[00:17] hey. harlowja [00:18] hey yo [00:20] when u want to go over https://openstacksummitmay2015vancouver.sched.org/event/b52d0c99f65d7713c61524b5a380812b [00:20] with vice executives [00:21] supyeah. [00:21] i do. [00:21] i had on my todo to bother you about tha ttoday [00:22] thank yo ufor beating me to it. [00:22] can we pick a time in my normal working hours tomorrow ? [00:22] :) === zz_natorious is now known as natorious === natorious is now known as zz_natorious === shardy_z is now known as shardy === rangerpbzzzz is now known as rangerpb [14:07] hi guys [14:07] does anyone know which cloud-init configuration I should use to disable pollinate to seed random ? [14:17] arnaud_orange, [14:17] # pollinate hangs without socket timeout if it can't reach network [14:17] random_seed: [14:17] command: null [14:17] will do it. [14:17] you can make 'command' anything you want. null is yaml's "None" [14:17] oh nice [14:17] let me try that [14:17] but you can make it [/bin/sh, -c, 'echo hi mom'] [14:21] Is there anything else to disable when my VM does not have access to internet, because cloud-init is taking very much time to bring my VM up [14:22] it's far better with your random_seed configuration, thanks smoser [14:22] but still taken 178sec to be up [14:23] what datasourc e? [14:23] openstack metadata [14:23] are you able to pastebin a /var/log/cloud-init.log ? [14:25] smoser: http://paste.ubuntu.com/11096632/ [14:30] arnaud_orange, it would appear you lose almost 2 minutes between lines 186 and 187. [14:31] yep [14:32] i would suggest disabling DataSourceGCE in /etc/cloud/cloud.cfg.d/90_dpkg.cfg [14:32] I think this is because the VM is not able to access http://metadata.google.internal./computeMetadata/v1/ is not resolvable [14:32] right. [14:32] what is that? [14:33] inside the image you have to change that. [14:33] GCE=GoogleCloudengine, right? [14:33] right. [14:33] remove GCE there. [14:33] please file a bug, as such a hang is not intended. [14:33] ok, if I do that, I won't be able to rely on the classic ubuntu cloud image [14:34] yeah, that sucks. :-( [14:34] file a bug, we can fix it. [14:34] yep [14:34] and SRU and then you can be happy again. [14:34] no way to fix it with yaml cloud-config? [14:34] no. [14:35] as this is config that is read before we can get to user-data [14:35] ie, it tells us how to find user-data. :) [14:35] understand [14:35] you could try to see why dns resolution would time out on that. [14:35] that does seem broken that a dns resolve would block for 120 seconds [14:35] the only networking provided at that point is from the dhcp server [14:36] so your dhcp server is possibly giving a dns server that is not reachable. [14:36] the dns server configured on the vm is 8.8.8.8 (google) [14:36] but the VM does not have access to internet [14:38] well, that'd seem like a misconfiguration :) [14:38] why is it 8.8.8.8 ? [14:38] this is configured on my tenant network [14:39] because, when I set a floating IP to my VM, the VM can access internet [14:39] but the floating IP is not set on startup [14:39] it's floating between VM [14:39] don't know if I am clear on that [14:41] right. but it seems like your network is busted still. [14:42] why give someone a dns server that they cannot reach [14:42] i guess so that later when they get a floating ip they can. [14:42] don't know, ask my cloud provider :p [14:42] who is cloud provider ? [14:42] cloudwatt [14:42] french company [14:42] but I think it might be the same on any openstack based cloud provider [14:43] no [14:43] thats definitely their configuration. and they could definitely give you access to a dns server internally that would dtrt. [14:43] i'd bug them about it. [15:02] the thing is that I only want my virtual machine to have access to other machines in the tenant, without access to the internet, so not having access a an external DNS is expected [15:05] i'd file a bug with them. [15:06] a.) telling you dns is 8.8.8.8 is hacky at best (rather than a dns server they run) [15:06] b.) telling you a dns server is that you can't get to is kind of silly/broken [15:14] ok, I found a way to remove the 8.8.8.8, so a) is not an issue anymore, [15:14] they provide their own nameserver [15:14] but, again, my tenant router have only one interface, in the private network [15:14] so without floating IP, no internet access [15:15] there is no way to disable every internet request from cloud-init point of view? [15:15] without modifying the ubuntu image, [15:15] ? [15:28] well, nto really. [15:28] but if you have dns configured correctly, then the dns should fial [15:28] and it will move on [15:28] quickly [15:29] you're right === zz_natorious is now known as natorious === natorious is now known as zz_natorious === zz_natorious is now known as natorious === natorious is now known as zz_natorious === zz_natorious is now known as natorious [17:08] 2015-05-12 17:06:07,091 - util.py[WARNING]: Failed to create user foobar [17:09] would be nice to know why :( [17:10] oh, I guess adduser failed for some reason [17:11] devicenull, /var/log/cloud-init-output.log maybe [17:11] yea, that's where I got that error message [17:12] not cloud-init.log [17:12] cloud-init-output.log [17:12] # less /var/log/cloud-init-output.log [17:12] yes, that's where that message was! [17:14] well thats not right. [17:14] turns out it didn't like the 'primary-group' setting I had [17:14] when the primary-group didn't exist [17:14] log should not be going to -output.log [17:15] oh wait. [17:15] or i guess *its* output, and on warn, it wrote that to stdout === natorious is now known as zz_natorious === zz_natorious is now known as natorious === shardy is now known as shardy_z === natorious is now known as zz_natorious