intermittent failed boot for lucid ec2 image

Bug #494595 reported by Scott Moser on 2009-12-09
26
This bug affects 4 people
Affects Status Importance Assigned to Milestone
linux-ec2 (Ubuntu)
High
Scott Moser
Lucid
High
Scott Moser

Bug Description

I launched the 20091209 ec2 build in us-west-1 on i386 and the system did not come up. The system is not reachable via ssh internal or externally, and ping fails from other us-west-1 system on its internal IP.

console-output is attached, but not seemingly helpful. This bug was reported from another instance of the same AMI.

ProblemType: Bug
Architecture: i386
Date: Wed Dec 9 16:44:52 2009
DistroRelease: Ubuntu 10.04
Ec2AMI: ami-033f6e46
Ec2AMIManifest: ubuntu-images-testing-us-west-1/ubuntu-lucid-daily-i386-server-20091209.manifest.xml
Ec2AvailabilityZone: us-west-1b
Ec2InstanceType: m1.small
Ec2Kernel: aki-633c6d26
Package: linux-image-2.6.32-300-ec2 2.6.32-300.1
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: User Name 2.6.32-300.1-ec2
SourcePackage: linux-ec2
Tags: lucid ec2-images
Uname: Linux 2.6.32-300-ec2 i686

Scott Moser (smoser) wrote :
description: updated
Scott Moser (smoser) wrote :

Just for comparison, here is console output of good boot.
Note, the ureadahead message seems to just be a red herring as it appears in both.

Scott Moser (smoser) wrote :

Note: A 'ec2-reboot-instances' for the failed boot's instance id brought the system up.

tags: added: iso-testing
Chuck Short (zulcss) wrote :

I get this as well it only seems to affect us-west-1.

Chuck Short (zulcss) wrote :
Thierry Carrez (ttx) on 2009-12-11
Changed in linux-ec2 (Ubuntu Lucid):
status: New → Confirmed
importance: Undecided → High
Scott Moser (smoser) wrote :

The idea popped into my head that this is not kernel related. I think its more likely related to upstart and order of ssh and/or ec2-init running.

I plan on looking some more at this today.

Thierry Carrez (ttx) on 2009-12-18
Changed in linux-ec2 (Ubuntu Lucid):
assignee: nobody → smoser (scott-moser)
assignee: smoser (scott-moser) → nobody
assignee: nobody → Scott Moser (smoser)

Is there any solution? This problem makes my system unusable.

Scott Moser (smoser) wrote :

Adam, have you seen this on the alpha2 release ? In my ~ 100 boots in testing it, I never saw this issue.

I have lucid updated to current, and bootup stops just after udev. Some I could boot by booting into recovery mode and starting gdm manually, but now I do not get to recovery menu :(

(I keep it updated by chrooting from live-cd environment)

Dr. Tyrell (dr.tyrell) wrote :

This started happening to me last night after update. I also ride the bleeding edge Lucid repos.

Dr. Tyrell (dr.tyrell) wrote :

FYI,
I have normal Lucid x64, no ec2, and Athlon 64 II 4000+, Acer with IDE and PATA drives.

On Sun, 17 Jan 2010, Dr. Tyrell wrote:

> FYI,
> I have normal Lucid x64, no ec2, and Athlon 64 II 4000+, Acer with IDE and PATA drives.

Could you please open another bug for your boot issues?
This bug is really only to track the ec2 issues, which probably is not
related to what others are seeing. I'm sorry that the subject matched.

Scott Moser (smoser) wrote :

I've not seen this recently, as I noted above alpha2 booted over 100 times (total) across the 3 regions without a failure.
Marking it fix-released.

Changed in linux-ec2 (Ubuntu Lucid):
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers