4

I have a WordPress install on Beanstalk with EC2 and RDS. I'm very new to AWS. Yesterday the site became unresponsive and had a status of 'No data received from the instance'. When I try to request full logs it times out. I can't FTP or SSH into the instance because it also times out. The only thing I view is the System Log but I didn't see anything that jumped out at me.

Looking at Events, it seems the problem started with 4xx errors that turned into 5xx errors.

Here are the entries:

2019-02-14 12:10:55 UTC-0500    WARN    Environment health has transitioned from Ok to Severe. None of the instances are sending data.  
2019-02-12 22:54:41 UTC-0500    INFO    Environment health has transitioned from Warning to Ok.  
2019-02-12 22:53:41 UTC-0500    WARN    Environment health has transitioned from Ok to Warning. 100.0 % of the requests are failing with HTTP 5xx.  
2019-02-12 22:45:42 UTC-0500    INFO    Environment health has transitioned from Warning to Ok.  
2019-02-12 22:44:42 UTC-0500    WARN    Environment health has transitioned from Ok to Warning. 100.0 % of the requests are failing with HTTP 5xx.  
2019-02-12 14:21:19 UTC-0500    INFO    Environment health has transitioned from Severe to Ok.  
2019-02-12 14:20:19 UTC-0500    WARN    Environment health has transitioned from Ok to Severe. None of the instances are sending data.  
2019-02-12 10:31:36 UTC-0500    INFO    Environment health has transitioned from Severe to Ok.  
2019-02-12 10:30:36 UTC-0500    WARN    Environment health has transitioned from Ok to Severe. 97.2 % of the requests are erroring with HTTP 4xx.

I can't figure out what is the underlying problem. Any help or guidance would be appreciated.

Thomas
  • 4,225
  • 5
  • 23
  • 28
mdailey77
  • 141
  • 1
  • 4
  • Given I couldn't find a solution, I rebooted the server. It was a last resort because the server reverts back to the content that was there in the beginning. I think it might have been a memory issue but I can't be sure. I was able to view a screenshot and one of the last entries had 'kill process (httpd)..." – mdailey77 Feb 16 '19 at 18:06
  • A good question to ask is: Why does a reboot of your server revert content in WordPress? This reason may be off topic to the question, but still that shouldn't really be a concern. Ideally a production server can be rebooted without any negative consequences, except maybe unavailability. You have found the [EC2 Screenshot](https://aws.amazon.com/blogs/aws/ec2-instance-console-screenshot/) functionality -- this is very useful to diagnose an unresponsive instance. Finally, you can probably identify what caused the crash by investigating your logs in /var/log/messages. – Aaron St. Clair Feb 19 '19 at 21:06
  • I should have explained it better. I meant rebooting the server reverts it back to the state when I did my first deploy. My first deploy was from a wordpress git repo. At the time of the issue I couldn't get to the logs. – mdailey77 Feb 20 '19 at 15:11

1 Answers1

3

I experienced similar issues like not being able to SSH, but also not being able to see the logs in the beanstalk console. Problem was a memory shortage due to my t2.mini instance. Change the instance type to t2.small or even to t2.medium and the problem should be solved (unless your memory shortage is caused by some coding errors)

Greg Holst
  • 131
  • 2