=== dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk === progre55_ is now known as progre55 [15:12] hey guys. I was setting up a node from a server installation cd, but after a restart falls to (initramfs) saying "Gave up waiting for root device. Common problems: .... etc" [15:12] any suggestions, please? === dendro-afk is now known as dendrobates === rgreening_ is now known as rgreening [16:31] hi guys. when running an instance, I get these exceptions (http://pastie.org/1160744) on the node logs, and it terminates. Any suggestions please? [16:33] brb === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates === dendrobates is now known as dendro-afk [17:36] Are there any tools built into cloud-utils to unboot an image for pushing it to ec2? [17:40] veenenen, no. [17:40] as much as i hate un-booting, i'd be open to such a thing [17:40] if you wanted to contribute :) [17:41] Once I get this all working, I'd be happy to contribute it back [18:43] hello all [18:44] guys can u just provide me a full documentation of ubuntu cloud so that i can understand it fully [18:44] plz help me... [18:47] hello everone [18:48] I am using ubuntu cloud for setting up the cloud at my college... but didnt knw hw to move further... [18:48] i have just installed the server... [18:49] plz help me this is my project... plz help [18:51] Crazy, did you think people were just ignoring you because of your name ? :) [18:51] have you looked at https://help.ubuntu.com/community/UEC ? [18:51] the easiest way to get up and running is to do the uec install from the cd [18:51] hmm ya bro i have looked at it [18:52] ya i have installed that succesfully [18:52] so what do you need? [18:52] bt pbm is how i put my software in the cloud [18:52] suppose i have to put adobe photoshop there [18:53] well, it all starts by getting an AMI (amazon machine image) (or emi) up [18:54] how smoser? [18:54] the easiest way to get one is probably from the UEC store (i think there is a store tab in the UEC ui) [18:54] install the lucid ubuntu ami [18:54] alternatively [18:55] download from http://uec-images.ubuntu.com/releases/lucid/release/ [18:55] and use 'uec-publish-tarball' on ubuntu-10.04-server-uec-amd64.tar.gz or ubuntu-10.04-server-uec-i386.tar.gz [18:55] ok thanks... but do you have any proper documantation so that i can learn from my own [18:57] plz smoser u look to me the most helping person here... [18:58] should i really change my name, does people here help by seeing there names [18:59] https://help.ubuntu.com/community/UEC/CDInstall steps you through most of that. [18:59] up to getting an image [18:59] (i was only joking. you were here as Ankit and then left and came back as Crazy) [19:04] hmmm okz bro... [19:04] ya bro may i discuss further wd you [19:04] what i have done till yet [19:04] srry for my poor grammar [19:06] i'm sorry, but i really can't spend much time trouble shooting here today. [19:07] ok smoser thanks for helping me... i will download that image and come back if any pbm will be there, thanks... bro it was my first time on IRC so dnt knw hw to go on... [19:07] bye [19:49] It worked! Local kvm vm to live ec2 vm. [20:34] veenenen, wow! [20:34] dude! [20:34] that rocks [20:34] you used the maverick stuff i pointed you at ? [20:35] i'm very interested in details [21:00] After I stopped trying to get fancy, all I had to do was start up the original image, add my ssh key and turn off text passwords. Then just did a standard shutdown. After that I mounted the img to the host and changed the fstab to the regular ec2 fstab. Used ec2-bundle-image (I had to remember to use the ext4 friendly kernel), ec2-upload-bundle, and ec2-register to wrap up the image and push it to ec2. After that, I just launched it and tada it worked [21:04] so why, then, did you ahve to boot it at all ? [21:04] and i dont know that you would have had to change fstab [21:04] this is if you started from a maverick image [21:05] I just wanted to make sure it could be booted on kvm [21:06] Ideally, I'd run the run the server locally to test it, and then when everything is working right, I'd shut it down and push it out. [21:06] So, while I didn't need to start it up now, I will in the future. [21:06] right. i understand your general interest. [21:06] ok. [21:06] you shouldn't have to mount the image in the host to change the fstab though [21:06] you were using maverick ? [21:07] yeah [21:07] in general, you'll always want to use the same kernel that the registered image on ec2 works with [21:07] for maverick images, that "kernel" is pv-grub [21:07] I'm using the one that's currently being used by the 32bit maverick ami [21:08] I'll have to switch it out tomorrow after the kernel updates are done. [21:09] Most of the time I'll have the image running, so I just want the fstab switched out when I'm pushing it. I'm just going to make the mount part of the push script. [21:13] veenenen, so the maverick amis wont change [21:13] the kernel is loaded from inside the image [21:13] oh, sweet [21:13] via pv-grub on ec2, and via that floppy on your local boot [21:13] that's perfect [21:13] (and it can be done via a floppy based loader on 10.10 UEC) [21:13] what did you change in the /etc/fstab ? [21:14] (you should still look at what aki was used for a given ami, but it is not going to change each time there is a new ubuntu kernel, it would only change if amazon released a new pv-grub, and we needed that) [21:15] http://paste.ubuntu.com/494395/ [21:16] I just grabbed the fstab of one of my instances on ec2 [21:16] you shouldn't need to do that. [21:17] cloud-init will write those ephemeral entries to fstab for you [21:17] in fact, you *dont want* to do that [21:17] lest you run into a bug like bug 634102 [21:17] Launchpad bug 634102 in cloud-init "t1.micro EC2 instances hang on reboot" [Medium,Fix released] https://launchpad.net/bugs/634102 [21:18] <_mup_> Bug #634102: t1.micro EC2 instances hang on reboot [21:18] <_mup_> Bug #634102: t1.micro EC2 instances hang on reboot [21:18] Launchpad bug 634102 in cloud-init "t1.micro EC2 instances hang on reboot" [Medium,Fix released] [21:18] <_mup_> Bug #634102: t1.micro EC2 instances hang on reboot [21:18] <_mup_> Bug #634102: t1.micro EC2 instances hang on reboot [21:18] _mup_, uvirtbot, FIGHT! [21:18] someone needs to stop that [21:18] haha [21:19] do I run cloud-init on the img from the host before I start pushing the image to amazon? [21:19] or when the guest is shutting down should it already be adding those? [21:20] I just know the image wasn't booting on amazon until I changed the fstab [21:21] and the error said something about sda1 [21:22] cloud-init runs "once per instance" based on the instance-id [21:22] hm.. [21:22] hm.. [21:22] you'r eprobalby right.... [21:23] but uncloud-init or cloud-init should not modify /etc/fstab like that [21:23] oh wait, maybe... you might have more luck with a newer image. i made a fix for that bug recently in maverick (monday or so) [21:24] I grabbed the image today [21:24] What's cloud-init written in? [21:24] programming language-wise [21:28] python [21:29] uncloud-init is even more hacky, its written in shell [21:43] Alright, I'll dig in and see what's going wrong sometime this week. [21:47] veenenen, thanks. [21:47] and i am very interested in your findings. [21:47] i really would like to have a generally functional "download image, boot in kvm, customize, push to ec2" path