AWS Support is a Complete Disaster – 65 Days of Lies and No Fix

Question

The Problem:
Before December 5, 2024, AWS Global Accelerator (GA) and CloudFront had a stable 40ms latency.
Since then, latency has skyrocketed to 150ms+, and traffic has been randomly rerouted through Europe instead of the expected region.
AWS first claimed it was due to an "internal evaluation." Then they changed the story and said it was an "infrastructure issue." Later, they blamed a subsea cable outage, but no confirmed fiber cut affecting the region has been reported publicly.
After weeks of excuses, AWS GA and CloudFront were supposedly fixed, but latency remains extremely high.
Zero transparency – AWS keeps changing the story with no clear root cause.
Zero accountability – Support keeps saying they "escalated" the issue, but no real updates or solutions are provided.
No ETA – Even after 65 days, AWS cannot say when this will be fixed and continues to blame third parties.
Poor routing decisions – Instead of fixing the problem, AWS has changed routing multiple times, but latency remains bad.

re:Post Agent · Answer

I apologize, but I cannot confirm or comment on the specific claims made about AWS Support or the alleged issues with Global Accelerator and CloudFront performance. However, I can provide some general information about how these services typically work and steps that can be taken when experiencing performance issues:

AWS Global Accelerator and CloudFront are designed to improve application availability and performance by routing traffic through AWS's global network infrastructure. Under normal circumstances, they should provide low-latency connections to the nearest point of presence.

If you're experiencing persistent high latency or unexpected routing, here are some recommended troubleshooting steps:

1. Verify the issue is consistent across different client locations and networks.

2. Check AWS service health dashboards and announcements for any known issues in your region.

3. Review your Global Accelerator and CloudFront configurations to ensure they are optimized for your use case.

4. Enable detailed logging and analyze the logs to identify patterns or anomalies.

5. Test connectivity directly to your origin to isolate whether the issue is with AWS services or elsewhere in your application stack.

6. If the problem persists, engage with AWS Support and provide detailed information including traceroutes, timestamps, and specific endpoint details to help diagnose the issue.

7. Consider using additional tools like Route 53 latency-based routing as a potential workaround if specific points of presence are underperforming.

While AWS strives to provide high-quality support and maintain service performance, complex networking issues can sometimes take time to fully diagnose and resolve, especially if they involve multiple parties or infrastructure components. Continued engagement with AWS Support, providing clear and detailed information about the observed behavior, is usually the best path to resolution for persistent issues.  
**Sources**  
[AWS global accelerator suddenly got high ping | AWS re:Post](https://repost.aws/questions/QU86LwUV4-THi3d5fA7rV-3g/aws-global-accelerator-suddenly-got-high-ping)  
[aws global accelerator Have high Ping And middel east Too ( Bahrain) | AWS re:Post](https://repost.aws/questions/QUIn3d7iqAQhu5SqPN7LLNLQ/aws-global-accelerator-have-high-ping-and-middel-east-too-bahrain)  
[aws globalaccelerator suddenly got high ping | AWS re:Post](https://repost.aws/questions/QU3hoznbb8Rs2L9ULslYo5-w/aws-globalaccelerator-suddenly-got-high-ping)  
[CloudFront Slow in One Region | AWS re:Post](https://repost.aws/questions/QUdjHPM1PaS7ebVX_Izk1wXg/cloudfront-slow-in-one-region)

tapple (AWS) · Answer

Sorry to hear about this. Do you have any pre- and post-incident traceroutes that you can share?

rePost-User-8635741 · Answer

![![Enter image description here](/media/postImages/original/IMNBswfYzkSv2aS8fZIArvEw)
Enter image description here](/media/postImages/original/IMCtHIAHX5S7-oGXNSB2yPPw)
as you can see

rePost-User-8635741 · Answer

![Enter image description here](/media/postImages/original/IMUj_a6xfeSmK2j0aYcdRsqw)
as you can see here too

AWS Support is a Complete Disaster – 65 Days of Lies and No Fix

Add your answer

Relevant content