[01:11] hi [01:13] I'm trying to install ubuntu server on EC2 following this guide: http://www.howtoforge.com/perfect-server-ubuntu-10.04-lucid-lynx-ispconfig-3-p3 , there's a section where we're supposed to edit /etc/network/interfaces file to setup static IP, I want to know if I have to do that on EC2 ? (i'm using elastic ip) [01:44] marc__, no, you should use dhcp on ec2 [06:40] i just started my first instance using the 64bit 10.04 ami [06:41] and i can't ssh it, connection refused [06:41] so this is probably more of an ec2 question than an ubuntu one... but what do i have to do to allow ssh [06:42] the ubuntu docs assume you already have an ubuntu machine to install the ec2-api-tools but i don't so i'm stuck in a bit of chicken and egg, using the aws console [07:10] ah, nvm [07:10] turned out to be a problem with the aws console in chrome, my edits to the default security group actually stuck/worked in firefox === popey__ is now known as popey [14:55] do I have to edit the hosts file in a ec2 instance? [15:54] marc__: I don't remember having to [15:58] veenenen, I think that tools like ISPConfig use the hostname, maybe it would be nicer to have a hostname like www1.mywebsite.com instead of ec2-xxxxxxx... but I'm not sure how to set the hosts file, I was thinking of using the public DNS? "ec2-xxx-xxx-xxx-xxx..compute-1.amazonaws.com www1.mywebsite.com" any chance that would work or will I lose the ability to connect to my instance? [15:59] no, you can change the hosts and hostname file as much as you want [15:59] they aren't used outside the server [15:59] it's really only for internal routing in the server [16:00] at least, in my experience [16:00] i generally switch out the hostname and main 172.0.1.1 record in the hosts file for the domain of the server [16:03] alright, I'll try that, thanks veenenen [17:48] is sshd enabled by default on the 10.04 ami's? [17:53] yeah [17:54] bummer [17:54] i stop'd and an ebs-based instance, and then started it again, but can't ssh to it now [17:54] i ran into the same issue [17:54] the ip/dns changed, that took me a moment to notice [17:55] have they fixed the fstab bug yet? [17:55] oh cool, is it a bug or am i missing something [17:55] i just started with both ec2 and ubuntu last night :) [17:55] they had a sdb mount in there that stopped the instances from rebooting [17:55] how would you even troubleshoot something like that [17:56] is there really no out-of-band console access? [17:56] on the first boot, i had to go in and remove it [17:56] like, how did the first person who discovered that do it [17:56] i had ssh on the initial boot, but lost it after a reboot [17:56] smoser caught it for me [17:57] gotcha so i just consider this instance a loss, nuke it, and make sure to comment out that line before rebooting next time [17:57] yup [17:57] hopefully they'll fix it at some point [17:57] they may be a tad busy right now though [17:58] whats the url for the ami's again [17:58] that should be in the topic here [17:58] http://uec-images.ubuntu.com/lucid/current/ [17:58] just keep this bookmark around [17:58] thanks [17:58] it's pretty helpful [17:59] is there a forum or a mailing list thats keyed in on this [18:16] i just noticed your url ends in /current/ and mine in /release/ [18:17] is there anywhere i can read the release policy [18:17] on what the difference is [18:34] hmm maybe its this and i just haven't been patient enough [18:34] https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/567334 [18:34] Launchpad bug 567334 in linux "blocked tasks delay cloud-init for 240 seconds" [Medium,Triaged] [19:56] jbartus, the reboot issue is https://bugs.launchpad.net/ubuntu/+source/cloud-init/+bug/634102 [19:56] Launchpad bug 634102 in cloud-init "t1.micro instances hang on reboot" [Medium,Confirmed] [19:59] jbartus, sorry, i hadn't read all above. you really shouldn't hit any "can't get to machine" if your security groups are set up correctly. [19:59] ssh should start by default, but after a stop and start, it will change IP addresses (as you've found). [19:59] the "won't reboot" really only will affect t1.micro, that has no ephemeral storage. [20:14] hey, thanks smoser [20:15] are you set now ? [20:15] i just tried reproducing it, new instance booted fine, commented out sdb, rebooted fine, un-commented sdb, rebooting right now [20:15] reading the bug report now [20:19] ok thats certainly it [20:20] so i just comment it out and presumably it'll be a non issue in some future ami [20:34] that only affects t1.micro [20:34] it will be a non-issue in future amis [20:35] and, will be fixed in a week or so by doing a "apt-get update && apt-get upgrade" after youve launched the instance === Daviey_ is now known as Daviey