 35,659 reports by around 10 a.m. ET, according to Forbes and Reuters, making it one of the platform's worst disruptions this year.
The outage hit mobile apps hardest — about 45–50% of reports involved the app — while feed and timeline failures made up roughly 30% and desktop issues accounted for around 15%, per Forbes. It followed a smaller disruption the day before that lasted about an hour, according to UNILAD.
Outage Erupts in Minutes, Cripples Key Features
The first reports trickled in just after midnight ET, mostly from users in Europe and Asia, according to DesignTAXI. Then at 9:30 a.m. ET, complaints exploded — jumping from 584 to over 15,000 in under 15 minutes, Forbes reported. Users saw "503 Service Unavailable" errors, empty timelines, and failed logins.
By 10:15 a.m. ET, the damage spread beyond X. Reddit logged 2,864 complaints, Zoom hit 3,245, and Microsoft Teams reached 1,312, all tracked on Downdetector, per Forbes. Internet monitoring group NetBlocks confirmed the X disruption was not a government block. It called the root cause "timeline backend API errors," according to The National News.
Cloudflare Alerts Point to Wider Infrastructure Breakdown
At 9:35 a.m. ET, Cloudflare — which powers over 20% of all websites — issued a status alert citing "elevated error rates and latency" across its global network, according to Newsweek. Cloudflare said it had "identified the issue" and was applying a fix by 10:30 a.m. ET. Experts at TechRadar said the simultaneous outages at X, Reddit, and Zoom pointed to a cloud infrastructure failure rather than a bug specific to X.
Some outlets speculated a possible DDoS attack — a flood of fake traffic designed to overwhelm servers — but no outlet confirmed one for this event. X owner Elon Musk stayed silent on the platform. He has previously blamed "massive cyberattacks" for similar failures, according to MyBroadband.
India Hit Again Hours After the US Recovered
While much of the Western world saw service return by midmorning, Indian users faced a fresh wave of disruptions in the evening. Between 7:30 and 8:00 p.m. IST, Downdetector India reports surged from 336 to 968 in three minutes, then peaked at 1,820 by 7:45 p.m. IST, according to Hindustan Times and The Economic Times. Users in New Delhi reported feeds that would not refresh and a mobile app that kept crashing.
Researchers noted that X's recovery appeared uneven across global regions. The platform's backend API may have remained stuck in a failure state in certain geographic nodes long after the core fix was deployed, per the Hindustan Times. X had not issued any formal statement or explanation as of the time of reporting.
Users Flee to Threads as Sunday Outage Raised Red Flags
For the third time this year, Meta's Threads saw a surge of users during an X outage. People used the hashtag #TwitterDown to swap notes and vent frustration, according to Forbes. The Independent reported users worried they had been hacked after being suddenly signed out of their accounts.
Sunday's prior outage — a smaller spike that lasted roughly one hour — had already put users on edge, per UNILAD. That back-to-back pattern, alongside X's reduced engineering staff since its 2022 acquisition, has fueled concern about the platform's long-term stability. X has not scheduled any public briefing on either disruption.
Sentiment
Mostly NegativePublishers
47
Articles
73
Reach
120
