Verizon Outage Dented Mobile Traffic, Not Enough to Fully Cut Data
Verizon mobile customers across multiple U.S. cities reported losing connectivity on Monday, September 30, 2024. Phones in affected areas displayed “SOS” instead of the standard signal indicator, and users said they couldn’t make or receive calls. Verizon’s mobile network operates as autonomous system AS6167 (CELLCO), which gave Cloudflare a useful vantage point for gauging when network problems actually started showing up in data traffic.
Though reports of trouble began around 09:00 ET (13:00 UTC), Cloudflare didn’t see a clear signal in HTTP request volume from the ASN until roughly two hours later. The first sign of trouble, a dip of about 5 percent compared to the previous week’s traffic at the same time, showed up around 11:00 ET (15:00 UTC). Request volume bottomed out at 13:45 ET (17:45 UTC), falling about 9 percent below expected levels. Verizon publicly acknowledged the issue just before 12:00 ET (16:00 UTC), saying engineers were working to identify and fix the problem.
Restoration began to show up in the data in the late afternoon. Verizon posted an update just after 17:00 ET (21:00 UTC) saying service had started to return, and request volumes crossed back over the prior week’s levels at 17:15 ET (21:15 UTC). By 19:18 ET (23:18 UTC), Verizon said the disruption had been fully resolved.
Looking at individual cities, the outage’s impact was uneven. Many metropolitan areas reported by the media as hardest hit showed the same general pattern as the national data. Los Angeles, Seattle, Omaha, Denver, Phoenix, Minneapolis, Indianapolis, and Chicago all showed declining request traffic from around 11:00 ET (15:00 UTC). Omaha stood out as the biggest loser, with traffic down about 30 percent relative to the prior week, even though it was already tracking lower heading into the outage. Other affected cities saw declines in the 10-20% range.
Not all cities followed the downward trend. Las Vegas traffic showed some volatility around the outage’s start but stayed close to prior-week levels, and it began exceeding them at 16:00 ET (20:00 UTC). Cincinnati was tracking slightly above last week’s volume before the outage and continued to do so throughout the disruption. New York and Atlanta both showed week-over-week increases during the outage window, but those cities were already running ahead of the prior week’s pace going into the event.
Based on the data, voice service on Verizon’s network appears to have taken a harder hit than data service. Individual city graphs showed declines in request traffic, but no location experienced a full outage, which suggests customers with data connectivity were still able to send and receive HTTP requests even as call functionality was impaired. Verizon had not disclosed a root cause for the network issues as of 19:15 ET (23:15 UTC).
