| === menn0 is now known as menn0-afk | ||
| === menn0-afk is now known as menn0 | ||
| === CyberJacc is now known as CyberJacob | ||
| === CyberJacob is now known as Guest84891 | ||
| === bogdanteleaga_ is now known as bogdanteleaga | ||
| === marlinc_ is now known as marlinc | ||
| BlackDex | Hello there. Is it possible to have MAAS assign ip's without having DHCP active on a specific subnet without putting things to static? | 07:52 |
|---|---|---|
| === essembe is now known as sbeattie | ||
| roaksoax | BlackDex: MAAS 1.9 can do static assigments, but it will also provide DHCP for that statically configured IP, for backwards compatibility | 13:13 |
| mup | Bug #1556952 opened: High CPU usage in browser when region is unavailable <ui> <MAAS:Triaged> <https://launchpad.net/bugs/1556952> | 13:55 |
| mup | Bug #1556963 opened: HandlerError suppressed in UI <ui> <MAAS:Triaged> <https://launchpad.net/bugs/1556963> | 14:28 |
| haasn | Is there a way to boot a maas node into a rescue system _without_ releasing it? I accidentally made a change that made it unconnectable via ssh, and redeploying would take a while. I could either wait until I get physical access, or just PXE boot a rescue shell and fix the sshd_config or something | 14:46 |
| haasn | Does maas make this easy? | 14:47 |
| mag009_ | morning | 15:31 |
| mag009_ | is it possible to trigger a build for maas 2.0 | 15:32 |
| mup | Bug #1557030 opened: Digital Logers Power Type Fails Power On <arm64> <digital> <logger> <pdu> <MAAS:New> <https://launchpad.net/bugs/1557030> | 16:16 |
| roaksoax | mattyw: ppa:maas/next | 16:25 |
| roaksoax | err | 16:25 |
| roaksoax | mag009_: ^^ | 16:25 |
| roaksoax | mag009_: or xenial repositories | 16:25 |
| mag009_ | ya that one | 16:26 |
| mag009_ | I just installed using the trunk repo | 16:26 |
| mag009_ | it fixed the following issue : https://bugs.launchpad.net/maas/+bug/1547275 | 16:26 |
| roaksoax | mag009_: yeah, alpha2 is in ppa:maas/next-proposed | 16:31 |
| mag009_ | ah crap wasn't there last week ? | 16:32 |
| mag009_ | i'm sure I checked last thursday | 16:32 |
| mag009_ | :D | 16:32 |
| mag009_ | thanks | 16:32 |
| roaksoax | mag009_: we uploaded it on friday, and haven't made the annoucement yet | 16:32 |
| mag009_ | ah! :) | 16:32 |
| mag009_ | anyway I managed to get it working manually but now I'm facing another issue | 16:33 |
| roaksoax | mag009_: which one is it? | 16:33 |
| mag009_ | seem the commissioning is broken | 16:33 |
| roaksoax | mag009_: https://bugs.launchpad.net/maas/+bug/1555570 | 16:33 |
| mag009_ | when cloud-init call : ?op=get_preseed it return no url | 16:33 |
| mag009_ | 'NoneType' object has no attribute 'url' | 16:34 |
| mag009_ | it call : MAAS/metadata/latest/by-id/4y3h83/?op=get_preseed and the result is that | 16:34 |
| roaksoax | mag009_: or https://bugs.launchpad.net/maas/+bug/1554566 | 16:34 |
| roaksoax | mag009_: ah yea, so it is the latter | 16:34 |
| roaksoax | mag009_: ok, so there's a work around | 16:35 |
| roaksoax | mag009_: go to the Node listing page, click on 'Controllers', then Click on the rack controller | 16:35 |
| mag009_ | ok im there | 16:35 |
| mag009_ | what next? | 16:36 |
| roaksoax | mag009_: on the interfaces section of the rack controller, look for the NIC that has an IP address directly connected to the vlan/subnet your machines are DHCP'ing from | 16:36 |
| mag009_ | yes its there | 16:36 |
| roaksoax | mag009_: ok, what Fabric is it on ? | 16:36 |
| mag009_ | ah change the fabric | 16:36 |
| mag009_ | its duplicated with another | 16:36 |
| roaksoax | mag009_: yeah, change the fabric to Fabric-0 and see what happens ? | 16:36 |
| roaksoax | mag009_: can you attach a screenshot of the original one to the bug though please ? | 16:37 |
| mag009_ | screenshot of ? | 16:37 |
| mag009_ | the controller page | 16:38 |
| roaksoax | mag009_: the interfaces section | 16:38 |
| roaksoax | mag009_: either way, moving that specific interface into Fabric-0 should fix your issue | 16:38 |
| mag009_ | so if i understand correctly it has to be in fabric-0 | 16:40 |
| mag009_ | its just a workaround for now | 16:41 |
| mag009_ | still the same result | 16:41 |
| mag009_ | http://pastebin.com/d71fcLs7 | 16:42 |
| mag009_ | here's the log | 16:42 |
| mag009_ | regiond.log | 16:42 |
| roaksoax | mag009_: strange... the work around works for me but for noone else I've asked to try it | 16:44 |
| roaksoax | mag009_: ok, we'll look into it | 16:44 |
| mag009_ | if you need me let me know I'm also looking at the code to see what the issue is | 16:45 |
| roaksoax | mag009_: thanks! | 16:46 |
| roaksoax | mag009_: oh... did you re-enable DHCP once you moved the interface to Fabric-0 | 16:50 |
| mag009_ | I'm using unmanaged | 16:51 |
| mag009_ | i have a external dhcp that managed the dhcp side | 16:52 |
| mag009_ | and my maas instance is on that network | 16:52 |
| roaksoax | mag009_: ah! that may be an issue since 2.0 won't support external DHCP | 16:52 |
| mag009_ | are you kidding me ? | 16:52 |
| mag009_ | I loved maas because it was bare minimal :) all my dhcp dns is out-of-band | 16:53 |
| mag009_ | our dhcp server are hosted on each top of rack | 16:53 |
| mag009_ | we use cumulus switches | 16:53 |
| roaksoax | right... we no longer perform testing against external DHCP and with this new networking support and such, official support for external DHCP is being dropped. It may still work but we no longer support it | 16:54 |
| mag009_ | but still I don't think it is related to the dhcp I wouldn't be able to enlist my node if that was the case | 17:07 |
| roaksoax | mag009_: no, this issue is not related to DHCP but rather how we tell the node what url to access | 17:08 |
| roaksoax | mag009_: and there's a bug on how we doing that | 17:08 |
| mag009_ | ok found it | 17:22 |
| mag009_ | I've enabled the dhcp in maas for that subnet and it worked | 17:22 |
| mag009_ | now I can access the url | 17:22 |
| mag009_ | this goes against our setup tho | 17:22 |
| roaksoax | mag009_: the bug is there regardless, so it does need to get fixed | 17:29 |
| mag009_ | i can pass that step in the commissioning | 17:34 |
| mag009_ | but it fail near the end | 17:34 |
| mag009_ | found why ... | 17:42 |
| mag009_ | is there a way to not have maas manage the dns ? | 17:42 |
| mag009_ | it uses the controller as nameserver | 17:42 |
| mag009_ | which is wrong... it should be the one I've set on the dhcp | 17:43 |
| mup | Bug #1557100 opened: [2.0a1] Fabric IDs (and thus default names) can get very large if rack continually fails to register <MAAS:Triaged> <https://launchpad.net/bugs/1557100> | 18:28 |
| mup | Bug #1557100 changed: [2.0a1] Fabric IDs (and thus default names) can get very large if rack continually fails to register <MAAS:Triaged> <https://launchpad.net/bugs/1557100> | 18:40 |
| mup | Bug #1557100 opened: [2.0a1] Fabric IDs (and thus default names) can get very large if rack continually fails to register <MAAS:Triaged> <https://launchpad.net/bugs/1557100> | 18:52 |
| mup | Bug #1557103 opened: After having disabled secondary rack, dhcpd fails <MAAS:New> <https://launchpad.net/bugs/1557103> | 18:52 |
| mup | Bug #1557144 opened: [2.0a1] Nodes fail to boot when enlisting with Xenial <MAAS:New> <https://launchpad.net/bugs/1557144> | 19:59 |
| mup | Bug #1554814 changed: [2.0a1]: Can't auto enlist with two Rack Controllers, works with one <MAAS:Invalid> <https://launchpad.net/bugs/1554814> | 22:26 |
| mup | Bug #1557204 opened: maas dnsresources create/update cannot assign a dhcp assigned IP address <MAAS:New> <https://launchpad.net/bugs/1557204> | 22:50 |
| mup | Bug #1557204 changed: maas dnsresources create/update cannot assign a dhcp assigned IP address <MAAS:Won't Fix> <https://launchpad.net/bugs/1557204> | 23:05 |
| mup | Bug #1555570 changed: Trying to commission a machine results into booting to the enlistment environment <MAAS:New> <https://launchpad.net/bugs/1555570> | 23:35 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!