3

I created a new AWS EC2 instance (error occurs with AWS Linux & AWS Linux 2). And I can SSH in fine to start installing thing but every time after 5-10 minutes I can no longer type in the terminal and eventually get a Connection Reset message.

After this the SSH server as well as Apache no longer takes connections, and all the monitoring graphs on the EC2 console go down to 0. I've checked the SSHD logs and it just says that it closed the connection after receiving a Signal 15.

CPU Graph

Any idea what's happening here or what other logs I can check to see what's happening?

EDIT: Same things happens when I try to spin up an Ubuntu Server instance. Seems to be more of an EC2 problem in general vs any Linux distro.

Community
  • 1
LightDisk
  • 31
  • 4
  • 1
    You mention "start installing thing". Does this happen if you create a new instance, SSH in, and *do nothing*? – ceejayoz Jan 25 '18 at 22:51
  • Yep, just launched a new EC2 instance. SSH'd in and started a ping. Withing about 4 minutes I was kicked off and can't reconnect. – LightDisk Jan 26 '18 at 15:28
  • 1
    And these are the absolute standard, straight-from-Amazon AMIs, not a custom AMI *based* on Amazon's Linux? If so, seems like a "contact support" scenario - it's definitely not normal and I'd think if it were widespread we'd hear a million people complaining. – ceejayoz Jan 26 '18 at 15:43
  • Unfortunately I don't want to pay AWS for real technical support. I'm working on getting a post on the AWS forums, but apparently there's a delay between creating the forum account and being able to post. Thanks for the advice, I'll edit the post with the AWS forum link when I get it. – LightDisk Jan 26 '18 at 16:09
  • 2
    Are you on a corporate network that might be detecting your SSH connections and blocking them or something? The behavior you're describing, with standard AMIs and doing nothing to them, is *exceedingly* odd and I suspect isn't Amazon's fault. Have you tried connecting to these servers from another internet connection (home, coffee shop) to see if they're accessible there? – ceejayoz Jan 26 '18 at 16:29
  • Nope, just my standard home network. I have several other EC2s on my business and personal accounts that I can still connect to fine. The only thing I can think of that would be different is I'm just an IAM user on an associates account to help him set it up and configure it correctly right now, but I don't think that would cause it to randomly fail instead of getting some permission error. – LightDisk Jan 26 '18 at 17:04
  • 1
    I'd point to your ISP or your home router causing this issue before I'd point to Amazon, tbh. Maybe try disabling any firewalls on your router + any application firewalls you may have? This is really odd behavior. Do you see any changes on the AWS side? Security groups? – Bart Jan 26 '18 at 20:46

0 Answers0