I think you mean region and not availability-zone since availability
zone names mean different things for different AWS accounts.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
Blueprint changed by Eric Hammond:
Whiteboard changed:
Blueprint:https://blueprints.launchpad.net/ubuntu/+spec/servercloud-q-awstools
Etherpad: http://pad.ubuntu.com/uds-q-servercloud-awstools
-
- content below copied to Etherpad 2012-04-30, please make updates to
- etherpad
Blueprint changed by Eric Hammond:
Whiteboard changed:
- Ubuntu has packages for only two sets of AWS tools:
+ Ubuntu has packages for some AWS tools:
* EC2 API Tools [package: ec2-api-tools]
* EC2 AMI Tools [package: ec2-ami-tools]
* RDS - Relational Database Service [package
Scott:
- With the CNAME solution, the requests still go to the internal IP
address for standard EC2 instances.
- I don't imagine that many non-EC2 people would try to configure their
Ubuntu systems to use the EC2 repositories.
- Canonical would get charged the same network fees for people
+1 for cloudcontrol's recommendation to use CNAMEs. I've been
recommending this to Canonical since we were discussing the initial
setup of EC2 dedicated repositories. It would have avoided a couple
issues that have happened since and would help prevent future problems
as AWS releases new
Though I don't like the current way Oneiric manages /etc/hosts (and
submitted related bug #890501) I agree with Scott that it is how Oneiric
works on EC2 and changes could cause existing installations to break.
In fact, I have automated system code that works around the bug which
would break if
Public bug reported:
In bug #892554, Kees Cook (kees) makes a great suggestion that cloud-
init could output the public ssh host keys to the console output. This
could then be read by automated software outside of the instance and
added to a known_hosts file using the IP address and/or hostname
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
https://bugs.launchpad.net/bugs/893400
Title:
cloud-init: Output machine usable public ssh host key (for
known_hosts)
To manage notifications about this bug go
Love the known_hosts suitable output format idea! It doesn't even need
to be an option. Split that off into bug #893400.
The current ticket can remain for adding the ECDSA ssh key fingerprint.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is
I've ammended the original example to use cat instead of cut as it
looks like the specific number of fields in the key may vary for some
older formats (rsa1) and it removes the objection that I invented
anything. I had been hoping to exclude the comment field, but agree
it's not worth the
** Description changed:
In bug #892554, Kees Cook (kees) makes a great suggestion that cloud-
init could output the public ssh host keys to the console output. This
could then be read by automated software outside of the instance and
added to a known_hosts file using the IP address
** Description changed:
cloud-init sets up /etc/hosts with a default value for 127.0.1.1 looking
something like:
- 127.0.1.1 ip-10-202-61-233.ec2.internal ip-10-202-61-233
+ 127.0.1.1 ip-10-202-61-233.ec2.internal ip-10-202-61-233
I edit /etc/hosts to change this
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
https://bugs.launchpad.net/bugs/890501
Title:
EC2 cloud-init overwrites 127.0.1.1 in /etc/hosts on every reboot
To manage notifications about this bug go to:
Public bug reported:
cloud-init sets up /etc/hosts with a default value for 127.0.1.1 looking
something like:
127.0.1.1 ip-10-202-61-233.ec2.internal ip-10-202-61-233
I edit /etc/hosts to change this value to something that makes more
sense to my internal software (e.g., Apache), say:
Scott: Learn something new every day. I will give this a try.
** Changed in: cloud-init (Ubuntu)
Status: New = Incomplete
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
** Changed in: cloud-init (Ubuntu)
Status: Incomplete = Invalid
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
https://bugs.launchpad.net/bugs/890501
Title:
EC2 cloud-init overwrites 127.0.1.1 in
Scott: Your suggested approach works for me even though I can't find any
documentation on how this is supposed to work. Everything I read says
that 127.0.1.1 should be the canonical hostname of the instance. Do you
happen to have any pointers to the information you are alluding to with
IPv6 and
Public bug reported:
When I start a standard Ubuntu 11.10 Oneiric instance (ami-a7f539ce) on
Amazon EC2 I can ssh in just fine with the default .ssh/authorized_keys
file.
If I prefix the ssh key in .authorized_keys with the option no-pty
then attempts to ssh fail with the error:
PTY
This may also be a problem on non-EC2 Oneiric; I have no way of testing
that.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to openssh in Ubuntu.
https://bugs.launchpad.net/bugs/876168
Title:
EC2 Oneiric ssh no-pty triggers PTY
Closing as invalid. I wasn't testing correctly with a non-pty ssh.
** Changed in: openssh (Ubuntu)
Status: New = Invalid
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to openssh in Ubuntu.
https://bugs.launchpad.net/bugs/876168
Sorry, I should have been clear in the original bug report that I was
submitting it on behalf of Amazon and another customer and did not
experience it myself on that particular instance or AMI.
Also, I'm not sure that lack of a public IP address as described in
#615545 is sufficient to determine
Amazon recommends fixing this through DNS instead of through software on
the instance.
Instead of resolving eu-west-1.ec2.archive.ubuntu.com directly to an A
record of the internal IP address starting with 10., Canonical should
change it to resolve to a CNAME of the external elastic IP address
Public bug reported:
DNS names like eu-west-1.ec2.archive.ubuntu.com (apt repository for eu-
west-1 on EC2) are currently resolving to private IP addresses (e.g.,
10.).
An EC2 instance running in VPC cannot access these repositories.
More details and possible fixes at:
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
https://bugs.launchpad.net/bugs/824947
Title:
EC2 apt repository DNS resolution on VPC instances
To manage notifications about this bug go to:
Public bug reported:
Binary package hint: cloud-init
The EC2/UEC images currently allow users to ssh to root@ but do not give
the user shell access. Instead, a helpful message is displayed to the
user explaining they need to ssh to ubuntu@ and they are disconnected
after 10 seconds.
This is
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
https://bugs.launchpad.net/bugs/798505
Title:
Tighten permissions on root@ ssh with EC2/UEC images
To manage notifications about this bug go to:
In my experience in 2008-2009, just because networking is up on an EC2
instance does not mean that 169.254.169.254 is going to accept
connections and requests for meta-data and user-data. You need to wait
for this to become available. I had code to do this in Ubuntu AMIs I
built back then.
On second glance, cloud-init may actually be retrying the connection and
the problem is that sometimes it takes longer than the current number of
retries.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cloud-init in Ubuntu.
I would like to see a solution like the following for this and other
files generated at startup (e.g., apt sources):
If the user has not modified the file since it was created from a
template, then the system should feel free to continue maintaining that
file by regenerating it from the template
*** This bug is a duplicate of bug 651370 ***
https://bugs.launchpad.net/bugs/651370
Confirmed with my AWS account using same AMI in eu-west-1 and
m2.2xlarge. Console log attached.
** Changed in: ec2-api-tools (Ubuntu)
Status: New = Confirmed
** Attachment added: console.log
*** This bug is a duplicate of bug 651370 ***
https://bugs.launchpad.net/bugs/651370
** This bug has been marked a duplicate of bug 651370
ec2 kernel crash invalid opcode [#1]
* You can subscribe to bug 651370 by following this link:
Public bug reported:
Binary package hint: cloud-init
Attempting to log in to the root user on an Ubuntu AMI returns the
message:
Please login as the ubuntu user rather than root user.
This has two minor problems:
1. There is a missing the before the word root.
2. Adding quotes around the
EBS volumes and snapshots create this sort of problem regularly for me,
generally with conflicting duplicate XFS UIDs which I have to override.
The best solution for EC2 AMIs may be to always accept that /dev/sda1 is
the boot disk.
--
attaching a volume to maverick instance may boot off it
*** This bug is a duplicate of bug 649742 ***
https://bugs.launchpad.net/bugs/649742
This is simply a software version issue. The ec2-describe-tags is in
the ec2-api-tools version 1.3-57419, but that version is not available
on Ubuntu 10.04 at this time.
The latest version on Ubuntu 10.04
This is especially important for remotely controlled servers which have
no console access (e.g., Amazon EC2).
--
ssh server doesn't start when irrelevant filesystems are not available
https://bugs.launchpad.net/bugs/583542
You received this bug notification because you are a member of Ubuntu
Scott: Is this fixed in Karmic or just Lucid?
--
ec2-init overwrites user/vmbuilder provided /etc/apt/sources.list
https://bugs.launchpad.net/bugs/502490
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to vm-builder in ubuntu.
--
Scott: Thanks for checking. The original AMI no longer exists, and I am also
unable to reproduce this on the latest Lucid daily:
ami-e9b95680
ubuntu-images-testing-us/ubuntu-lucid-daily-i386-server-20100301.manifest.xml
I have marked the bug invalid.
** Changed in: euca2ools (Ubuntu)
Public bug reported:
Binary package hint: euca2ools
On a fresh copy of the latest 32-bit Ubuntu Lucid daily AMI
ami-6a28c503
ubuntu-images-testing-us/ubuntu-lucid-daily-i386-desktop-20100110.manifest.xml
I installed a copy of ec2-ami-tools (to get the EC2 certificate)
echo deb
** Attachment added: Dependencies.txt
http://launchpadlibrarian.net/37676678/Dependencies.txt
** Description changed:
Binary package hint: euca2ools
On a fresh copy of the latest 32-bit Ubuntu Lucid daily AMI
- ami-6a28c503
-
I tried to test the fix, but ran into bug 505482.
--
euca2ools: euca-bundle-vol strips leading zero (0) from user id
https://bugs.launchpad.net/bugs/479823
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to euca2ools in ubuntu.
--
For reference, this article enumerates the steps I took to run
vmbuilder:
http://alestic.com/2010/01/vmbuilder-ebs-boot-ami
--
Too much image copying when using vmbuilder for EC2
https://bugs.launchpad.net/bugs/502495
You received this bug notification because you are a member of Ubuntu
Public bug reported:
I'm using the latest lp:vmbuilder source to build EBS boot AMIs for EC2.
I noticed that there is an awful lot of image file copying being done
which seems unnecessary when using vmbuilder for this purpose.
1. vmbuilder installs the image files into a temporary root
Public bug reported:
I'm using vmbuilder on 32-bit and 64-bit EC2 instances with identical
arguments except for --arch, --kernel, and --ramdisk. The 32-bit EC2
images are created fine, but the 64-bit images are practically empty
except for a few /var directories.
I have tracked this down to the
** Attachment added: Dependencies.txt
http://launchpadlibrarian.net/36525905/Dependencies.txt
--
vmbuilder: Creates mostly empty EC2 images on 64-bit
https://bugs.launchpad.net/bugs/493510
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed
No sooner entered as a bug than I think I've tracked this down to an
invalid part file I was using:
root 10240 a1
root 2048 a1
/mnt 1 b
Note the extra root entry.
I don't know if this can be used to create validation code to prevent
other people from making this dumb mistake, but I'm
Set to Confirmed since Scott confirmed the bug. Setting to Medium
since this seems like a severe impact on a non-core application.
** Changed in: vm-builder (Ubuntu)
Status: New = Confirmed
** Changed in: vm-builder (Ubuntu)
Importance: Undecided = Medium
--
vmbuilder: Deletes most
Public bug reported:
I'm running vmbuilder on the latest EC2 Karmic AMI. When I Ctrl-C
interrupt the run, it deletes almost all of the devices in /dev making
the system somewhat unusable.
Here's a log demonstrating this. I have reproduced the problem on a
fresh AMI by interrupting it in about
** Attachment added: Dependencies.txt
http://launchpadlibrarian.net/36488152/Dependencies.txt
--
vmbuilder: Deletes most of /dev on interrupt
https://bugs.launchpad.net/bugs/493020
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to
On IRC, smoser noted that he had also seen this behavior on nectarine.
--
vmbuilder: Deletes most of /dev on interrupt
https://bugs.launchpad.net/bugs/493020
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to vm-builder in ubuntu.
--
In case it matters, I ran ubuntu-bug on the Karmic instance right before
I ran the vmbuilder command so that all of the information in this bug
would be captured from the system. Attempts to run ubuntu-bug fail
after /dev is cleared out.
--
vmbuilder: Deletes most of /dev on interrupt
I agree that given the goals of euca2ools, the default installation of
the upstream source should not force AWS-specific options on users. The
primary goal of this ticket is to have euca2ools on Ubuntu EC2 AMIs
default to AWS/EC2 resources so that when a user runs an Ubuntu AMI,
euca2ools commands
I'm removing this partly incorrect statement in my original bug report:
- euca-register requires setting $EC2_ACCESS_KEY and $EC2_SECRET_KEY
instead of also checking $AWS_ACCESS_KEY_ID and
$AWS_SECRET_ACCESS_KEY_ID as ec2-register does.
euca2ools is based on the access key and secret key, while
Public bug reported:
This problem happens using vmbuilder on trunk.
When I specify --ec2-bundle (see entire command below) and I do not
specify --ec2-kernel and --ec2-ramdisk, then the following error is
generated:
2009-11-23 04:32:58,233 DEBUG : Oh, dear, an exception occurred
2009-11-23
** Attachment added: Dependencies.txt
http://launchpadlibrarian.net/35962337/Dependencies.txt
--
vmbuilder: --ec2-bundle triggers AttributeError: 'Ubuntu' object has no
attribute 'suite'
https://bugs.launchpad.net/bugs/486956
You received this bug notification because you are a member of
Public bug reported:
This problem happens with the vmbuilder on trunk.
When running vmbuilder I passed in an option like
--execscript setup-karmic-server
where setup-karmic-server was an executable file in my current
directory.
vmbuilder output the following messages which give a warm
** Attachment added: Dependencies.txt
http://launchpadlibrarian.net/35962327/Dependencies.txt
--
vmbuilder: --execscript in current directory is not found
https://bugs.launchpad.net/bugs/486955
You received this bug notification because you are a member of Ubuntu
Server Team, which is
Public bug reported:
Binary package hint: euca2ools
As I understand it, one of the goals in Ubuntu is for euca2ools to be a
drop in replacement for Amazon's EC2 AMI (and API) command line tools.
The following are differences where I noticed that I had to set
additional environment variables
** Attachment added: Dependencies.txt
http://launchpadlibrarian.net/35902309/Dependencies.txt
--
euca2ools: Requires more environment variables than EC2 AMI tools
https://bugs.launchpad.net/bugs/486128
You received this bug notification because you are a member of Ubuntu
Server Team, which
Thierry: This is the first time I've typed any euca2ools commands, so
I'm not sure where else there might be a problem. I don't think may
commands would require the user id anyway.
Let me know if there are any specific ones you'd like me to test.
--
euca2ools: euca-bundle-vol strips leading
Torsten: What AMI id are you starting with?
--
ec2-bundle-vol and ec2-upload-bundle result in non accepted manifest
https://bugs.launchpad.net/bugs/470355
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to eucalyptus in ubuntu.
--
Thanks for reporting this. Can you provide the AMI id you started with?
** Tags added: ec2-images
--
Re bundled karmic image fails to boot
https://bugs.launchpad.net/bugs/479621
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to eucalyptus
*** This bug is a duplicate of bug 450044 ***
https://bugs.launchpad.net/bugs/450044
Thanks Neil. This does appear to be a duplicate of bug 450044, so it is
being marked as such.
Torsten, please look at the other bug report to see if there is any
missing information that you can provide, or
Scott: Since you're solving this with an exec script, could you also
remove the incorrect /etc/hostname file while you're at it?
Even though I haven't found any standard packages which depend on
/etc/hostname it still bugs me to have an incorrect value stored there
by default.
--
ec2-images
Thierry: Half joking, I'll point out that anybody can have root on an
Ubuntu system for an hour for only $0.10 with EC2.
More seriously: If the benefit of the rsync approach is only to increase
the number of inodes, perhaps the original mke2fs could be run with the
-N option to increase the
AMIs for Amazon EC2 should be 10 GB (10240 * 1024 * 1024 bytes)
--
UEC images could be smaller
https://bugs.launchpad.net/bugs/439868
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to vm-builder in ubuntu.
--
Ubuntu-server-bugs mailing
** Changed in: vmbuilder
Assignee: Eric Hammond (esh) = Chuck Short (zulcss)
** Changed in: vm-builder (Ubuntu)
Assignee: Eric Hammond (esh) = Chuck Short (zulcss)
** Changed in: vmbuilder
Status: Fix Committed = Invalid
** Changed in: vm-builder (Ubuntu)
Status: Fix
Scott: What do you mean by not change anything else? If the user
modifies sources.list would a rebundled AMI overwrite their changes or
not?
I'm also not clear which ami_id is referenced in the last step. If it
is the AMI being created, then the user does not know the id yet.
--
Are we sure that group admin should have sudo rights by default in
Karmic? It doesn't in Jaunty.
When this bug was submitted, the AMI in question had a comment in
/etc/sudoers which claimed that admin should have sudo privs, but the
actual line to implement this was missing.
The latest Alpha-6
Upon further review, I stand corrected and the bug stands.
--
Sudoers file is misconfigured in AMI ami-5059be39
https://bugs.launchpad.net/bugs/423497
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to vm-builder in ubuntu.
--
I'm marking this Low based on my understanding of the importance
metrics, but since this can break a user's EC2 image on rebundling, I'd
love to see the fix released.
** Changed in: ec2-init (Ubuntu)
Importance: Undecided = Low
--
ec2-set-defaults should be 'run_once_per_ami'
** Changed in: vmbuilder
Status: New = Invalid
--
Karmic i386 EC2 kernel emulating unsupported memory accesses
https://bugs.launchpad.net/bugs/427288
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to vm-builder in ubuntu.
--
there presumably aren't very many modules for the EC2 kernel.
I guess this raises the questions of what modules we are talking about,
and this is an area where I wouldn't know what to cut out. I personally
depend on some which may not be considered part of the core modules
including: fuse, xfs,
Public bug reported:
It is fairly standard practice to include kernel modules in images for
Amazon EC2, but the most recent Karmic AMIs do not include them:
ami-a40fefcd
canonical-alphas-us/karmic-i386-alpha5.1.manifest.xml
ami-a20fefcb
** Also affects: vm-builder (Ubuntu)
Importance: Undecided
Status: New
** Changed in: vmbuilder
Status: Confirmed = Invalid
** Changed in: vm-builder (Ubuntu)
Status: New = Confirmed
--
Allow multiple exec scripts
https://bugs.launchpad.net/bugs/289998
You received this
** Changed in: vmbuilder
Status: Fix Committed = Fix Released
** Changed in: vm-builder (Ubuntu)
Status: Fix Committed = Fix Released
--
vmbuilder EC2: Retry ec2-upload-bundle on failures
https://bugs.launchpad.net/bugs/287860
You received this bug notification because you are a
** Changed in: vmbuilder
Status: Confirmed = Invalid
--
Add ability to build Debian virtual machines
https://bugs.launchpad.net/bugs/235562
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to vm-builder in ubuntu.
--
Public bug reported:
Binary package hint: ec2-ami-tools
The package ec2-ami-tools is outdated on Hardy, Intrepid, Jaunty,
Karmic. This prevents the usage of the latest EC2 features.
** Affects: ec2-ami-tools (Ubuntu)
Importance: Undecided
Status: New
--
Updgrade ec2-ami-tools
77 matches
Mail list logo
