반응형
1.1. Re-install a Compute-0-0
# ssh-agent $SHELL
# shoot-node compute-0-0
1.2. Forcing a Re-install at Next PXE Boot(compute-0-0)
# rocks set host boot compute-0-0 action=install
# rocks list host boot
HOST ACTION
isc: -------
compute-0-0: install
compute-0-1: os
# ssh compute-0-0 "sync;shutdown -r now"
2.1 Re-install All Compute Nodes
# rocks run host /boot/kickstart/cluster-kickstart
2.2 Forcing a Re-install at Next PXE Boot(All Compute Nodes)
# rocks set host boot compute action=install
# rocks list host boot
HOST ACTION
icluster: -------
compute-0-1: install
compute-0-0: install
# rocks run host "sync;shutdown -r now"
2.3 스크립트
#! /bin/bash
for node in `rocks list host | grep compute | cut -f 1 -d :`
do
rocks set host boot $node action=install
rocks run host $node "sync;shutdown -r now"
done
exit
반응형
'HPC > Rocks Clusters' 카테고리의 다른 글
로그인 노드 설치 (0) | 2012.07.04 |
---|---|
Adding Packages to Compute Nodes (0) | 2012.07.04 |
Enabling RSH on Compute Nodes (0) | 2012.06.10 |
Changing the Root password - Compute nodes (0) | 2012.06.10 |
노드 자동 재설치 비활성화 (0) | 2012.06.10 |
최근댓글