[01:45] Bug #1674864 opened: MAAS deployment failed due to data source user-data 500 Internal Server Error, after using set default gateway [02:45] Bug #1674864 changed: MAAS deployment failed due to data source user-data 500 Internal Server Error, after using set default gateway [02:45] Bug #1674866 opened: Default gateway (per-subnet) should have a metric value === frankban|afk is now known as frankban === MrPekka_ is now known as MrPekka [08:58] Hello [11:07] is there anyone active in this channel? [11:08] ive logged in multiple time and see people online but no one ever types any messages [11:09] bryanb229, it's active [11:09] some times are quieter than others [11:18] Bug #1674959 opened: [UI] Close Add Pod button is redundant on pods page [11:48] Bug #1674959 changed: [UI] Close Add Pod button is redundant on pods page [12:00] Bug #1674959 opened: [UI] Close Add Pod button is redundant on pods page [12:03] Bug #1674959 changed: [UI] Close Add Pod button is redundant on pods page [12:52] Bug #1674991 opened: [2.1] Cannot skip 2nd stage of intro config journey [13:04] Bug #1674991 changed: [2.1] Cannot skip 2nd stage of intro config journey [13:43] Bug #1675017 opened: [CLI] interface link-subnet command cannot refer to interface by device name [13:49] Bug #1675017 changed: [CLI] interface link-subnet command cannot refer to interface by device name [13:58] Bug #1675017 opened: [CLI] interface link-subnet command cannot refer to interface by device name [14:46] Bug #1670499 changed: maas 2.1.3 nodes do not deploy, Juju reports 'started but not deployed'

[14:46] Bug #1671605 changed: Unbootable system after installation [14:46] Bug #1673726 changed: scsi ids with vmware are not correctly used [16:07] Hello all, is there a way to debug the commissioning process in 2.1? I've been stuck for a couple days and about to call it quits :/ [16:09] dfin, logs are a good place to start [16:09] Right, but I can't log in to the commissioned machine and I couldn't get the backdoor hack to work [16:10] Specifcially, I can see in the MAAS logs that the initial PXE is going well, but the node I'm trying to provision just sits there at a login prompt [16:11] dfin, and the web UI says 'Commissioning failed'? [16:11] yes, after it hits the 20 min timout [16:12] dfin, what kind of machine is powering the node? bare-metal, virtual [16:13] BM - Dell 6730xd [16:13] *r730 [16:15] MAAS itself is in a VMware cluster on the same vlan [16:15] dfin, what power type was chosen? [16:15] IPMI 2, which is working [16:16] At least, it powers on the machine [16:16] and I can make it power the machine off via the web ui [16:17] when will the maas patch released into beta release for the disk boot order [16:18] there is a patch available and i need it [16:18] https://bugs.launchpad.net/maas/+bug/1673724 [16:19] dfin, any firmware updates available for the BMC? [16:20] pmatulis: I don't think so - let me check [16:21] pmatulis: I am on the latest firmware version [16:24] I _suspect_ that something is up with the scripts run on squashfs, but I can't get in there to check. Just a suspicion though... [16:30] dfin, i'm sorry i don't know what else i can add. hopefully someone else here can chime in. i wouldn't give up just yet though. there's a good reason why it's not working [16:30] dfin, do you have other machines that work? [16:30] pmatulis: no, this is the pilot machine [16:30] dfin, maybe try a quick KVM-backed or VMware-backed node [16:31] to see if there's a wider problem [16:31] I'll do that [16:37] pmatulis: whatever the case, thanks for your help [16:40] yeah, same thing with a VMware node. I'll see if it times out [16:40] is DNS working? pretty sure commissioning requires it [16:42] DNS is what way. Like the MAAS DNS? [16:42] I have a working DNS server for our internal domains [16:44] nodes should be able to contact the internet and have DNS working. network-wise. i'll be back later [16:44] sure [16:46] check the proxy logs during commissioning. you should see requests go out [16:46] i'm pretty sure [16:51] is someone from the devs here? [16:52] ybaumy: Yes, though they might be out due to different TZ. [16:53] zeestrat: i need a patch for maas beta applied [16:53] https://bugs.launchpad.net/maas/+bug/1673724 [16:54] when can i expect a release for this [16:55] ybaumy, that's going to be in the next release, out soon [16:56] brendand: thats great to hear because its really important. would i get that earlier if i had a support aggreement with canonical? [17:05] ybaumy, if you're willing to use a ppa you can get it now but would have to accept that it hasn't been tested fully yet [17:09] brendand: im using devel [17:09] ybaumy, which ppa? [17:10] brendand: http://ppa.launchpad.net/maas/next/ubuntu xenial main [17:13] Bug #1675095 opened: [2.2] MAAS selecting the incorrect NTP server for deployed machines [17:29] jamespage: ii'll ask here, if i delete NICs in maas, can i get them back later without knowing the MAC addresses? === frankban is now known as frankban|afk [18:37] Bug #1675123 opened: [2.2, UI] Interfaces UI shows two items for the subnet [18:46] Bug #1675123 changed: [2.2, UI] Interfaces UI shows two items for the subnet [18:58] Bug #1675123 opened: [2.2, UI] Interfaces UI shows two items for the subnet [20:49] Does anyone have any experience deploying coreos with maas? [21:10] Bug #1675123 changed: [2.2, UI] Interfaces UI shows two items for the subnet [21:36] smgoller: maas doesn't support coreos [21:42] roaksoax: thanks [21:43] so I notice when juju creates containers on machines it allocates via maas, the containers show up on the maas node details page. Is there a way to manage that outside of juju? [21:43] I guess the simple question is, if I create lxd containers on maas-deployed machines, will they show up? [21:51] smgoller: if you register them in MAAS they would [21:51] smgoller: what juju does is basically " [21:51] "hey maas, register this container that I've just created and give it an static IP address" [21:52] smgoller: and maas adds a container, and using the mac it provides it an "static" IP address that's delivered by DHCP (or juju configures it statically) [21:52] if that makes sense [21:52] so maas is doing the actual container creation [21:53] er [21:53] sorry [21:53] * smgoller goes to look at the juju maas drivers [21:53] smgoller: juju is doing the container creation and telling MAAS about the existance of the container [21:54] right [21:56] any info on how juju tells maas about the container? [21:57] smgoller: it just makes a API call. I'd think it adds a device with a parent [21:57] smgoller: and then asks for an IP with link-subnet [22:22] I have two subbnets on the same untagged fabric. Although deploying works fine giving it the right auto-assigned address, comission fails because it picks an dynamic IP in the opposite subnet than its in [22:22] the result is that it gets its DHCP response and address during comission, but it cannot connect, because its the wrong address for that subnet [22:23] I am I running into limitations, or bugs? [22:36] looks like it might have worked when I seperated that second subnet into its own fabric [22:47] Good evening everyone! [22:49] Quick question, what kind of "scripts" the commissioning scripts can be? As they seems to be launched by cloud-init, I would say nearly anyone but I need to be sure as I want to use it to manage the Hardware RAID configuration. [23:03] anyone?