AWS S3 CLI - Could not connect to the endpoint URL

Question

$ aws s3 ls

Could not connect to the endpoint URL: "https://s3.us-east-1a.amazonaws.com/"

What could be the problem?

This happened with me when trying to create a cloudsearch domain on `us-east-2`, I had to use `us-east-1`? — Jason Goemaat, Dec 11 '17 at 20:03

score 370 · Accepted Answer · answered Nov 03 '16 at 21:02

370

You probably have something wrong in your default profile for the default region.

Check your file at ~/.aws/config, you have something like

[default]
region=us-east-1a
...

Fix the region to region=us-east-1 and then the command will work correctly

answered Nov 03 '16 at 21:02

Frederic Henri

51,761
10
113
139

4

I had a similar wrong entry in ~/.aws/config which was created by 'aws configure' command. By default it suggested my region name 'Mumbai' which I accepted. This was the issue. Once I replaced it with ap-south-1, the command (aws s3 ls) started working. Thanks. – Anurag Apr 04 '18 at 05:11
6

I had a similar error, and found that it could also be solved by specifying the region as a command-line option: `aws s3 ls --region us-east-1`. – Kurt Peek Aug 29 '19 at 22:54
Does not only mean that, in my case Cloudformation was temporary down – Matteo Oct 27 '20 at 06:51

score 11 · Answer 2 · answered Apr 06 '18 at 08:03

11

first you use 'aws configure' then input the access key, and secret key, and the region. the region you input would be important for this problem. try to input something like 's3.us-east-1', not 's3.us-east-1a'. it will solve the issue.

answered Apr 06 '18 at 08:03

fraya

111
1
2

score 11 · Answer 3 · answered Dec 06 '18 at 22:32

11

If none of solutions given above works,also check your permissions and firewall settings. In my case adding proxy environment variables did the job.

For Linux or mac

$ export HTTP_PROXY=http://<YOUR PROXY IP>:<PORT>

$ export HTTPS_PROXY=http://<YOUR PROXY IP>:<PORT>

For Windows

set HTTP_PROXY=http://<YOUR PROXY IP>:<PORT>

set HTTPS_PROXY=http://<YOUR PROXY IP>:<PORT>

answered Dec 06 '18 at 22:32

imTheManager

219
3
5

this was my issue. One can test the proxy by running **curl** ```curl -I https://sts.amazonaws.com ``` – Andre Leon Rangel Aug 09 '22 at 06:16

score 7 · Answer 4 · answered Nov 14 '18 at 12:59

Some AWS services are just available in specific regions that do not match your actual region. If this is the case you can override the standard setting by adding the region to your actual cli command.

This might be a handy solution for people that do not want to change their default region in the config file. IF your general config file is not set: Please check the suggestions above.

In this example the region is forced to eu-west-1 (e.g. Ireland):

aws s3 ls --region=eu-west-1

Tested and used with aws workmail to delete users:

aws workmail delete-user --region=eu-west-1 --organization-id [org-id] --user-id [user-id]

I derived the idea from this thread and it works perfect for me - so I wanted to share it. Hope it helps!

score 6 · Answer 5 · answered Jan 04 '19 at 22:01

You should specify the region in your CLI script, rather than rely on default region specified using aws configure (as the current most popular answer asserts). Another answer alluded to that, but the syntax is wrong if you're using CLI via AWS Tools for Powershell.

This example forces region to us-west-2 (Northern California), PowerShell syntax:

aws s3 ls --region us-west-2

score 6 · Answer 6 · answered Jan 10 '19 at 19:05

6

Probably, there is something wrong with the default region while configuring aws. In your case, the URL says "https://s3.us-east-1a.amazonaws.com/"

In your command prompt,

aws configure, enter your keys, Now fix your region from us-east-1a to us-east-1.

Kindly check the syntax according to the CLI you are using. This will be helpful.

answered Jan 10 '19 at 19:05

Sonal

579
5
7

That wierd but it helped me. My region was us-east-2c. I removed the c from 2c and it resolved. – Ganesh S Dec 08 '21 at 10:56

score 3 · Answer 7 · answered Mar 29 '21 at 16:05

This worked for me.

Use debug option to get clear idea of bug

aws ec2 describe-instances --instance-ids (myid) --region ap-south-1 --debug

I got following issue.

EndpointConnectionError: Could not connect to the endpoint URL: "https://ec2.ap-south-1b.amazonaws.com/"

I tried to Ping but it was not working

ping ec2.ap-south-1b.amazonaws.com ping: ec2.ap-south-1b.amazonaws.com: Name or service not known

I checked for invalid value for Region when using aws configure

but it was configured properly

[default] region = ap-south-1

Then , I went to **IAM -> Users - > (yourusername) - > Add Permissions

and I added "AdministratorAccess" Policy**.

After that everything worked for me.

MainThread - urllib3.connectionpool - DEBUG - Starting new HTTPS connection (1): ec2.ap-south-1.amazonaws.com:443 MainThread - urllib3.connectionpool - DEBUG - https://ec2.ap-south-1.amazonaws.com:443 "POST / HTTP/1.1" 200 7176

score 2 · Answer 8 · answered Nov 07 '19 at 13:58

2

You should do the following on the CLI : 1. aws configure'
2. input the access key 3. input secret key 4. and then the region i.e : eu-west-1 (leave the a or b after the 1)

answered Nov 07 '19 at 13:58

adam ali

21
2

score 2 · Answer 9 · edited Feb 09 '21 at 12:29

2

In case it is not working in your default region, try providing a region close to you. This worked for me:

   PS C:\Users\shrig> aws  configure
   AWS Access Key ID [****************C]:**strong text**
   AWS Secret Access Key [****************WD]:
   Default region name [us-east1]: ap-south-1
   Default output format [text]:

edited Feb 09 '21 at 12:29

Benjamin Zach

1,452
2
18
38

answered Feb 09 '21 at 10:28

Shrikant Gawande

21
2

score 2 · Answer 10 · answered Mar 28 '22 at 09:35

2

Weirdly, in my case restarting my laptop has helped. I've been using VPN lately, I think it messed up some network settings causing this error.

answered Mar 28 '22 at 09:35

Zhenya

6,020
6
34
42

score 1 · Answer 11 · answered May 31 '19 at 18:41

Assuming that your profile in ~/aws/config is using the region (instead of AZ as per your original question); the other cause is your client's inability to connect to s3.us-east-1.amazonaws.com. In my case, I was unable to resolve that DNS name due to an error in my network configuration. Fixing the DNS issue solved my problem.

score 1 · Answer 12 · answered Dec 27 '20 at 06:47

Check the .aws directory under home directory. Windows: C:\Users<home-name>.aws Linux: ~/.aws
Under this directory, you will find the config as well as credentials file. It will have the information from the aws configure that you may have run before. IF not, then
Run aws configure Enter the access key - secret key - enter secret key region - (ap-southeast-1 or us-east-1 or any other regions) format - (json or leave it blank, it will pick up default values you may simply hit enter)
From the Step 2, you should see the config file, open it, it should have the region. Please ensure there is region specified.
You may now run the following command to list the buckets aws s3 ls It should work fine.

score 1 · Answer 13 · answered May 09 '22 at 04:00

1

Check your environment settings:

set |& grep REGION

I had a typo in my AWS_REGION and AWS_DEFAULT_REGION environment variables, so it was trying to access a non-existant region.

answered May 09 '22 at 04:00

Under4Mhz

73
8

score 0 · Answer 14 · answered Sep 13 '19 at 18:53

Couple things I've done to fix this :

Updated my CLI and it given this error (previous error was "aws connection aborted error 10013")
Tried to nslookup aws s3 endpoing : nslookup s3.us-east-2.amazonaws.com

DNS request timed out. timeout was 2 seconds. Server: UnKnown Address: 192.168.10.1

-> hmmm very weird

Went to windows network troubleshooting and selected to test access to specific page. It informed that Windows firewall blocked the connection. Fixed this
Received a new error , after fixing the request through firewal :

An error occurred (RequestTimeTooSkewed) when calling the ListBuckets operation: The difference between the request time and the current time is too large.
Updated my date & time to automatic -> Fixed

score 0 · Answer 15 · edited Jan 20 '20 at 10:34

Everyone has different defaults, and interestingly it will change after time. As an example, first I was on global, and then after 15 minutes it shows Ohio (which is us-east-2).

The best approach is to check it during your work -- in console of your AWS working area, just set it on the right above side near your name on top bar check your region name and click on the down arrow to see your region.

In AWS CLI type aws configure or aws2 configure, give your access and secret id, then during default region, write your region and press Enter.

You will definitely get access to specific region set and it will work.

Nicole Douglas · Answer 16 · 2020-07-06T12:54:37.423

0

The solution to my problem was to run:

    sudo aws configure

Enter your credentials and then run:

    sudo aws s3 ls

A different solution was to make sure that the region in the .aws/config file is the same as the endpoints

edited Jul 06 '20 at 12:54

answered Apr 23 '20 at 19:05

Nicole Douglas

579
4
14

score 0 · Answer 17 · answered Aug 12 '20 at 16:33

0

On Windows run the aws configure command again and reset region without the letter for example if your region is us-east-1a you need to change it to us-east-1 in order for it to work.

answered Aug 12 '20 at 16:33

techjunkie

171
1
4

score 0 · Answer 18 · answered Jan 29 '23 at 14:30

0

If you are using ec2 just:

export AWS_DEFAULT_REGION=your-region

answered Jan 29 '23 at 14:30

Yasin Türk

11
3

score 0 · Answer 19 · answered Mar 01 '23 at 13:30

0

If you are sure that you did everything correctly, If you are using a Company PC, disconnect VPN connection.

This worked for me.

answered Mar 01 '23 at 13:30

Ali UÇAR

1

score 0 · Answer 20 · answered Mar 15 '23 at 05:46

0

Check VM security group rules and add outbound rule for S3 access .

answered Mar 15 '23 at 05:46

Muthukumar M

1

AWS S3 CLI - Could not connect to the endpoint URL

20 Answers20

Linked