[00:10] so the error initially is about bootif: siocgifindex: no such device. [00:10] wondering if this could have anything to do with running M.2 drives? === CyberJacob is now known as zz_CyberJacob === frankban|afk is now known as frankban [10:42] Bug #1611711 opened: [API] block-device 'add/remove_tag' operations use GET method, not POST [11:24] Bug #1611726 opened: Documentation for machine create API doesn't say that power_type is required [13:03] Bug #1340920 changed: Extra slash in /etc/apt/sources.list; prevents connecting to some servers [13:33] Bug #1611761 opened: [2.0 RC3 ] Unhandled Error - maasserver.models.regioncontrollerprocess.DoesNotExist

[13:45] I am using MAAS with auto pilot and it only auto detects one node out of 12. Is there a place to look to help mass find all the nodes? [13:46] Or get maas to re-try auto discovery? [13:47] newmember: what do you mean by auto-discovery ? [13:48] when installing maas it appears to look at the network for nodes, it found one. [13:48] i see all the nodes picked up an IP address form maas [13:49] from maas [13:54] newmember: maas doesn;t look at the network for nodes [13:54] newmember: if a machine that maas knows nothing about, DHCP's from MAAS [13:54] we will know that /something/ has DHCP'd [13:55] newmember: but typically, MAAS "auto" discovers machines when machines DHCP from MAAS *and* PXE boot from MAAS [13:58] newmember, you basically need to PXE boot all your 12 nodes against MAAS [15:53] Hi everyone, what is the default password that MAAS sets for a node? preseeds master file doesn't specify any password [15:54] is there a way that we can use for setting up password for the nodes that we deploy? === frankban is now known as frankban|afk [17:07] junaidali, there's no default password -- use the registered ssh key to log in [17:07] junaidali, the default user on ubuntu is called ubuntu, and on centos is centos [17:08] thanks. Is there any setting to set the password for every node that we're provisioning? [17:10] password for ubuntu user* [17:21] junaidali, no, you'll need to do that through a curtin hook. I don't know if I think that's a sane thing to do though [17:41] Thanks kiko [17:42] If I want to add CentOS images to my MAAS setup, is that possible? [17:42] The docs seem to imply that it is, but don't describe how. [17:44] in MAAS 2.0, it is pretty easy. which version of MAAS are you using? [17:45] junaidali: I'm using 2.0.0~rc3+bzr5180-0ubuntu2~16.04.1 [17:47] * nturner sees 2.0.0~rc4+bzr5187-0ubuntu1~xenial1 is now available and updates... [17:49] nturner: centos images are in the images page, you can just import them [17:51] hmmm [17:54] here's what I see on that page: https://snag.gy/XounFl.jpg [17:56] Am I tracking the wrong thing? Under Settings my Boot images Sync URL is https://images.maas.io/ephemeral-v2/releases/ [17:56] Which is whatever MAAS defaulted to. Maybe the default has changed for new installs but upgrades don't change it? [17:57] blake_r: What Sync URL are you using? [17:57] nturner, can you change releases to daily? [17:57] sure... [17:58] well, golly, that was easy! [17:58] Thanks all. [18:39] thanks kilo [18:39] I am using the autopilot deployment from Ubuntu [18:39] I installed MAAS [18:39] I created 12 nodes [18:40] When I look on MAAS all I see is the one node created in MAAS none of the other nodes are created in MAAS [18:40] I would have thought that they all auto create in MAAS [18:40] they are all getting DHCP from MAAS [18:41] Maybe autopilot uses the vmware api to get the 'node' names and load them in MAAS, maybe its not a DHCP discovery thing [19:16] newmember, you don't need to create the nodes manually [19:16] newmember, and in fact, I argue you shouldn't [19:18] kiko, okay [19:25] newmember, instead, just PXE-boot the machines and see them appear [19:27] newmember, if they don't appear, you have a problem that needs fixing [19:28] kiko, I think I have a problem that needs fixing :) [19:29] newmember, do the machines PXE boot successfully? [19:31] gets an ip address [19:31] booting under maas [19:31] some long line of stuff [19:31] then stops at "boot:" [19:32] there is a note: could not find kernel image [19:36] newmember, could you get us a screenshot? [19:41] kiko, great idea [19:41] http://picpaste.com/Capture-50dZaauj.PNG [19:42] newmember, yeah, that's definitely odd. what releases do you have selected in your images page? [19:43] checking [19:43] newmember: it is either you are missing the i386 image , or the import of images hasn't completed [19:45] newmember, I'm slightly surprised that enlistment is trying to use the i386 image [19:45] newmember, what hardware are you enlisting? [19:48] I used the autopilot ova from ubuntu [19:48] newmember, what's happening is this: [19:48] - pxelinux's ifcpu64 module is determining your machines are 32-bits only [19:49] - it then tells MAAS to boot using a 32-bit kernel [19:49] - the 32-bit images are not yet imported in MAAS, so it fails [19:49] I think everything has to be 64bit [19:50] newmember, definitely -- the question is why is this not working in your situation [19:51] newmember, are these actual machines, or VMs? [19:51] vms on esxi6 [19:51] ah [19:51] newmember, and these vms, are they configured to be 32-bit or 64-bit? [19:51] i imported the ova from ubuntu [19:51] looking [19:52] hmmm! [19:53] newmember, I am a bit clueless about our autopilot vmware install (and vmware) -- so the ova contains the machine definitions already? [19:53] just on a call, please hang on [19:54] sure [19:54] but thanks [19:59] its 64bit [19:59] http://picpaste.com/Capture2-ybPaWJdf.PNG [20:05] newmember, well.. that doesn't actually say the machine is 64-bit [20:05] newmember, anyway, can you import the i386 images and see if it works? [20:11] newmember, I need to split but will look through the channel traffic later tonight or tomorrow morning [20:12] cheers [20:13] newmember: while it may be 64 bit VM, pxelinux may be identifying it as i386 [20:13] newmember: i've seen that before [20:15] dpb1, ping [20:15] dpb1, have you seen the above? [20:15] not specifically with vmware though [20:17] we haven't seen that specifically, no [20:19] would be interested to see versions though, if you get to that. [20:19] kiko: just so you know, the autopilot doesn't do anything directly with vmware. we treat it as a regular maas node. [20:19] the instructions we have also need to be followed exactly [20:24] dpb1, right, but do we provide the machine definitions (in machine-readable or doc form), or are they to be manually created? [20:28] Bug #1611949 opened: cryptic error when PXE-boot requires an image not yet imported [20:31] Bug #1611949 changed: cryptic error when PXE-boot requires an image not yet imported [20:40] Bug #1611949 opened: cryptic error when PXE-boot requires an image not yet imported [22:50] Bug #1611980 opened: Upgrade from MAAS 2.0 RC3 to RC4 has pre-removal script error messages [22:59] Bug #1611980 changed: Upgrade from MAAS 2.0 RC3 to RC4 has pre-removal script error messages [23:05] Bug #1611980 opened: Upgrade from MAAS 2.0 RC3 to RC4 has pre-removal script error messages [23:32] Bug #1611984 opened: MAAS images page doesn't correctly report image status