New Question
0

V-Magine install fail "A start job is running for dev-mapp.."

asked 2017-03-07 06:31:38 +0200

eckdd gravatar image

updated 2017-03-07 09:18:00 +0200

Server 2016. Install fails after timing out connected via SSH upon rebooting after install the LIS. Meanwhile, the console hangs at the following:

https://goo.gl/photos/LGLK7f1YG6aG92RP6

v-magine.log: http://paste.openstack.org/raw/601688/ v-magine-console.log: http://paste.openstack.org/raw/601689/

edit retag flag offensive close merge delete

Comments

Hi, this looks like a CentOS bug. The only related info I could find is: https://bugzilla.redhat.com/show_bug.cgi?id=979695 Does it have the same issue after rebooting the controller VM? Do you have the same issue by repeating the setup?

alexpilotti gravatar imagealexpilotti ( 2017-03-07 11:20:12 +0200 )edit

Hi, I've just started using v-magine today and I'm getting the same error. My environment is Server 2016 with Gui + Hyper-V installed. I applied the extra command to install tinyrpc into install-rdo.sh I even tried to delete everything and restart v-magine and it reproduced the error.

Deon Cui gravatar imageDeon Cui ( 2017-03-07 12:01:44 +0200 )edit

Hi, after the error occurs, what is the state of the VM? Does is it boot eventually? What about if you reboot it? Thanks.

danardelean gravatar imagedanardelean ( 2017-03-07 12:14:51 +0200 )edit

If this issue happens regularly, it's probably due to some CentOS updates. We're deploying a few test environments now to see if we can reproduce the issue.

alexpilotti gravatar imagealexpilotti ( 2017-03-07 12:25:32 +0200 )edit

The VM never finishes booting, this is after 1 hour of waiting. Rebooting the VM just repeats the same boot sequence, where it is basically hung waiting for the start job to finish.

Deon Cui gravatar imageDeon Cui ( 2017-03-07 12:45:46 +0200 )edit

1 answer

Sort by » oldest newest most voted
2

answered 2017-03-07 13:08:00 +0200

alexpilotti gravatar image

updated 2017-03-07 15:19:13 +0200

The LIS team managed to reproduce the issue on an updated vanilla RHEL / CentOS 7.3 and confirmed that it's a bug with the latest RHEL / CentOS kernel update that just rolled out (see issue: https://github.com/LIS/lis-next/issue...). While this is being taken care of on the LIS side, the fastest workaround is to revert to the previous kernel at the end of "install-rdo.sh", before v-magine installs LIS. Updates will follow as soon as we finish testing the workaround.

Updated LIS (Linux Integration Services) components are needed as the ones that come with the RHEL / CentOS kernel don't fully support Windows Server 2016 / Windows 10.

Update:

Reverting to the previous kernel solves the issue. This can be done by adding the following at the end of resources/install-rdo.sh :

# This is needed due to a bug in LIS with kernel-3.10.0-514.10.2.el7
exec_with_retry 5 0 /usr/bin/yum install -y kernel-3.10.0-514.6.2.el7

An updated v-magine.zip package with this workaround will be released as soon as tests are done.

The updated kernel that broke LIS was released yesterday (March 6th 2017) on CentOS: https://lists.centos.org/pipermail/ce...

Update 2

https://cloudbase.it/downloads/v-magi... now includes the workaround. We'll remove it as soon as the issue is fixed in LIS.

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

Question Tools

2 followers

Stats

Asked: 2017-03-07 06:31:38 +0200

Seen: 2,307 times

Last updated: Mar 07 '17