[02:33] Bug #1726202 opened: maas rackd tftp bootstrap.py: 'Nonetype' object has no attribute 'stopListening' [03:12] so i am new to maas and am trying to setup a home labe for testing it would someone be able to help me get this working [03:20] ...anyone [04:18] Bug #1708918 changed: maas node failed commisioning , failed to connect node via ssh === frankban|afk is now known as frankban [09:59] hello do you know how to setup wake-on-lan power type on maas 2.2 ? [10:01] https://stgraber.org/2017/04/02/using-wake-on-lan-with-maas-2-x/ this is not working anymore === cnfer is now known as cnf [16:40] Bug #1726474 opened: psycopg2.IntegrityError: update or delete on table "maasserver_node" violates foreign key constraint "maasserver_event_node_id_xxx_fk_maasserver_node_id" on table "maasserver_event" DETAIL: Key (id)=(xx) is still referenced from table "maasserver_event". [16:40]

=== frankban is now known as frankban|afk [19:13] Bug #1726534 opened: [2.3, HWTv2] Fio does not always gather metrics [20:00] ryebot: juju reload-spaces [20:41] roaksoax: http://paste.ubuntu.com/25803994/ <-- how long should 2.3.0~beta2 take to enlist a new vm? it would seem that it's hating my setup again [20:42] * lamont will turn on more debugging later, but it may be quite later [20:43] lamont: should be no different than 2.2 or older... check that /etc/maas/rackd.conf is pointing to an IP and not localhost, which could be messing communication with the metadata server [20:44] lamont: or actuall, also check /etc/maas/regiond.conf has the correct maas_url too [20:44] I am trying MAAS with virtualbox. My MAAS server has 2 ip address (one for internet, one for dhcp). The node is booting fine with DHCP/PXE but then cloud-init seems to be trying to get an image/file using the other ip (the public one, which it has no access to). [20:44] Is there a way to specify which ip the booting node must use for accessing MAAS? [20:45] maas_url: http://localhost:5240/MAAS [20:45] * lamont updates [20:45] edmz: same as for lamont, make sure /etc/maas/rackd.conf doesn't point to localhost and points to an IP the machines can access to [20:47] roaksoax : thank you [20:50] roaksoax: when the maas_url in regiond.conf was a hostname, everything hated life. [20:51] roaksoax: which was left over from having ipv4 and 6 active on different vlans on the same rack [20:52] I'll poke at it some more at some point [20:54] roaksoax: it worked. [20:54] lamont: probably dns resolution ? [20:54] I really should purge/reinstall maas on this cluster, since it's the left over from whatever the last hacking I was doing was. with the IP there, the commissioning worked. [20:54] lamont: but yes there's still some leftovers there [20:54] roaksoax: yeah, possibly that [20:54] edmz: cool! [20:55] is there a place where I can read about the support/limits of using servers that dont have IPMI/BMC? [20:56] edmz: supported power types are listed here: https://docs.ubuntu.com/maas/devel/en/nodes-power-types [20:57] roaksoax : currently I am testing with virtualbox. My real servers do have BMC [20:59] edmz: you can use "manual" power type [21:00] That is what I currently have. After the boot process the booting node just closes suddenly. I am wondering if it's because of lacking bmc/ipmi. [21:01] and I get a "failed comissioning" status. [21:02] edmz: uhmm are they pxe booting and then something happens, and then it changes to failed commissioning ? [21:04] I only see 1 node in the nodes page. It always has "failed commisioning" status. [21:04] If I boot that node, while the boot process is going on nothing changes on the nodes page. [21:05] The boot processs continues past the part where it is downloading packages from ubuntu servers [21:05] then the virtualbox window suddenly closes. [21:05] I managed to save a video of the output with Virtualbox. [21:06] I see, it is actually running shutdown. [21:08] the only odd thing I see is "ci-info: no authorized ssh keys fingerprints found for user ubuntu" [21:10] edmz: right so the commissioning process actually starts doing something [21:10] did you see if any of the commissioning scripts failed [21:10] edmz: go to the machine details page and look at 2 things: [21:10] 1. 'Commissioning' tab and 'Event's tab [21:10] that should give you more details on what may be wrong [21:11] i just deleted the machine. Originally i had to add it via the nodes page "Add Hardware" option. Then specifying the MAC address. [21:11] I was abou to do that again. [21:12] Maybe I am missing a step. [21:12] edmz: so auto enlistment may have been failing due to the localhost thing [21:12] if that's what you mean [21:12] yeah [21:13] so should I try to boot it without adding it? so it tries to auto enlist it? [21:14] edmz: yeah, [21:14] but i'm pretty sure that's the issue [21:16] yeah, that did it [21:16] Status new. [21:21] roaksoax : I really appreciate your help. It worked. [21:25] edmz: cool! glad it does