OpenStack Compute (nova)

CRITICAL nova [-] [Errno 13] Permission denied: '/usr/lib/python2.6/site-packages/nova/..//keys'

Asked by Truong Pham

I just installed Diablo on CentOS6 and getting permission denied: '/usr/lib/python2.6...

Nova-network.log:
2012-01-27 11:31:19,285 CRITICAL nova [-] failed to create /usr/lib/python2.6/site-packages/bncloud01.MainThread-5636
2012-01-27 11:39:27,286 AUDIT nova [-] Starting network node (version 2011.3-workspace:tarmac-20110727233925-40wmpvkvas7j1p93)
2012-01-27 11:39:27,286 CRITICAL nova [-] failed to create /usr/lib/python2.6/site-packages/bncloud01.MainThread-6090
2012-01-27 11:46:36,779 AUDIT nova [-] Starting network node (version 2011.3-workspace:tarmac-20110727233925-40wmpvkvas7j1p93)
2012-01-27 11:46:36,780 CRITICAL nova [-] failed to create /usr/lib/python2.6/site-packages/bncloud01.MainThread-6485

Nova-api.log:

2012-01-27 11:46:34,625 AUDIT nova [-] zone_name : nova
2012-01-27 11:46:34,625 AUDIT nova [-] memcached_servers : None
2012-01-27 11:46:34,625 AUDIT nova [-] logging_default_format_string : %(asctime)s %(levelname)s %(name)s [-] %(message)s
2012-01-27 11:46:34,625 AUDIT nova [-] fake_rabbit : False
2012-01-27 11:46:34,626 AUDIT nova [-] rabbit_host : localhost
2012-01-27 11:46:34,626 AUDIT nova [-] use_syslog : False
2012-01-27 11:46:34,626 AUDIT nova [-] osapi_path : /v1.0/
2012-01-27 11:46:34,626 AUDIT nova [-] ec2_path : /services/Cloud
2012-01-27 11:46:34,626 AUDIT nova [-] rabbit_use_ssl : False
2012-01-27 11:46:34,626 AUDIT nova [-] rabbit_retry_interval : 10
2012-01-27 11:46:34,627 AUDIT nova [-] node_availability_zone : nova
2012-01-27 11:46:34,627 AUDIT nova [-] volume_manager : nova.volume.manager.VolumeManager
2012-01-27 11:46:34,627 AUDIT nova [-] start_guests_on_host_boot : False
2012-01-27 11:46:34,627 AUDIT nova [-] rpc_thread_pool_size : 1024
2012-01-27 11:46:34,627 AUDIT nova [-] api_paste_config : api-paste.ini
2012-01-27 11:46:34,627 AUDIT nova [-] rabbit_virtual_host : /
2012-01-27 11:46:34,628 AUDIT nova [-] ajax_console_proxy_port : 8000
2012-01-27 11:46:34,628 AUDIT nova [-] image_service : nova.image.glance.GlanceImageService
2012-01-27 11:46:34,628 AUDIT nova [-] control_exchange : nova
2012-01-27 11:46:34,628 AUDIT nova [-] compute_manager : nova.compute.manager.ComputeManager
2012-01-27 11:46:34,628 AUDIT nova [-] network_topic : network
2012-01-27 11:46:34,852 CRITICAL nova [-] [Errno 13] Permission denied: '/usr/lib/python2.6/site-packages/nova/..//keys'

Question information

Language:
English Edit question
Status:
Solved
For:
OpenStack Compute (nova) Edit question
Assignee:
No assignee Edit question
Solved by:
Vish Ishaya
Solved:
Last query:
Last reply:
Revision history for this message
Truong Pham (truongdpham) said : 		#1

So if I change from nova:nobody to root:root in /etc/init.d/openstack-nova-api:

/sbin/start-stop-daemon --start -b -c root:root --make-pidfile --pidfile $pidfile --exec /usr/bin/nova-$suffix -- --flagfile=$flgfile --logfile=$logfile --pidfile $pidfile

then it would start up....

Is it a bug?

Thanks,
-TP

Revision history for this message
Vish Ishaya (vishvananda) said : 		#2

try setting --state_path and --lock_path to writable directories.

They default to where the code is located, but you probably want something like:

--logdir=/var/log/nova
--state_path=/var/lib/nova
--lock_path=/var/lock/nova

(make sure that the directories exist and are writable by the nova user

Revision history for this message
Truong Pham (truongdpham) said : 		#3

Thanks for the response Vish!

I already have those setting in nova.conf:

# egrep 'logdir|state-path|lock_path' nova.conf
--logdir=/var/log/nova
--lock_path=/var/lib/nova/tmp
# egrep 'logdir|state_path|lock_path' nova.conf
--logdir=/var/log/nova
--state_path=/var/lib/nova
--lock_path=/var/lib/nova/tmp
[root@bncloud01 nova]# ls -ld /var/lib/nova/; ls -ld /var/lib/nova/tmp
drwxr-xr-x 10 nova nobody 4096 Jan 27 12:56 /var/lib/nova/
drwxr-xr-x 2 nova nobody 4096 Jan 27 12:56 /var/lib/nova/tmp

I actually had this working on before on the test machine. I repeated the same process and running into this issue...
Weird?
-TP

Revision history for this message
Best Vish Ishaya (vishvananda) said : 		#4

Are you sure the binaries are finding your nova.conf?

Maybe it isn't readable by your user or isn't in the right directory?
You can force it with --flagfile=/path/to/nova.conf when you start the service

On Jan 27, 2012, at 1:35 PM, Truong Pham wrote:

> Question #186040 on OpenStack Compute (nova) changed:
> https://answers.launchpad.net/nova/+question/186040
>
> Truong Pham posted a new comment:
>
> Thanks for the response Vish!
>
> I already have those setting in nova.conf:
>
> # egrep 'logdir|state-path|lock_path' nova.conf
> --logdir=/var/log/nova
> --lock_path=/var/lib/nova/tmp
> # egrep 'logdir|state_path|lock_path' nova.conf
> --logdir=/var/log/nova
> --state_path=/var/lib/nova
> --lock_path=/var/lib/nova/tmp
> [root@bncloud01 nova]# ls -ld /var/lib/nova/; ls -ld /var/lib/nova/tmp
> drwxr-xr-x 10 nova nobody 4096 Jan 27 12:56 /var/lib/nova/
> drwxr-xr-x 2 nova nobody 4096 Jan 27 12:56 /var/lib/nova/tmp
>
> I actually had this working on before on the test machine. I repeated the same process and running into this issue...
> Weird?
> -TP
>
> --
> You received this question notification because you are a member of Nova
> Core, which is an answer contact for OpenStack Compute (nova).

Revision history for this message
Truong Pham (truongdpham) said : 		#5

Thanks Vish Ishaya, that solved my question.