Ask Your Question

Rebooting with DevStack?

asked 2013-09-25 15:15:58 -0500

rz2p gravatar image

I am getting mixed results, so I will ask the question.

I have OpenStack setup through Devstack scripts. I have about 20 images, and about 9 running VMs and I need to reboot the host.

From what I have read, after rebooting, I can run ./ followed by a ./rejoin-stack and APPARENTLY this will somehow re-register all 20 of my images(VMs too?).

Could someone confirm this for me?

My Goal:

Reboot the system and retain the users, the images and all those settings. I am not overly worried about the VMs themselves, as I can always re-create them.

Thanks for any input


edit retag flag offensive close merge delete

2 answers

Sort by ยป oldest newest most voted

answered 2013-09-26 03:48:35 -0500

unmesh-gurjar gravatar image

Once the system hosting devstack installation is rebooted, you only need to execute the script (if you run, it will recreate all the databases, thus losing data of all the images, users, instances, etc). The script will attempt starting all the services (that were earlier installed using the The users, images should still be available after the host reboots. Not sure about whether all the instances that were running as the system went down will be up again. PS: Some services like Cinder volume might not start right away after a reboot (since you need to create the volume group).

edit flag offensive delete link more



running 'sudo losetup -f /opt/stack/data/stack-volumes-backing-file' before makes the volume group come online and cinder-volume will start ok.

darragh-oreilly gravatar imagedarragh-oreilly ( 2013-09-26 13:06:09 -0500 )edit

Thanks Unmesh! One final question.When I go to reboot the system, must I ./ first? then ./ to bring it back? Or simply shut the server down? I am just having trouble understanding how the screens can be resumed after the server was powered off? I dont know much about screen.

rz2p gravatar imagerz2p ( 2013-09-26 15:47:00 -0500 )edit

If you are rebooting the system, you need not run script. Internal details: when you run the script, it populates a file stack-screenrc with the commands to start each service (you asked devstack to enable). So, after a reboot, when you run, it reads stack-screenrc and restarts the services.

unmesh-gurjar gravatar imageunmesh-gurjar ( 2013-09-27 01:59:22 -0500 )edit

I had to run @darragh-oreilly's command for /opt/stack/data/stack-volumes-default-backing-file and /opt/stack/data/stack-volumes-lvmdriver-1-backing-file instead. I also had to restart the httpd, mysqld, rabbitmq-server and openvswitch services.

sfinucan gravatar imagesfinucan ( 2015-06-16 09:02:27 -0500 )edit
1 is no longer available since stable/mitaka version

webknjaz gravatar imagewebknjaz ( 2016-07-06 05:55:40 -0500 )edit

answered 2015-01-08 15:41:50 -0500

joeq gravatar image

I have to resurrect this but I am having a similar issue. I have successfully installed devstack and tested it with horizon and other tools. After a reboot, I run and the output of the script is as follows.

2015-01-08 01:33:59.381 DEBUG heat-api-cloudwatch [-] ************************** from (pid=4264) log_opt_values /usr/lib/python2.7/site-packages/oslo_config/ 2015-01-08 01:33:59.381 INFO heat.api.cloudwatch [-] Starting Heat CloudWatch API on 2015-01-08 01:33:59.383 INFO eventlet.wsgi.server [-] Starting single process server 2015-01-08 01:33:59.384 DEBUG eventlet.wsgi.server [-] (4264) wsgi starting up on from (pid=4264) write /opt/stack/heat/heat/common/

It is refusing connecitons and I am unable to understand where the problem lies. The only way to fix is to rerun which takes forever. Please advise

[stack@win2k3 devstack]$ heat stack-list Unable to establish connection to [stack@win2k3 devstack]$ curl localhost curl: (7) Failed connect to localhost:80; Connection refused [stack@win2k3 devstack]$

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools



Asked: 2013-09-25 15:15:58 -0500

Seen: 17,451 times

Last updated: Sep 26 '13