By using AWS re:Post, you agree to the Terms of Use

Slowness within the AWS (EC2 and Workspaces) in Internet browsing itself, sometimes almost impossible to access a simple website.



We are experiencing browsing internet related slowness within the AWS (EC2 and Workspaces), browsing in simple websites itself is slow.

We had the same problem on 03/2022, we hired support business, but they didn't solve the problem and it went back to normal on its own.

Monitoring via CloudWatch, we noticed that the slowness is related to the NAT Gateway, as the documentation says the following:

"If the value of ConnectionEstablishedCount is less than the value of ConnectionAttemptCount, clients behind the NAT gateway tried to establish new connections for which there was no response."

And as you can see in the attached graphs, whenever there is a slowdown, the ConnectionEstablishedCount metric is lower than the ConnectionAttemptCount metric

Enter image description here

Internet browsing itself is slow, sometimes almost impossible to access a simple website, and this is directly impacting our internal use and customer service, as all our internal and support applications require the internet.

It is also worth mentioning that we migrated our entire On-Premise structure to AWS in October/2021, and this is the second time this has happened.

We even set up a new NAT Gateway in another AZ but it didn't work and it was still slow.

Today we have 6 EC2 servers and 20 Workspaces in the same VPC and using the same NAT Gateway, and at night I transfer backups from EC2 to S3.

Has anyone ever experienced this? Do you know what it could be and how to fix it?


1 Answer

Although this may not be related: Are you using a S3 Gateway Endpoint for your backups? If not, you should. It will mean that traffic from your instances to S3 will not have to go through NAT Gateway. This will reduce cost as well as reduce the amount of traffic through NAT Gateway that is going to S3.

Otherwise, at this distance it is difficult to determine what might be causing the traffic spike. I'd recommend enabling VPC Flow Logs to try and determine if there is something in the VPC which is causing the issue. You can use Athena to query the flow logs.

profile picture
answered a month ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions