[00:09] hey roaksoax === CyberJacob is now known as CyberJacob|Away === zz_mwhudson is now known as mwhudson === claytonk is now known as ninjix === ninjix is now known as claytonk === mwhudson is now known as zz_mwhudson [05:30] bigjools: howdy!! [05:49] roaksoax: hey - just wondered what the state of the rabbitmqlib stuff was? [05:50] regarding the crashes [05:58] hello maas team, what does No PXE template found in u'/etc/maas/templates/pxe' means? [05:59] i failed to deploy node the nodes is trying to install OS, and I got this error from pserve.log [06:00] rawang: are the templates there? [06:01] bigjools, yes, they are all there [06:01] rawang: can you paste the pserv log and the maas log please [06:01] bigjools, sure [06:06] hi bigjools , i can paste the logs, but would you be happy to log in the the boston lab directly, I can PM your creds, or you'd prefer to see the paste ? :) [06:08] rawang: logging in is nicer for sure :) [06:08] bigjools, thank you [06:20] bigjools, pserve.log http://pastebin.ubuntu.com/6754647/ [06:21] bigjools, maas.log http://pastebin.ubuntu.com/6754651/ [06:30] bigjools: so we are now default to use python-amqp, but that's set by maas [06:30] roaksoax: I saw, that's great, thanks [06:31] rawang: it looks for templates using this schema: [06:31] config.{purpose}.{arch}.{subarch}.template [06:31] config.{purpose}.{arch}.template [06:31] config.{purpose}.template [06:31] config.template [06:31] bigjools: other than that, is the lander working fine? this afternoon it didn't seem so [06:31] rawang: are you sure nothing got removed? [06:31] roaksoax: the package lander is bust, I landed yours manually [06:31] bigjools, i don't think so :) [06:31] bigjools: ok cool, thanks! [06:32] roaksoax: it lives in the lab so I am going to move it to canonistack [06:32] bigjools: ok, rvba also reported the trunk lander wasn't working either [06:32] rawang: can you tell me what arch you're booting, and prove an ls of the /etc/maas/templates/pxe dir [06:32] bigjools, it's all there, 8 temples [06:32] bigjools, amd64 [06:32] ubuntu@maas:~$ ls /etc/maas/templates/pxe/ [06:32] config.commissioning.armhf.template config.install.armhf.template config.local.amd64.template config.local.template [06:32] config.commissioning.template config.install.template config.local.i386.template config.xinstall.template [06:33] roaksoax: yes it was broken earlier, all the canonistack instances had been shut off [06:33] roaksoax: so much for the freaking cloud [06:33] lol [06:34] rawang: I'm afraid I have no idea what's up [06:34] it looks fine [06:35] bigjools, i see, thanks for the help :) [06:35] rawang: I want to work out what's up though [06:35] me too :) [06:35] bigjools, maybe you can setup the routes, and connect to the box to see what's up if you have time right now? :) [06:36] bigjools: hehe!! [06:36] bigjools: nayway, i'm off to bed. Have a good one man [06:36] rawang: accessing the box is not a problem, it's the key [06:36] rawang: cheers, night! [06:36] errr [06:36] roaksoax: : cheers, night! [06:37] bigjools, actually it's use username/password to login, but i have no idea why it's asking key [06:37] bigjools, maybe you can try 10.230.167.2? [06:37] bigjools, to see if login works? [06:38] rawang: ssh will be configured to disallow p/w auth [06:39] rawang: same problem with that IP [06:39] hmm... [06:41] bigjools, https://pastebin.canonical.com/102979/ [06:41] bigjools, it's trying public,password [06:42] rawang: gah, it's batuan that's kicking me out [06:42] bigjools, oh lord :) [06:43] bigjools, probably need to fix batuan issue first. :) [06:44] yep [06:47] jtv2: can you suggest why rawang gets this in his pserv? http://pastebin.ubuntu.com/6754647/ [06:47] apparently the templates are there [06:47] jtv2, yeah, :) [06:51] rawang: what state is the node in that you're booting? [06:52] bigjools, when I boot it, the server become from "Ready" to "allocate to root", [06:52] rawang: so you're using juju or starting in the UI? [06:52] bigjools, and the server boots, pending at "try to load pxelinxu.conf/, and wait for time out [06:53] bigjools, yes [06:53] ummm stumped [06:53] bigjools, and when it's time out, it's loading amd64/generica/commissioning/linux.gz etc [06:54] oh it starts loading the image? [06:54] bigjools, and when the kernel / initrd loads, it's booted up ,and it's show "maas-enlist-node" as the hostname.... [06:54] as in, it gets past the pxe config? [06:54] bigjools, yeah, when it's time out, it's loading linux.gz and initrd.gz from comissioning directory [06:55] and end up with show "maas-enlist-node" , that confuse me :) [06:55] rawang: https://bugs.launchpad.net/maas/+bug/1064212 [06:55] Ubuntu bug 1064212 in MAAS "If a machine is booted manually when in status "declared", TFTP server tracebacks" [Low,Triaged] [06:56] bigjools, hm, it's still a open bug [06:56] rawang: exactly :) [06:57] freeflying, ^^ [06:57] rawang, we don't have any node in Declared [06:58] or Ready [06:58] freeflying, yeah, it's a different, but someone also mentioned same applies in Ready state [06:58] did you manually power up something? [06:58] bigjools, yes, manually power on the server after juju deploy something [06:59] bigjools, as maas doesn't set up the ipmi correctly [07:00] rawang: in that case you are powering up the wrong one [07:00] I suggest that you set the ipmi properly [07:00] bigjools, i don't think so, as i can see the hostname and IPMI ip of that server [07:00] rawang: that is the only reason you will get this message, can you double check [07:01] bigjools, ok [07:05] bigjools, yep, i have double checked that I started the correct server, and it keeps reporting the same error [07:06] bigjools, output that error many times, exceptions.AssertionError: No PXE template found in u'/etc/maas/templates/pxe'! [07:07] rawang: can you set up the IPMI details properly and let it do its own thing [07:08] bigjools, ok, let me try it [07:13] bigjools, hmm, seems like this time it works,but maas didn't detect the IPMI, and didn't create maas user, i put the admin user in the nodes' configs [07:13] rawang: it only detects IPMI at enlistment time [07:13] otherwise you need to enter manually [07:14] if it worked this time then you were turning on the wrong one before :) [07:14] bigjools, yes, what I mean is during the enlist stage, it didn't create maas user in server's IPMI,and didn't put the correct value in the node's config [07:14] rawang: what equipment? [07:15] rawang: ok can you report a bug about it and someone might fix it :) [07:15] ok. [07:15] bigjools, but it's not successfully install the OS using use-fastpath-install [07:16] bigjools, when the server reboot, it stops "Can not get disk parameter \n boot: " [07:19] rawang: file a bug on curtin about that [07:19] bigjools, ok [07:20] rawang, lets switch back to debian installer === CyberJacob|Away is now known as CyberJacob === jtv2 is now known as jtv [07:58] gmb: do you want your report-machine-info branch reviewed as it stands now? [08:01] jtv: Nope, sorry. I'll make it WIP. [08:02] OK [08:02] I noticed there was a "### REMOVE" in there. [08:25] rvba: morning. [08:25] did you get this when building the dev env on trusty? [08:25] make: bin/pip: Command not found [08:29] bigjools: yes [08:29] rvba: ok. Only just got around to playing with it since I spent all freaking day dealing with broken landers [08:29] (Plus the problem with postresql [make install-dependencies wants to install version 9.1]) [08:30] rvba: yeah see my question on maas-devel [08:30] Yep, saw that. [08:42] hi bigjools, i have tried to use default install (preseed), but the OS installation, it stops at "boot: ", and only "local" is available, the error is "can no get disk parameters", do you have any idea? [08:43] rawang: no idea, sorry. You need a server guy. [08:43] bigjools, i see, thanks :) [09:29] when pserv generate pxelinux.cfg for a node, it use one of the mac address or system_id of the node [09:30] freeflying: it uses the mac address that the pxe agent requests [09:30] bigjools, pxelinux.cfg/01-ac-f2-c5-ed-86-14 like this? [09:30] yes [09:30] bigjools, but why it has 7 numbers [09:31] the 01 is ignored, I forgot why it's there [09:32] bigjools, but why I saw the node request something like pxelinux.cfg/714ec577-d892-4e06-81c6-08674c9452bf [09:32] no idea what that is [09:32] bigjools, and turns out to not found such one, then the node try to get pxelinux.cfg/01-ac-f2-c5-ed-86-14 [09:33] that's fine, it's designed to work like that [09:44] bigjools, btw, guess all the boot parameter will be generated by maas dynamically via chan.c32, am I right === rvba changed the topic of #maas to: Discussion of upstream development of Ubuntu's Metal as a Service (MAAS) tool | Please read http://bit.ly/1j716Fc | MAAS documentation: http://bit.ly/1eIPFAg === rvba changed the topic of #maas to: Discussion of upstream development of Ubuntu's Metal as a Service (MAAS) tool | Please read http://bit.ly/1j716Fc if you're having a problem with a MAAS instance | MAAS documentation: http://bit.ly/1eIPFAg [13:27] hi i have a problem: after i have attached 2 nodes to my MAAS, they stay in the state "commissioning" [13:28] i have already tried to synchronice HW clocks, but this does not solve the problem [13:29] the 2nd suggested possibility, written here: Possible Cause: Network drivers http://maas.ubuntu.com/docs/troubleshooting.html, i dont know if this could be the problem and i dont know which steps i should do to install a "linux friendly network adopter". [13:30] but i guess its sth else [13:33] noone? :( [14:31] a question: where can i find the "Global Kernel parameters" described here: http://maas.ubuntu.com/docs/kernel-options.html [14:31] in my maas web interface i cannot see such a dialog... [15:25] i have a problem: i have attached 2 nodes to my MAAS, but they stay in the state "commissioning" [15:25] the never get to state "ready" [15:32] tomixxx: Are these machines that you have physical access to? [15:33] yes @ gmb [15:33] tomixxx: Can you check whether they're PXE booting from the MAAS server? [15:34] gmx: how exactly do i check this? [15:35] tomixxx: Well, one way would be to attach a screen and see to the machine and see what it's actually doing at boot time. [15:35] tomixxx, I encounter this problem from time to time, which is for me a problem with interface order detection, especially with the fast installer [15:35] i have a screen attached to the computer [15:35] normally, if i boot the node, windows starts ;) [15:35] I alos had this behind of a firewall [15:36] tomixxx: H [15:36] sorry [15:36] tomixxx: How did you get the node into MAAS in the first place? Did you do it manually? [15:37] gmb: i have a usb boot stick with ubuntu server 12.04.03 lts on it [15:37] so i go into boot options, select booting from usb stick and then i chose the option with MAAS and in the following dialoges i could enlist on the MAAS server [15:38] gmb: that's how far i come so far... [15:38] tomixxx: Okay, give me a second... [15:38] gmb: ok [15:39] I have cpu count to 0 after comissionning any suggestion to fix this ? [15:41] tomixxx: What happens if you boot from the USB stick again after you've enlisted the machines? [15:42] @gmb: give me a second, i will try [15:42] i come into the language selection screen [15:44] @gmb: should i select a language and go on? [15:44] @gmb: i mean its the same screen when installing ubuntu [15:45] tomixxx: Did you get the language selection screen before you got to the enlist with MAAS stuff before? [15:45] @gmb: yes [15:45] tomixxx: Okay, then yes, carry on [15:45] ok, selected "english" [15:46] (I haven't used the Avahi stuff myself, and it's sadly not very well documented) [15:46] @gmb: now i can "install ubuntu serveR", "multiple server install with maas" and so on. [15:46] tomixxx: If you follow "multiple server install" what options does it present? [15:47] @gmb: well, after clicking "multiple server install.." i get another language selection screen [15:47] @gmb: but it differs now from style a bit [15:47] ok [15:47] @gmb: i guess i should select english again? [15:48] Yes [15:48] @gmb: btw: at this point, additionally, i have to open a 2nd console and mount the /dev/sdb to /cdrom [15:48] @gmb: if i do not this, ubuntu gives an error, cannot boot from cdrom [15:48] ?! [15:48] How annoying. [15:49] @gmb: its the problem described here: http://askubuntu.com/questions/127398/usb-drive-install-of-ubuntu-12-04-server-fails-cant-find-components-from-cd-r [15:49] Ick. [15:50] @gmb: however, applied the mount-fix, selected english as language and europe->austria as location [15:50] ok [15:50] @gmb: local settings united states - en_US.UTF-8 [15:51] k [15:51] @gmb: after some init procedures, i come to the dialog where i can choose a hostname [15:51] ok [15:52] now, i have three options listed : "ubuntu MAAS Server - (xx.xx.xx.xx:80)", "Specify MAAS by name or address" and "create a new MAAS on this server" [15:52] tomixxx: Are these the same options you saw before? [15:52] @gmb: yes [15:52] Hmm. [15:53] in meanwhile: the node is still "commisionen" from the run before, if i refresh the web-interface on the MAAS-server-node [15:54] tomixxx: Sadly, I don't know how to help you from here because I don't know the avahi-maas stuff at all. Can you file a question on http://askubuntu.com/questions/ask?tags=maas and someone with knowledge of it will lget back to you as soon as they can. [15:55] tomixxx: A node can be "Commissioning" without MAAS actually being able to do anything with it (e.g. if MAAS doesn't know how to cycle change the power state of the node and the node is switched off). [15:56] @gmb: ok [15:56] @gmb: i have alrady set power type to "wake-on-lan" and set mac address [15:57] Hmm. [15:57] tomixxx: Okay. Ask the question anyway and we'll endeavour to get back to you quickly. [15:58] @gmb: ok, thank you very much so far! [16:02] ok, question is online so far, plz let me know if u need further details: http://askubuntu.com/questions/405937/2-added-nodes-to-maas-server-but-their-state-does-not-change-from-commissioning [16:18] tomixxx, Can you check, in the MAAS server, if the power_on task has been triggered? It's in /var/log/celery*.log. I'm not sure but it could be a bug using the wake-on-lan power option. Another thing to bear in mind is that MAAS will probably destroy the contents of your nodes disk, so if you're dual booting the nodes, your windows install might be gone. [16:20] @matsubara: ok, i'll chekc the log [16:20] log line looks something like this: http://pastebin.ubuntu.com/6756972/ [16:21] hmm, i have no such file [16:22] tomixxx, oops, my bad, /var/log/maas/celery.log [16:22] kk [16:22] ok, i ve openend the file now [16:24] and i can find similiar log entry like u have postet indicating "power_on succedded:None" [16:25] tomixxx, so, after the machine enlisted, it automatically turned off, right? Then you accepted it in the MAAS web UI, it changed to commissioning. Did the machine turn on automatically? [16:26] so, that might be the reason (i.e. wake-on-lan not working) [16:26] @matsubare: yes, machine turned off, but never turned on again [16:27] log-file: pastebin.ubuntu.com/6757001/ [16:28] tomixxx, so, what I'd suggest is to first test that wake-on-lan work between your machines. See https://help.ubuntu.com/community/WakeOnLan on how to test it. [16:28] kk [16:30] ok, iam an idiot, i have uploaded the full logfile to pastebin... sorry for long-loading times... [16:40] tomixxx, np, I'll take a look in a bit. I'm in a call right now [17:07] tomixxx, so, I took a look at the log you pasted and it seems to me the problem lies in MAAS trying to power on the node but failing (although the log says the task succeed). I think the code doesn't really check if the node has been power on and the succeed message means that the task has been run without errors. [17:08] tomixxx, so, my suggestion is to try to test wake on lan outside of MAAS first and make sure it works, and then retrying enlistment/commissioning in MAAS [17:09] @matsubara: kk, at the moment, i try to establish "wake-on-lan". i follow this article, since i have only windows pre-installed on the nodes: http://windows7-issues.blogspot.co.at/2011/03/wake-on-lan-wol-for-windows-7-made-easy.html [17:12] tomixxx, you can use the live cd from the usb stick, I guess [17:14] tomixxx, https://help.ubuntu.com/community/Installation/FromUSBStick#live-usb [17:16] ok, ty [17:18] thanks for help so far, iam offline now but ill join this channel tomorrow again === bjf is now known as bjf[afk] === zz_mwhudson is now known as mwhudson === mwhudson is now known as zz_mwhudson === zz_mwhudson is now known as mwhudson === mwhudson is now known as zz_mwhudson === zz_mwhudson is now known as mwhudson