AskUbuntu_ | Problem installing MAAS nodes on Intel NUC | http://askubuntu.com/q/489786 | 01:27 |
---|---|---|
l1fe | sorry to be a bother, but has anyone setup a maas node on an intel NUC? | 01:51 |
mwhudson | l1fe: was it you how posted the askubuntu question that got linked in here? | 01:54 |
mwhudson | http://askubuntu.com/q/489786 | 01:54 |
mwhudson | l1fe: i haven't but i know it's been done | 01:54 |
mwhudson | l1fe: what are you using for power control? | 01:54 |
l1fe | mwhudson: that would be me | 01:54 |
l1fe | mwhudson: I'm using wake on lan (it's a haswell NUC) | 01:55 |
mwhudson | ok | 01:55 |
mwhudson | when you say that you boot the device and it looks like the old install | 01:56 |
mwhudson | how are you booting it? | 01:56 |
l1fe | so first boot is with PXE, that installed the image (i'm guessing) that enlisted the node | 01:56 |
l1fe | second boot, was also with PXE | 01:57 |
l1fe | and i think that's when it was commissioning | 01:57 |
l1fe | 3rd boot was normal boot to hdd | 01:57 |
* mwhudson tries to remember maas terminology | 01:58 | |
mwhudson | i don't think that's quite how it works though | 01:58 |
mwhudson | the image isn't installed until you provision the node | 01:58 |
mwhudson | with maas, you're not expected to boot the device yourself ever -- maas does it | 01:59 |
l1fe | yeah, unfortunately it doesn't seem to WoL | 01:59 |
l1fe | and the end of every step seems to be the NUC shutting down, while maas thinks the node is "ready" | 01:59 |
mwhudson | that's expected | 02:00 |
mwhudson | the shutdown thing i mean | 02:00 |
mwhudson | i think maas really expects some more uh enterprisey oob management | 02:00 |
mwhudson | like ipmi or hp ilo | 02:00 |
l1fe | right, it just never starts it back up - guessing that's why the folks who I think got NUC setup with maas used the non-haswell version with AMT enabled | 02:01 |
mwhudson | yes | 02:01 |
mwhudson | certainly i know that's why the orange box has that particular version... | 02:01 |
l1fe | hmm that's kind of disappointing...guess I'll have to see if I can make this work somehow | 02:06 |
mwhudson | you can always use the human pdu approach :) | 02:07 |
l1fe | haha, that's basically what I've been doing - but the success rate seems to be lacking... | 02:09 |
=== vladk|offline is now known as vladk | ||
=== vladk is now known as vladk|offline | ||
=== CyberJacob|Away is now known as CyberJacob | ||
=== vladk|offline is now known as vladk | ||
=== vladk is now known as vladk|offline | ||
=== frobware- is now known as frobware | ||
=== vladk|offline is now known as vladk | ||
=== g_ is now known as Guest88496 | ||
gmb` | allenap: Unless jtv has other ideas, you could start with adding the new config template / generation stuff for DHCPv6 | 12:37 |
allenap | gmb`: For DHCP? | 12:38 |
=== gmb` is now known as gmb | ||
* rvba agree with gmb. The DHCP stuff is probably the next big thing to get sorted. | 12:39 | |
rvba | agrees* even | 12:39 |
gmb | Yarp. | 12:39 |
gmb | allenap: Yes. | 12:39 |
allenap | gmb: Mucho gracias. | 12:40 |
rvba | allenap: the trick is that we need to get two servers up and running. | 12:40 |
gmb | De nada. | 12:40 |
gmb | allenap: What rvba said. The config stuff is the easy part, but we’ll need two instance of dhcpd… which isn’t that hard either, but that’s potentially the trip-up point. | 12:40 |
bigjools | seen this? http://www.test-ipv6.com/ | 12:46 |
rvba | Yep. I've used it when testing that my ISP was supporting IPv6 out of the box. | 12:49 |
bigjools | rvba: my modem doesn;t support ipv6 :/ | 13:02 |
rvba | /o\ | 13:02 |
bigjools | rvba: stuck with tunnelling then | 13:04 |
rvba | Yep | 13:04 |
=== plars_ is now known as plars | ||
=== ftp21 is now known as d4rkn3t | ||
d4rkn3t | hello dear, I neet help with MAAS/JUJU is there someone can help me? thanks | 13:54 |
d4rkn3t | I've run the command "juju bootstrap --upload-tools -e maas --debug" during the all debug JUJU try to connect to a node "Node02Cluster01Svr:22". the node change its status from ready to allocated, with the OS in running. after 10 minutes the error is this "ERROR juju.cmd supercommand.go:305 waited for 10m0s without being able to connect: ssh: Could not resolve hostname node02cluster01svr: Name or service not known". In the Region Contro | 13:54 |
d4rkn3t | ller I've set DNS and DHCP service. | 13:54 |
AskUbuntu_ | juju bootstrap using maas unable to ssh into nodes | http://askubuntu.com/q/490000 | 13:59 |
* rvba replies on http://askubuntu.com. | 14:04 | |
=== jfarschman is now known as MilesDenver | ||
=== CyberJacob is now known as CyberJacob|Away | ||
=== vladk is now known as vladk|offline | ||
l1fe | can someone go over the different states that maas can be in? currently i have four nodes in "allocated to root" - does this mean they are ready to be used or is there something else that needs to happen? | 18:33 |
l1fe | i tried to juju bootstrap, but I keep on getting 409 conflict with no nodes available | 18:33 |
jhobbs | l1fe: allocated to root means they should be started and running | 18:34 |
jhobbs | l1fe: since they're all allocated, there are no more for juju to bootstrap with | 18:34 |
jhobbs | l1fe: you'll want to stop them if you want to make them available to juju; they should be in the Ready state then | 18:35 |
l1fe | hmm ok | 18:35 |
l1fe | i must be really confused about how maas and juju interact | 18:35 |
l1fe | sorry for the obtuse questions :) | 18:35 |
=== jfarschman is now known as MilesDenver | ||
=== vladk|offline is now known as vladk | ||
=== vladk is now known as vladk|offline | ||
AskUbuntu_ | deploying charms using juju fails with tcp connection timed out | http://askubuntu.com/q/490141 | 21:57 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!