Helion OpenStack Community installation fails after timing out

asked 2014-05-14 23:04:14 -0600

Naveen Chandrashekar gravatar image

After successfully creating tripleO directory withing root's home directory, was able to successfully start seed vm by executing "HP_VM_MODE=y bash -x ~root/tripleo/tripleo-incubator/scripts/hp_ced_start_seed.sh" command.

Then was able to login to the seed VM by executing "ssh".

Withing the seed VM , executed "bash -x ~root/tripleo/tripleo-incubator/scripts/hp_ced_installer.sh" command to start with undercloud and overcloud deployment. During the process of deployment, noticed that there were no issues while undercloud deployment was in progress, but at the end of overcloud deployment process, encountered the below timimng out error.

++ '[' 0 -eq 0 ']'
++ break
++ set -e
++ wait_for 30 10 nova service-list --binary nova-compute '2>/dev/null' '|' grep 'enabled.*\ up\ '
Timing out after 300 seconds:
COMMAND=nova service-list --binary nova-compute 2>/dev/null | grep enabled.*\ up\

Has anyone come across similar issue and got it resolved?

edit retag flag offensive close merge delete


Did you try to increase the timeout value eg: from 300 to 600 and try again to see if this issue is gone?

9lives gravatar image9lives ( 2014-05-15 09:54:21 -0600 )edit

I am unable to figure out as to where I can modify the time out value in the script.

Naveen Chandrashekar gravatar imageNaveen Chandrashekar ( 2014-05-16 02:05:14 -0600 )edit

It is not the timeout... see below.

++ glance image-create --name user --public --disk-format qcow2 --container-format bare --file /root/tripleo/user.qcow2
| Property         | Value                                |
| checksum         | ba56450f1d7e13e7bbf492e77a1bfdf1     |
| container_format | bare                                 |
| created_at       | 2014-05-19T20:50:15                  |
| deleted          | False                                |
| deleted_at       | None                                 |
| disk_format      | qcow2                                |
| id               | bee49368-97f0-465d-8b62-f8838663475f |
| is_public        | True                                 |
| min_disk         | 0                                    |
| min_ram          | 0                                    |
| name             | user                                 |
| owner            | 9218c546b1d7445cb653289a37b4b957     |
| protected        | False                                |
| size             | 245010944                            |
| status           | active                               |
| updated_at       | 2014-05-19T20:51:23                  |
| virtual_size     | None                                 |
++ '[' 0 -eq 0 ']'
++ break
++ set -e
**++ wait_for 300 10 nova service-list --binary nova-compute '2>/dev/null' '|' grep 'enabled.*\ up\ '
Timing out after 3000 seconds:**
COMMAND=nova service-list --binary nova-compute 2>/dev/null | grep enabled.*\ up\
richard.murphy@hp.com gravatar imagerichard.murphy@hp.com ( 2014-05-19 17:45:53 -0600 )edit

ok, from your comment, the nova service-list command has been timeout, could you try again and login the seed vm then execute the command "nova --debug service-list --binary nova-compute" manually to see what happened? I guess it might be the proxy issue that blocks the keystone interaction, just make a check.

9lives gravatar image9lives ( 2014-05-19 22:49:17 -0600 )edit

Hmm... weird.

    root@hLinux:~# nova --debug service-list --binary nova-compute
DEBUG (shell:792) You must provide a username or user id via --os-username, --os-user-id, env[OS_U          SERNAME] or env[OS_USER_ID]
Traceback (most recent call last):
  File "/opt/stack/python-novaclient/novaclient/shell.py", line 789, in main
  File "/opt/stack/python-novaclient/novaclient/shell.py", line 616, in main
    raise exc.CommandError(_("You must provide a username "
CommandError: You must provide a username or user id via --os-username, --os-user-id, env[OS_USERN          AME] or env[OS_USER_ID]
ERROR (CommandError): You must provide a username or user id via --os-username, --os-user-id, env[          OS_USERNAME] or env[OS_USER_ID]
richard.murphy@hp.com gravatar imagerichard.murphy@hp.com ( 2014-05-20 08:27:20 -0600 )edit

3 answers

Sort by ยป oldest newest most voted

answered 2014-05-20 09:23:56 -0600

9lives gravatar image

From what the nova --debug service-list --binary nova-compute shows the script cannot find the necessary env variables to interact with keystone, please make sure the OS_USERNAME and OS_PASSWORD are correctly set.

Hope that helps!


edit flag offensive delete link more


The env variables are set inside of the shell script and do not survive after the script exits. One of the Overcloud nodes always starts, reboots, then never starts again.. wondering if that has somethign to do with it.

richard.murphy@hp.com gravatar imagerichard.murphy@hp.com ( 2014-05-21 08:08:14 -0600 )edit

answered 2014-06-22 04:18:38 -0600

Peter Toft gravatar image

The issue with installation of the virtual preview on AMD processors will be fixed in the next release of the virtual installer. The specific (Nova) issues are:

Defect: https://bugs.launchpad.net/nova/+bug/... Fix: https://review.openstack.org/#/c/96148/

The date of the next virtual preview release is TBD, but will probably be in July.

Thanks for reporting and characterizing this issue.

edit flag offensive delete link more

answered 2014-06-04 10:25:42 -0600

cswiger gravatar image

updated 2014-06-06 19:02:00 -0600

Has there been a patch or fix yet? We have installed a fresh Ubuntu 14.04LTS, ran the scripts in the instructions and get to the same stopping point.

6/6 - verified it's an amd thing - did an install on xeon and works just fine! The above was on an HP Opteron workstation xw9400 w/ 64gb memory and 12 cores.

DevStack works on the amd.

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools



Asked: 2014-05-14 23:04:14 -0600

Seen: 786 times

Last updated: Jun 22 '14