Ask Your Question
0

Glance Times out on Image-create

asked 2015-09-29 09:48:34 -0600

sdub gravatar image

Trying to follow these instructions: http://docs.openstack.org/openstack-ops/content/user_facing_images.html (http://docs.openstack.org/openstack-o...)

I have a deployment of icehouse on vagrant VMs. Whenever I try and add a cirros os image to glance it waits 6 minutes and then times out with this error:

Error communicating with http://192.168.122.10:9292 timed out

Here's the logging information for that one command:

2015-09-29 10:23:41.328 17969 INFO glance.wsgi.server [-] Starting 4 workers
2015-09-29 10:23:41.334 17969 INFO glance.wsgi.server [-] Started child 17978
2015-09-29 10:23:41.340 17978 INFO glance.wsgi.server [-] (17978) wsgi starting up on http://0.0.0.0:9292/
2015-09-29 10:23:41.343 17969 INFO glance.wsgi.server [-] Started child 17979
2015-09-29 10:23:41.350 17979 INFO glance.wsgi.server [-] (17979) wsgi starting up on http://0.0.0.0:9292/
2015-09-29 10:23:41.356 17969 INFO glance.wsgi.server [-] Started child 17980
2015-09-29 10:23:41.357 17980 INFO glance.wsgi.server [-] (17980) wsgi starting up on http://0.0.0.0:9292/
2015-09-29 10:23:41.363 17969 INFO glance.wsgi.server [-] Started child 17981
2015-09-29 10:23:41.372 17981 INFO glance.wsgi.server [-] (17981) wsgi starting up on http://0.0.0.0:9292/
2015-09-29 10:24:45.399 17981 INFO urllib3.connectionpool [-] Starting new HTTP connection (1): 192.168.122.10
2015-09-29 10:24:45.984 17981 INFO urllib3.connectionpool [-] Starting new HTTP connection (1): 192.168.122.10
2015-09-29 10:24:46.055 17981 INFO urllib3.connectionpool [-] Starting new HTTP connection (1): 192.168.122.10
2015-09-29 10:24:46.060 17981 INFO keystoneclient.middleware.auth_token [-] Auth Token confirmed use of v2.0 apis
2015-09-29 10:24:46.061 17981 INFO urllib3.connectionpool [-] Starting new HTTP connection (1): 192.168.122.10
2015-09-29 10:24:46.068 17981 INFO urllib3.connectionpool [-] Starting new HTTP connection (1): 192.168.122.10

I can share my glance configuration but it's pretty basic, so I'll forgo that for now. If I use nc to test the address that's timing out, it works:

[root@vbit00 vagrant]# nc -zv 192.168.122.10 9292
Connection to 192.168.122.10 9292 port [tcp/armtechdaemon] succeeded!

However, if I try and ping the whole hostname I get this:

[root@vbit00 vagrant]# [root@vbit00 vagrant]# ping http://192.168.122.10:9292
ping: unknown host http://192.168.122.10:9292

Thinking it may be a routing problem, I just flushed my entire iptables but that didn't help either. I found out what was prepending the http:// to the controller address (it was keystone_init.py), and I removed the http:// but that didn't work either.

Some more interesting output:

[root@vbit00 vagrant]# netstat -antp | grep 9292
tcp        0      0 0.0.0.0:9292                0.0.0.0:*                   LISTEN      18460/python
[root@vbit00 vagrant]# ps -ef | grep  18460
glance   18460     1  3 10:45 ?        00:00 ...
(more)
edit retag flag offensive close merge delete

Comments

are you using ceph or filesystem for image saving?

Moe gravatar imageMoe ( 2015-09-29 21:48:51 -0600 )edit

I am using ceph.

sdub gravatar imagesdub ( 2015-09-30 08:13:24 -0600 )edit

1 answer

Sort by ยป oldest newest most voted
0

answered 2015-09-29 11:05:18 -0600

sdub gravatar image

The problem existed with ceph. Neither of my OSDs were connecting to my monitor so I had to go and sort that out. Turned out my bootstrap keyring wasn't the same so I had to fix that. I did that by following this: http://lists.ceph.com/pipermail/ceph-users-ceph.com/2014-February/037521.html (http://lists.ceph.com/pipermail/ceph-...)

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2015-09-29 09:48:34 -0600

Seen: 314 times

Last updated: Sep 29 '15