Subsea cable repair website latency: how to test your site before Asia's EAC1 repairs begin

Subsea cable repair website latency is something you can measure before it affects customers. Taiwan has scheduled repairs on the EAC1 cable, and an early-October cut between the Philippines and Singapore showed how quickly rerouted traffic can slow overseas websites. Test your site from the regions that matter, record a baseline, and decide what to change while conditions are normal. This article was last updated on 6 November 2026.

What happened: EAC1 repairs scheduled after an early-October cable cut

A Taiwan Maritime Port Bureau Notice to Mariners, released and last verified on 4 November 2026, schedules repair operations for faults on East Asia Crossing 1 (EAC1), section C, off eastern Taiwan. The work runs from 10 November to 20 December, following an August fault about 150 kilometres off Taiwan's eastern coast that disrupted communications. SubTel Forum is explicit that a scheduled window is not evidence that repairs are finished or that service has returned.

This is a separate event from the October incident. In early October, a submarine cable cut between the Philippines and Singapore disrupted internet services across the Philippines, and Globe, PLDT and Converge rerouted traffic through alternative international links, as Newsbytes.ph reported. Globe said it had restored services, having rerouted traffic over the weekend after the disruption reduced available bandwidth and slowed speeds, though it expected fibre repairs to take several weeks. See BusinessWorld. The reports reviewed do not link the two events or name the cable involved in the October cut.

Why a cable repair matters for your website, domain and email

Your site can be perfectly healthy and still feel slow, because visitors reach it over international links you do not control. When traffic is moved off its usual route, data may travel farther or share capacity with more users, according to Rappler's report. Your uptime monitor, which probably checks from one location, may show green throughout.

The effects differ by service:

The route your visitors' ISP uses is not something you can fix. Where your server sits, and how much of your content is served from nearby caches, is something you can influence.

How to test website latency from Asia before repairs start

Test now to get a baseline. Without one, you cannot tell a repair-related slowdown from your site's normal behaviour.

1. Pick your real audience regions. Use your analytics to list the countries where customers actually are, for example Singapore, the Philippines, Hong Kong, Taiwan and Japan.
2. Run a multi-location test. Use any reputable online tool that tests from several cities, or a small VPS in each region. A single test from your office tells you little.
3. Measure the stages, not just total time. From a Linux or macOS shell in each location, run:

curl -o /dev/null -s -w "dns:%{time_namelookup} connect:%{time_connect} tls:%{time_appconnect} ttfb:%{time_starttransfer} total:%{time_total}\n" https://example.com/

4. Trace the path. Run mtr -rw example.com (or traceroute) to see where delay builds up. A sudden jump at an international hop points to the path, not your server.
5. Repeat at different times. Test during local business hours and in the evening, because congestion on rerouted paths often peaks at busy times.
6. Record the results in a spreadsheet with date, region and each timing.

How to read the numbers: a high connect value relative to a nearby location indicates distance or path problems. A high ttfb with a normal connect suggests slow application work, such as uncached WordPress pages, rather than the network.

Server location in Asia and latency: what you can and cannot change

Hosting location in Singapore affects latency for Southeast Asian visitors because the physical distance and number of international hops shrink. It is not a cure-all. If most of your customers are elsewhere, a regional server could make things worse for them. Decide using your analytics and your baseline, not on assumption.

Decision criteria

• Audience concentrated in one region: host close to it, and check whether the path from that region to your server crosses affected links.
• Audience spread across continents: keep one well-run origin and put a CDN or caching layer in front of static content.
• Dynamic, uncacheable pages (carts, dashboards): origin location matters most, so test it carefully.

The data centres and network latency test page shows the locations available and lets you check latency from your own connection. If your site outgrows shared resources while you move, our guide to shared vs VPS vs cloud vs dedicated hosting explains the trade-offs.

What to do now: a practical checklist

Reroute traffic cable outage planning is mostly about preparation you can do in an afternoon.

• Record baseline latency and TTFB from each key region (see the steps above).
• Set up monitoring from more than one location, with alerts on response time rather than only up/down.
• Turn on page caching, and compress and size images, so each page needs fewer round trips.
• Lower DNS TTLs a few days before any planned move, so a change takes effect quickly. See our guide to changing nameservers.
• Confirm your SSL certificate is valid and auto-renewing, because a failed TLS handshake over a slow path looks like an outage.
• Make sure you have a current backup and a tested restore before changing anything.
• Tell customers and staff in affected regions that slowness may be network-related, and give them a phone or email fallback.
• Check where your email is hosted and whether remote staff rely on webmail.

If testing shows your current location is a poor fit, free website migration can move the site, and WordPress hosting or a VPS suits sites that need more control over caching and location.

Background and what to watch next

Repairs depend on specialist ships and conditions at sea, so published windows can move. Taiwan's notice covers 10 November to 20 December, and the earlier Philippines–Singapore repairs were described as taking several weeks. Treat both as estimates.

What to watch: updated Notices to Mariners, advisories from your own visitors' ISPs, and your regional test results. If numbers worsen only at certain hours or from certain countries, that points to path congestion. If they worsen everywhere, look at your own server first. Our article on keeping your website and email online during a DNS outage covers the related failure mode.

Want to see how our locations perform from where your customers are? Run the data centres and network latency test and compare results with your baseline.

Frequently asked questions

Will the EAC1 repairs make my website slower?

The sources do not say. The notice only schedules the work and does not describe expected impact on any website. Test your own site from your key regions before, during and after the window, and compare against your baseline.

How do I tell whether slowness is my server or the network path?

Compare connect time and time to first byte. A high connect time from one region, with a normal result from another, points to the path. A high first-byte time with a normal connect time points to slow page generation on your server.

Should I move my site to Singapore because of a cable cut?

Not on the strength of one incident. Move only if your analytics show most visitors are in Southeast Asia and your tests show a clear gain. Otherwise, caching and a CDN usually help more than a relocation.

Can I stop traffic being rerouted?

No. Rerouting is done by the network operators between your visitors and your server. You can only reduce exposure by serving more content from nearby caches and by choosing a server location that suits your audience.

How often should I test latency?

Take a baseline now, then test again at different times of day, such as daily or weekly during the repair window. Automated multi-location monitoring is easier than manual checks.