Ask Your Question

Dashboard overview stuck in "loading"

asked 2017-10-06 09:41:51 -0500

jep gravatar image

updated 2017-10-06 09:43:25 -0500

I have a problem that started after the upgrade to Pike. When I click on the dashboard's "Overview", it gets stuck in loading, and the nova-api logs show this line 4 times per second in a loop, over and over:

2017-10-06 09:42:18.170 7383 INFO nova.osapi_compute.wsgi.server [req-9cf0dd87-bfec-40eb-ad50-bb0aa646b822 2bcbe1b9694d45d0b2114d3d18408be8 eb43f6dfa2764e2eb8bb8d4d95e1b1e0 - default default] "GET /v2.1/eb43f6dfa2764e2eb8bb8d4d95e1b1e0/os-simple-tenant-usage?start=2017-10-05T00:00:00&end=2017-10-06T23:59:59&marker=4a1b34f5-9fdd-4204-a527-e873c558b679&detailed=1 HTTP/1.1" status: 200 len: 4606 time: 0.2725942

After restarting openstack-nova-api, dashboard shows the message:

Error: Unable to retrieve usage information.
Error: Unable to retrieve limit information.

These lines from the CLI work ok though:

openstack usage show --project admin --start 2017-10-05 --end 2017-10-06
openstack usage list --start 2017-10-05 --end 2017-10-06

(Not sure how to specify an instance in those commands, to emulate the "marker" part of the URL in the loop)

Any help on troubleshooting this issue will be greatly appreciated!

[Running Pike on CentOS 7, installed manually, keystone, horizon, nova, neutron, cinder, swift, glance, ceph]

edit retag flag offensive close merge delete

2 answers

Sort by ยป oldest newest most voted

answered 2017-10-06 12:11:49 -0500

jep gravatar image

I found the problem. It was related to the cells setup. Somehow I had 2 cells in addition to cell0, causing all kinds of issues, including my hypervisor-list showing each hypervisor twice.
I few commands that helped troubleshoot it were:

nova-manage cell_v2 list_cells
nova-manage cell_v2 map_instances --cell_uuid

edit flag offensive delete link more

answered 2017-10-27 02:25:20 -0500

yujian gravatar image

I also encountered the same problem, and finally found the reason:

After installing nova-compute, the 'nova-status upgrade check' was executed on the controller node, and the 'nova-manage cell_v2 simple_cell_setup' was executed according to the check prompt, which eventually resulted in the above results.

The normal steps are:

nova-manage cell_v2 discover_hosts

nova-status upgrade check

Do not need to execute 'nova-manage cell_v2 simple_cell_setup'

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2017-10-06 09:41:51 -0500

Seen: 993 times

Last updated: Oct 27 '17