| bradm | I'm looking at the backup recommendations for maas, and it says to stop postgresql, maas-{dhcp,regiond,rackd} during the backups, is that just to make sure nobody is making changes to it while you back up? seems tricky to backup the db with it down, too :) | 01:27 |
|---|---|---|
| maestropandy | Hi all, while deploying maas node, it got failed, how to investigate and troubleshoot, i read in maas documents asked to get into rescue mode ? | 05:52 |
| jose-phillips | maestropandy commision goes right? | 07:23 |
| maestropandy | jose-phillips: commision goes well | 07:27 |
| jose-phillips | your server have some special configuration | 07:28 |
| jose-phillips | ? | 07:28 |
| jose-phillips | or just a normal server with hard drive | 07:28 |
| maestropandy | in deployment its machine is restarting multiple times which i can able to see in console | 07:28 |
| maestropandy | its normal boot from pxe thats no special | 07:28 |
| jose-phillips | on the logs | 07:29 |
| jose-phillips | inside maas you dont have any message? | 07:29 |
| maestropandy | i have checked there, no error logs | 07:29 |
| maestropandy | but node is restarting multiple times | 07:30 |
| jose-phillips | ok try selection the server | 07:31 |
| jose-phillips | on actions | 07:31 |
| jose-phillips | select rescue mode | 07:31 |
| jose-phillips | turn on the machine and try to access in ssh | 07:32 |
| maestropandy | now node got failed deployment, shall i do rescue mode now ? | 07:33 |
| jose-phillips | yep | 07:33 |
| maestropandy | failed deployment now allowing to rescue mode | 07:33 |
| maestropandy | i think cant take rescue action on failed deployment node | 07:34 |
| jose-phillips | at least on maas 2.3 yes | 07:34 |
| jose-phillips | i already tried | 07:34 |
| maestropandy | have u tried with failed deployment node or while deploying, please confirm | 07:35 |
| jose-phillips | when the machine flag deployment failed | 07:35 |
| jose-phillips | i set on rescue mode | 07:35 |
| jose-phillips | and is booting up on rescue mode | 07:35 |
| maestropandy | mine is 2.2.2 version | 07:35 |
| maestropandy | shall i reset node via console and do ? | 07:36 |
| jose-phillips | but your node is on "failed deployment" | 07:36 |
| jose-phillips | or in "deployment" | 07:36 |
| maestropandy | now my node is one failed deployment | 07:37 |
| jose-phillips | and if you select rescue mode what happend? | 07:37 |
| maestropandy | the selected node cannot be resuce mode, please update your selection | 07:37 |
| jose-phillips | mhmm i already tried on 2.3 and works | 07:38 |
| maestropandy | any other option | 07:40 |
| maestropandy | ?? | 07:41 |
| jose-phillips | flag broken node | 07:41 |
| jose-phillips | and try to rescue mode | 07:41 |
| maestropandy | okay | 07:41 |
| jose-phillips | acording the documentation im right | 07:41 |
| jose-phillips | 2.2 can just do on deployed or broken node | 07:42 |
| maestropandy | now its working | 07:42 |
| maestropandy | took rescue mode | 07:42 |
| jose-phillips | in 2.3 | 07:42 |
| jose-phillips | apply for failed node | 07:42 |
| maestropandy | after entering rescue mode next what to do ? | 07:42 |
| jose-phillips | idk if your hard drive will be mounted or not | 07:43 |
| jose-phillips | i recommend you see if you have something on /var/log/syslog on your hard drive | 07:43 |
| jose-phillips | to see if something is failing there | 07:43 |
| jose-phillips | wbat version of os are you trying to install | 07:43 |
| maestropandy | no what am asking is now status in "entering rescue mode" on MAAS UI, shall i go to MAAS server and see that system logs / | 07:44 |
| jose-phillips | no i mean on the server that is failing | 07:44 |
| maestropandy | mean see via console ? we have rsyslog configured with MAAS, so from maas server, i can see syslog of that server | 07:45 |
| maestropandy | how long status will be in resuce mode, still MAAS UI showing, "entering rescue mode" | 07:45 |
| jose-phillips | i don't think so you are running on emphemeral disk | 07:46 |
| maestropandy | u mean ubutu right | 07:46 |
| jose-phillips | yep | 07:46 |
| jose-phillips | is ubuntu right? | 07:46 |
| jose-phillips | i recommend you check the servers log in console via ssh | 07:46 |
| jose-phillips | at least is what i try to do | 07:46 |
| maestropandy | yes i am seeing the same through rsyslog collected on MAAS server | 07:47 |
| maestropandy | also for another node i found "IPMI reachalbity issue " for that whether firmware upgrade will do needful how to resolve in that case | 07:48 |
| maestropandy | jose-phillips: in rescue mode logs, i can see "timed out waiting for reply fom ntp.ubuntu.com | 07:49 |
| maestropandy | jose-phillips: found, "hardware error" with CPU | 07:56 |
| jose-phillips | thats sound more reasonable | 07:58 |
| jose-phillips | is all that you have | 07:59 |
| jose-phillips | is not more detailed? | 07:59 |
| jose-phillips | another question when the node is "deploying" after restart just stuck on grub? | 08:00 |
| === frankban|afk is now known as frankban | ||
| mup | Bug #1739761 opened: Unable to deploy Precise on MaaS 2.3.0 (6434) <MAAS:New> <https://launchpad.net/bugs/1739761> | 11:53 |
| mup | Bug #1739761 changed: Unable to deploy Precise on MaaS 2.3.0 (6434) <MAAS:Won't Fix> <https://launchpad.net/bugs/1739761> | 12:56 |
| === frankban is now known as frankban|afk | ||
| jose-phillips | hi | 17:39 |
| jose-phillips | guys | 17:39 |
| jose-phillips | question | 17:39 |
| jose-phillips | is posible modify a tag? | 17:39 |
| jose-phillips | i creat a tag for kernel options per node | 17:39 |
| jose-phillips | but i need change the kernel_opts arguments on the tag this is posible? | 17:39 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!