1 Answer
- Newest
- Most votes
- Most comments
0
If reverting the instance type does not recover the system, there may be a problem that cannot be repaired automatically. Check the serial console for Error messages.
The following page may be helpful. https://repost.aws/knowledge-center/ec2-linux-status-check-failure-os-errors
Relevant content
- asked 3 years ago
AWS OFFICIALUpdated 2 years ago
If i reverting the instance type to initial type it works normally
I'm sorry, I misread that.
The primary response is the same as the previous one: if you can connect to EC2, connect and check the logs; if you cannot connect to EC2, check the logs from the serial console.
However, since the problem is recovered by changing the instance size, the cause is likely to be the difference in instance size, which I can assume to be a lack of CPU or memory capacity.
I think it's due to a lack of memory. If you're utilizing a Linux-based server like Amazon Linux, a clear indication of insufficient memory is the appearance of logs stating 'Out of memory'. This is a strong indicator that the instance is running out of memory.
In this case, you should review the settings of services that are started automatically to save memory or use t3a.medium as it is.
In the case of CPU, you can check the usage rate in the Monitoring tab (usually averaged every 5 minutes, so it may be less useful).