Deploy on CentOS failed (install CentOS again and again)

[Sergey Galkin]

api: '1.0'
astute_sha: c15623d05ccdf7ac10873e7a90df954de8726280
auth_required: true
build_id: 2014-11-26_13-35-04
build_number: '10'
feature_groups:
- mirantis
fuellib_sha: 25eb629f3c6a6ff41cf187e260fe4ff456cfc4e4
fuelmain_sha: 465afb6479a0b3c677040fb978cc109dcf62f774
nailgun_sha: bf9ddb9f9d5dbb09c4b50201ce176635791d7d3e
ostf_sha: a35f516f1606b0d03d51ff63bfe3fbe23de4b622
production: docker
release: '6.0'

Steps to reproduce
1. Deploy cluster on 99 HW nodes with centos+ceph+ceilometer+neutron gre (3 nodes as compute in HA)
Cluster deployment failed with error ' Timeout of provisioning is exceeded' . 33 nodes gone to offline.

As I can seen on some nodes by IPMI on this nodes CentOS installed again and again every new reboot

[Sergey Galkin (sgalkin)]

Snapshot https://drive.google.com/open?id=0B1LFPizteKUEYkN1bmxNN1hCVXc

[Łukasz Oleś (loles)]

This one is really strange. Maybe 10s after all packages are installed node reboots itself. I couldn't see any error on the screen but definitely installation process finished to early.

[Mike Scherbakov (mihgen)]

Can it be that boot record was not set? Did you check that server attempts to boot from disk in a first order?

[Sergey Galkin (sgalkin)]

Issues in the switches lab configuration