This post is more than 5 years old
5 Practitioner
•
274.2K Posts
0
5534
Unable to browse buckets on ECS from S3 browser
Hi all,
We've recently installed successfully ECS 3.1 for a customer, in which we need to perform an integration with DataDomain through CloudBoost.
After installing ECS and performing the initial configuration steps, the only missing piece to fully complete the installation is the browsing through the S3 browser and the subsequent integration with DataDomain.
As for Load Balancing, HA Proxy was installed following the steps provided in the White Paper. I found it a little confusing about the creation of a self-signed certificate in which it omits a few pieces of info that might be useful.
Whenever I try to browse by using the S3 browser, a different set of errors appear. I've tried by putting the name/credentials for the Load Balancer and I've also tried to do so directly towards the ECS. A few screenshots of the so-mentioned errors are attached.
Any thoughts on what might be causing this? Where can I look to further investigate? Is this something that support might be involved in considering the participation of a 3rd party product like HA Proxy?
Any thoughts or comments will be more than appreciated.
This is the object test user that has been created:
These are the inputs for the S3 browser (putting the info for the ECS array):
And this is the error message when trying to browse putting the :
I don't happen to have any screenshots for the "direct" approach, but it's basically a "unable to browse buckets" message.
benschumacher
75 Posts
0
January 29th, 2018 08:00
Can you have a look at this article: https://community.emc.com/docs/DOC-59224
The article above is going directly to an ECS node. You will want to try that first before attempting to use HAProxy. What type of ECS installation is this? Is this an actual ECS appliance or Community Edition? If following the article still doesn't work, can you run a simple curl command for me and copy/paste the results (example below)?
ben@USENSCHUMBL4C:~$ curl http://10.1.83.51:9020?ping -v
* Rebuilt URL to: http://10.1.83.51:9020/?ping
* Hostname was NOT found in DNS cache
* Trying 10.1.83.51...
* Connected to 10.1.83.51 (10.1.83.51) port 9020 (#0)
> GET /?ping HTTP/1.1
> User-Agent: curl/7.35.0
> Host: 10.1.83.51:9020
> Accept: */*
>
< HTTP/1.1 200 OK
< Date: Mon, 29 Jan 2018 16:17:28 GMT
< Content-Type: application/xml
< Content-Length: 290
<
* Connection #0 to host 10.1.83.51 left intact
http://s3.amazonaws.com/doc/2006-03-01/ "> LOAD_FACTOR 1 MAINTENANCE_MODE OFF
Data Node is Available
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
February 26th, 2018 07:00
Hi Ben,
Thanks a lot for your time and your answer. I ended up running some tests with Cyberduck and after testing those steps, we were able to log directly towards the ECS.
We will need to do the same with the LB this time, although we are not succeeding in logging in, so we'll need to check the LB (this runs with HA Proxy) configurations.
I was told that the customer will need to also register the newly-created ObjectBaseURL onto their DNS entries, but I'm not quite sure on which IP address shall be addressed to that record.
Any piece of advice will be most appreciated.
Thanks again for your help and guidance.