| stackedsax | plars: I have never heard of anything like that | 01:38 |
|---|---|---|
| plars | stackedsax: any suggestions for debugging? | 01:39 |
| stackedsax | not from me — this is the agent you're trying to start? | 01:39 |
| plars | stackedsax: the other thing that is really suspicious, is that when I try to juju deploy things to lxc: targets, I get weird errors: http://paste.ubuntu.com/12121232/ | 01:40 |
| plars | stackedsax: bootstrapping goes fine, and the base node or anything else that goes on baremetal works fine, but the lxc bits start to cause problems | 01:41 |
| plars | I've successfully deployed the charm on a test maas system at home though | 01:41 |
| stackedsax | so, above the boot image error, there's a problem wget'ing | 01:41 |
| stackedsax | certificate | 01:42 |
| stackedsax | common name ''*'' doesn''t match requested host name ''10.101.49.149 | 01:42 |
| stackedsax | aren't you just having a problem connecting out? | 01:42 |
| stackedsax | like the container's networking isn't right | 01:45 |
| stackedsax | anyway, I'm heading out, sorry I'm not more useful. | 01:45 |
| mup | Bug #1486455 opened: CI fails when removing the curtin_branch parameter <MAAS:Triaged by rbanffy> <https://launchpad.net/bugs/1486455> | 09:02 |
| mup | Bug #1486489 opened: netaddr.core.AddrFormatError: failed to detect a valid IP address from None <MAAS:Triaged by mpontillo> <https://launchpad.net/bugs/1486489> | 10:08 |
| mup | Bug #1486505 opened: No static IP address / DNS record allocated to a deployed machine after commissioning <MAAS:Triaged by mpontillo> <https://launchpad.net/bugs/1486505> | 11:08 |
| mup | Bug #1425234 changed: ubuntu precise listed as invalid choice in error message while changing OS <oil> <MAAS:Invalid> <https://launchpad.net/bugs/1425234> | 13:35 |
| plars | Good morning, anyone have any ideas on debugging the problem I mentioned yesterday? | 14:08 |
| plars | I'm hitting a weird error on maas, it says "Boot image import process not started. Nodes will not be able to provision without boot images. Visit the boot images page to start the import." | 14:08 |
| blake_r | plars: what version of MAAS? do you have iamges on the images page? | 14:10 |
| plars | blake_r: 1.8.0+bzr4001-0ubuntu2~trusty1 and yes... I've even managed to deploy systems, but the error won't go away. It still thinks something is wrong. Deployment to baremetal works fine, but when I try to use juju to deploy lxc instances on top of maas, it blows up | 14:11 |
| blake_r | plars: juju does not use the images from maas for the lxc containers | 14:16 |
| blake_r | plars: have you tried giving the the region a restart | 14:16 |
| blake_r | plars: "sudo service maas-regiond restart" | 14:17 |
| plars | blake_r: to clear the boot image error? I can try it | 14:20 |
| blake_r | plars: yes | 14:20 |
| plars | blake_r: awesome, that took care of it, thanks! | 14:21 |
| blake_r | plars: np, now for the LXC issue | 14:21 |
| plars | blake_r: I would imagine that it was not causing any problems, but just weird that it didn't clear | 14:21 |
| blake_r | plars: are you using a proxy server? | 14:21 |
| blake_r | plars: yeah no problems, but agree its wierd that it didnt go away | 14:21 |
| plars | blake_r: I've asked on #juju about that, and no I'm not using a proxy | 14:21 |
| blake_r | plars: yeah #juju is the best place to ask | 14:22 |
| plars | blake_r: the errors I get at deployment, if you're interested, look like: http://paste.ubuntu.com/12121232/ | 14:22 |
| blake_r | plars: okay sometimes proxies can cause an issue | 14:22 |
| plars | blake_r: it's complaining about certificates and also seems to be trying to download from a url that is clearly invalid | 14:22 |
| blake_r | plars: looks liek your DNS might be incorrect | 14:23 |
| blake_r | plars: did you set the upstream dns server in MAAS? | 14:23 |
| blake_r | plars: you need to set that on the Settings page | 14:23 |
| plars | blake_r: hmm, no I didn't. Let me give that a try | 14:24 |
| plars | blake_r: it says that's only for when maas is running it's own dns server though, and I don't have it set up that way | 14:25 |
| blake_r | plars: how did you configure your cluster interface? "DNS+DHCP"? | 14:34 |
| plars | blake_r: dns/dhcp are handled through an established dns/dhcp server in the lab. The number of nodes in maas is very small right now, so they are just added there, and pointed to the maas server as next-server for booting | 14:35 |
| roaksoax | plars: if you are running maas in a lxc container, apparmor might be preventing dhcp to run | 14:40 |
| blake_r | roaksoax: he is not running DHCP with MAAS | 14:41 |
| blake_r | plars: can you ssh into a deployed node and access the internet? | 14:41 |
| plars | roaksoax: no, maas is not in the lxc container | 14:41 |
| plars | roaksoax: maas is just being used to deploy the base node, on which juju is deploying lxc instances | 14:41 |
| roaksoax | plars: ah so lxc instances are failing to deploy | 14:42 |
| plars | roaksoax: yes | 14:43 |
| roaksoax | plars: that might be an issue with the proxy as mentioned before | 14:43 |
| roaksoax | plars: or with juju in general | 14:43 |
| roaksoax | plars: since maas doesn't really touch the LXC deployment | 14:43 |
| plars | roaksoax: I'm talking to someone about it on #juju now, I suspect it's not a maas problem but I was having an odd error in maas earlier, so I wanted to get that cleared also | 14:43 |
| plars | roaksoax: there is no proxy | 14:43 |
| roaksoax | plars: the only thing that MAAS would do is to give IP addresses via DHCP | 14:43 |
| blake_r | plars: please ssh into a deployed maas not to make sure it can access the internet | 14:43 |
| plars | blake_r: it can | 14:46 |
| blake_r | plars: okay then something wierd with juju then | 14:46 |
| blake_r | plars: seems like MAAS is deploying fine and the nodes can access the internet | 14:46 |
| plars | blake_r: yeah, I'm going to try downgrading juju | 14:47 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!