Currently I'm trying to deploy some websites on AWS with EC2 instance. Unexpectly after one reboot,it can no longer get back. As I'm an all-the-time newbie I have to upgrade my package temporarily so to ask for help. Meh,basic plan cannot ask for tech support. They are giving really good advice on this and I'm sharing here so to help other people,in the hope it can save some of your pocket money.

1. Launch a temporary instance using same AMI in the same region and Availability Zone as your current. 
2. Detach the root volume from original instance,call it o-i and attach the volume to the temporary instance,the t-i
3. Login to the t-i and confirm the device is attached.
4. create a mount point and mount the volume that you attached,for example sudo mount /dev/xvdf /mnt
5. Correct the instance configuration,unmount the volume and detach it.
6. Attach it to the o-i,make sure that you specify the correct device name for the root volume 
7. Start your instance o-i
8. Terminated the t-i if no longer required.

My problem is with "root: /etc/dhcp/dhclient-enter-hooks.d/avahi-autoipd returned non-zero exit status 1",and the solution is to remove the relevant package on t-i and that's it. You should be able to get the o-i back to life if that's the same awful stuff you are looking around.

What a day~