some times ,when we create a new EMI and launch a instance from this EMI, the instance is in RUNNING status,but could not be used. even could not be ping.
use euca-get-console-output to check the information will get follow logs:
Gave up waiting for root device. Common problems:
- Boot args (cat /proc/cmdline)
- Check rootdelay= (did the system wait long enough?)
- Check root= (did the system wait for the right device?)
- Missing modules (cat /proc/modules; ls /dev)
ALERT! /dev/sda1 does not exist. Dropping to a shell!
BusyBox v1.10.2 (Ubuntu 1:1.10.2-2ubuntu7) built-in shell (ash)
Enter 'help' for a list of built-in commands.
could you help why we always meet that issue?And how to do?
yes, our EMI is base on images eucalyptus provide.
And it can not be reproduce all the time but sometimes happened.
for example, we had use a instance A to create a EMI A, and when we use EMI A to lunch a new instance ,we meet the issue i said before, and if we use the same instance A again to create a new EMI B, and use EMI B to lunch a instance,It will be worked. so i think it may be not the instance's issue.
during create EMI ,the controllor and node machine is normal and not in busy status.
Hello,
do you have a single NC? I'm not sure why it would fail occasionally: it doesn't make sense. Either the image is misconfigured for your system and it fails every time, or it works every time. That error, as I mentioned, means that it cannot mount of find the root file system: you can check in the kernel log before that error if the image sees sda with its partitions. Otherwise you could try to inspect the image created by eucalyptus., and try to start by hand to see if you can re-create and thus debug the problem.
keep us posted on your progress.
cheers
graziano