Take the 2-minute tour ×
Ask Ubuntu is a question and answer site for Ubuntu users and developers. It's 100% free, no registration required.

juju bootstrap starts up the node.

The node loads, and then nothing. I get a juju status of no reachable server, it's like MaaS does not report to Juju that the node is ready for Juju. MaaS UI does show the node as allocated.

That is where it stays until I issue juju destroy-environment, which MaaS executes to the letter, shutting down the node and then shows the node as "ready" in the MaaS UI. I waited an hour and watched the node load before running the destroy command.

error details
no reachable server

structure: i have the maas-region and cluster running off one dell 1800 running ubuntu 12.04.4 server. eth0 is internet connected @ xxx.xxx.xxx.100,to the public switch (which gives me xxx.xx.xxx.100/MAAS and full internet access) eth1 is connected to private none managed gigabit switch, eth1 set to xx.xx.xx.1., eth1 is set to manged with maas dns/dhcp hosting the all 10 dell 1950 gen 2s. servers get internal addressing.

all machines commission and enlisted and become ready normally.

i install juju client on a machine operating in the public side on the connection, running ubuntu 12.4.4 64 bit

i can access the maas ui, and everything is looks normal. then the juju client from configuration to access becomes overwhelming.

1st errors come when for some reason juju tries to get the needed tools from amazomaws. (remember i am in maas). the 2nd is that if i don't tell it where to find the authorized keys it does not go. my environments.yaml is

default: maas environments: maas: type: maas maas-server: 'http://xxx.xxx.xxx/MAAS/' maas-oauth: '3Ey3RueWP2X4asFahJ:CmR3fPYaTBwxhvx87X:Sample' admin-secret: '' authorized-keys-path: /xxx/xxxxx/id_xxx.pub

after all correction i just get the above described.

share|improve this question
    
Sounds like a bug to me. AU is the place for questions. The place for bugs is launchpad or IRC... –  hbdgaf May 25 at 1:56
    
This sounds like a configuration error, not a bug (yet) –  Jorge Castro May 25 at 2:47
    
Could you be more descriptive about how to reproduce the error? Step by step if possible, and also add details about your topology. –  Braiam May 25 at 3:42
    
one last thing..there is and error on the node..reads 2014-05-25 14:00:00 ERROR juju.state open.g:93 TLS handshake failed: x509: certificate has expired or is not yet valid...this repeats it self every second until environments is destroy.. –  JUAN CABALLERO May 25 at 18:25
    
This sounds suspiciously like one of the timing issues. MAAS, like a lot of network software, is very dependent on time signatures - if the clocks on the nodes vary by more than 5 minutes, you get all sorts of errors because the clocks don't match closely enough. The later versions of MAAS address this by updating the clocks as part of the commissioning process, but that would be the first thing I would check. P.S. Which version of MAAS are you running? –  Nick Veitch Jun 2 at 14:22

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.