I've been messing with the i386 ebs-backed F15 on a micro-instance. Very impressed. Really spectacular.
Now I've been modifying and adapting it to use as a mail server. I'd like to create a modified ami. I've seen the material using boxgrinder on the wiki:
https://fedoraproject.org/wiki/Cloud_SIG/EC2_Creation
but this may require more cloud-foo than I possess.
I see on the AWS Management Console an instance action "create image". Would this be a coward's way out?
sean
On Tue, 4 Oct 2011, sean darcy wrote:
I see on the AWS Management Console an instance action "create image". Would this be a coward's way out?
It certainly seems that way.
Try it:
Launch an AMI.
Modify the instance in some way -- add packages using yum, enable it as a webserver, etc.
Go through that create image workflow in the console. It will give you a private AMI ID for a new AMI that you own.
Then launch more of that instance, and see if it is in the same state as the original.
I haven't tried it, but it seems like that is the way it will work.
--Max
On 10/04/2011 03:56 PM, Max Spevack wrote:
On Tue, 4 Oct 2011, sean darcy wrote:
I see on the AWS Management Console an instance action "create image". Would this be a coward's way out?
It certainly seems that way.
Try it:
Launch an AMI.
Modify the instance in some way -- add packages using yum, enable it as a webserver, etc.
Go through that create image workflow in the console. It will give you a private AMI ID for a new AMI that you own.
Then launch more of that instance, and see if it is in the same state as the original.
I haven't tried it, but it seems like that is the way it will work.
--Max
Not quite OT, but I've yum upgrade'd the F15 ami. That installed kernel-PAE-2.6.40.4-5.fc15.i686. Rebooted - still 2.6.38.8-32.fc15.i686.PAE. Then fixed grub to boot 2.6.40. Reboot. Still 2.6.38.
cat /boot/grub/grub.conf default=0 fallback=1 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu
title Fedora-15 (2.6.40.4-5.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.40.4-5.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.40.4-5.fc15.i686.PAE.img
title Fedora-15 (2.6.38.8-32.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.38.8-32.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.38.8-32.fc15.i686.PAE.img
Does aws actually reboot the instance? If not, how do you (can you) upgrade the kernel? If aws does really reboot, what am I doing wrong?
sean
Sean,
Here's what I did as a test case:
(1) Launched an AMI (Amazon Linux AMI ami-7f418316 in us-east-1).
(2) yum upgrade -y, yum install -y httpd, chkconfig httpd on, touch ~/foo.txt
(3) Through the console, went to the Create Image (EBS AMI) option, and went through that workflow. This gave me a new AMI ID (private just to myself, though it could be made public).
(4) Launched another AMI with that new AMI ID.
(5) Logged into the new AMI to see that the packages I had installed were there, httpd was running, and ~/foo.txt existed.
That seems quite a bit different from your struggles to upgrade the kernel. I wonder if we're now talking about different things.
What AMI ID are you using, in what region, and what exactly are you trying to do?
--Max
On Tue, 4 Oct 2011, sean darcy wrote:
On 10/04/2011 03:56 PM, Max Spevack wrote:
On Tue, 4 Oct 2011, sean darcy wrote:
I see on the AWS Management Console an instance action "create image". Would this be a coward's way out?
It certainly seems that way.
Try it:
Launch an AMI.
Modify the instance in some way -- add packages using yum, enable it as a webserver, etc.
Go through that create image workflow in the console. It will give you a private AMI ID for a new AMI that you own.
Then launch more of that instance, and see if it is in the same state as the original.
I haven't tried it, but it seems like that is the way it will work.
--Max
Not quite OT, but I've yum upgrade'd the F15 ami. That installed kernel-PAE-2.6.40.4-5.fc15.i686. Rebooted - still 2.6.38.8-32.fc15.i686.PAE. Then fixed grub to boot 2.6.40. Reboot. Still 2.6.38.
cat /boot/grub/grub.conf default=0 fallback=1 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu
title Fedora-15 (2.6.40.4-5.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.40.4-5.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.40.4-5.fc15.i686.PAE.img
title Fedora-15 (2.6.38.8-32.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.38.8-32.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.38.8-32.fc15.i686.PAE.img
Does aws actually reboot the instance? If not, how do you (can you) upgrade the kernel? If aws does really reboot, what am I doing wrong?
sean
cloud mailing list cloud@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/cloud
Sean,
You need to edit /boot/grub/menu.lst file after you upgrade the kernel, /boot/grub/grub.conf is not used on EC2 at all (pvgrub).
--Marek
On 5 paź 2011, at 03:04, sean darcy wrote:
Not quite OT, but I've yum upgrade'd the F15 ami. That installed kernel-PAE-2.6.40.4-5.fc15.i686. Rebooted - still 2.6.38.8-32.fc15.i686.PAE. Then fixed grub to boot 2.6.40. Reboot. Still 2.6.38.
cat /boot/grub/grub.conf default=0 fallback=1 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu
title Fedora-15 (2.6.40.4-5.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.40.4-5.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.40.4-5.fc15.i686.PAE.img
title Fedora-15 (2.6.38.8-32.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.38.8-32.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.38.8-32.fc15.i686.PAE.img
Does aws actually reboot the instance? If not, how do you (can you) upgrade the kernel? If aws does really reboot, what am I doing wrong?
sean
cloud mailing list cloud@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/cloud
On 10/07/2011 06:03 AM, Marek Goldmann wrote:
Sean,
You need to edit /boot/grub/menu.lst file after you upgrade the kernel, /boot/grub/grub.conf is not used on EC2 at all (pvgrub).
--Marek
On 5 paź 2011, at 03:04, sean darcy wrote:
Not quite OT, but I've yum upgrade'd the F15 ami. That installed kernel-PAE-2.6.40.4-5.fc15.i686. Rebooted - still 2.6.38.8-32.fc15.i686.PAE. Then fixed grub to boot 2.6.40. Reboot. Still 2.6.38.
cat /boot/grub/grub.conf default=0 fallback=1 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu
title Fedora-15 (2.6.40.4-5.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.40.4-5.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.40.4-5.fc15.i686.PAE.img
title Fedora-15 (2.6.38.8-32.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.38.8-32.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.38.8-32.fc15.i686.PAE.img
Does aws actually reboot the instance? If not, how do you (can you) upgrade the kernel? If aws does really reboot, what am I doing wrong?
sean
If I install kernel-2.6.40-6, edit menu.lst to point to the new kernel, and then reboot, I can not ssh into the instance.
The Management Console show this instance running, but it's unreachable.
Same thing if I create a new ami image.
Odd.
sean
Sean,
Take a look at the console log and if appropriate - share with us so we can help you debug it.
--Marek
On 10 paź 2011, at 23:00, sean darcy wrote:
On 10/07/2011 06:03 AM, Marek Goldmann wrote:
Sean,
You need to edit /boot/grub/menu.lst file after you upgrade the kernel, /boot/grub/grub.conf is not used on EC2 at all (pvgrub).
--Marek
On 5 paź 2011, at 03:04, sean darcy wrote:
Not quite OT, but I've yum upgrade'd the F15 ami. That installed kernel-PAE-2.6.40.4-5.fc15.i686. Rebooted - still 2.6.38.8-32.fc15.i686.PAE. Then fixed grub to boot 2.6.40. Reboot. Still 2.6.38.
cat /boot/grub/grub.conf default=0 fallback=1 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu
title Fedora-15 (2.6.40.4-5.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.40.4-5.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.40.4-5.fc15.i686.PAE.img
title Fedora-15 (2.6.38.8-32.fc15.i686.PAE) root (hd0,0) kernel /boot/vmlinuz-2.6.38.8-32.fc15.i686.PAE ro root=LABEL=79d3d2d4 initrd /boot/initramfs-2.6.38.8-32.fc15.i686.PAE.img
Does aws actually reboot the instance? If not, how do you (can you) upgrade the kernel? If aws does really reboot, what am I doing wrong?
sean
If I install kernel-2.6.40-6, edit menu.lst to point to the new kernel, and then reboot, I can not ssh into the instance.
The Management Console show this instance running, but it's unreachable.
Same thing if I create a new ami image.
Odd.
sean
cloud mailing list cloud@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/cloud
-
Marek Goldmann -
Max Spevack -
sean darcy