0

I have spent days on this. It's likely something simple. I've tried booting from ISO and creating MAAS, I've tried clean installing and following the openstack instructions for a clean install of ubuntu. Never any luck. When I go to commission a node, the machine turns on, and sits at commissioning. When I look at the node it PXE boots but that's about it. I've confirmed DNS appears to be ok (though there is no dns entry for the node-1 but a query for maas returns the correct IP) This is all running under vSphere 6.

First and foremost... here are the logs from the node:

cloud-init.log: http://pastebin.com/4XbcnbZB

clout-init.output.log: http://pastebin.com/nWZEc56r

Here are the warnings I see in cloud-init.log:

2016-01-03 22:00:01,156 - util.py[WARNING]: Failed accessing user data.

2016-01-03 22:00:03,398 - util.py[WARNING]: Running seed_random () failed

2016-01-03 22:00:03,523 - cc_resizefs.py[WARNING]: Device 'overlayroot' did not exist. cannot resize: dev=overlayroot mnt_point=/ path=/

Now for the controller... I see a bunch of WARNINGS in maas.log:

Jan 3 15:52:47 head maas.start-up: [WARNING] Database error during start-up (PostgreSQL error 40P01); pausing for 3 seconds.

Jan 3 15:52:47 head maas.import-images: [WARNING] I/O error while syncing boot images. If this problem persists, verify network connectivity and disk usage.

Jan 3 15:52:47 head maas.import-images: [WARNING] Finished importing boot images, the region does not have any boot images available.

Jan 3 15:52:49 head maas.start-up: [WARNING] Database error during start-up (PostgreSQL error 40P01); pausing for 3 seconds.

Jan 3 15:57:42 head maas.import-images: [WARNING] I/O error while syncing boot images. If this problem persists, verify network connectivity and disk usage.

Jan 3 15:57:42 head maas.import-images: [WARNING] Finished importing boot images, the region does not have any boot images available.

Jan 3 16:02:41 head maas.import-images: [WARNING] I/O error while syncing boot images. If this problem persists, verify network connectivity and disk usage.

Jan 3 16:02:41 head maas.import-images: [WARNING] Finished importing boot images, the region does not have any boot images available.

Now on the MAAS web page, it shows a valid image. On the clusters page it says connected and images are synced so those warnings could just be boot warnings? I'm really not sure what to try next.

0

This seems like a network communication issue. Looking at cloud-init logs:

Jan 3 22:00:03 ubuntu pollinate[750]: ERROR: Network communication failed [60]\n22:00:02.333112 * Hostname was NOT found in DNS cache

Also, looking at maas.log:

Jan 3 16:02:41 head maas.import-images: [WARNING] I/O error while syncing boot images. If this problem persists, verify network connectivity and disk usage.

If you are behind a proxy, this could be an issue. Can you verify that you have access to archive.ubuntu.com as well?

  • I noticed that too - I can access archive.ubuntu.com. I installed a package on node-1 just to make sure. In the DNS on the controller, the only entry is .maas which correctly points to the controller. node-1 and the controller itself (named head.maas) are not there. I'm not sure if they are supposed during commissioning. No proxy. – Huza Jan 4 '16 at 4:51
  • Also to add to the network config - it's all on a esxi virtual switch (not distributed and no physical adapter connected to it). The switch is set to allow promiscuous mode, MAC address changes, and forged transmits just to make sure. – Huza Jan 4 '16 at 4:57
  • Ran a constant ping as well as a file transfer from head to node for several hours. Nothing unexpected so I don't think it's an all-out network error. The pollinate error I believe is related to bugs.launchpad.net/maas/+bug/1424549 - SSL cert error during that process does show in cloud-init-output. – Huza Jan 5 '16 at 21:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.