quattor Framework for Managing Grid-enabled Large Scale Computing Fabrics

Configuration arranged in hierarchies of Pan templates

CERNCC name_srv1: 137.138.16.5

time_srv1: ip-time-1

lxbatch

cluster_name: lxbatchmaster: lxmaster01pkg_add (lsf5.1)

lxplus

cluster_name: lxpluspkg_add (lsf5.1)

disk_srv

lxplus001 eth0/ip: 137.138.4.246 pkg_repl (lsf5.1_debug) lxplus020

eth0/ip: 137.138.4.225lxplus029

• Information is arranged in templates: common properties are set only once

• Using templates it is possible to create hierarchies to match service structures

GUI for configuration access

Install server

base OS

dhcppxe

nfs/http

Vendor System installer RH73, RHES,

Fedora,…

System services AFS,LSF,SSH,accounting..

Installed softwarekernel, system, applications..

CCMNode Configuration Manager (NCM)

RPM, PKG

nfshttp

ftp

Software Repositories

packages

(RPM, PKG)SWReppackages

CDB

Managed nodes

Install Manager

Node (re)install

cacheSW package Manager (SPMA)

Installation server

Node Management

Modularity, extensibility• Scalable and multi-

vendor Software Package Manager (RPM, PKG)

• Service specific configuration components make the necessary system changes

• Installation manager steers and configures the vendor system installer

Configuration Database

Central control• Primary configuration is kept centrally (and replicated on

the nodes)• Transactional validation, version control• A single source for all configuration information

GUI

Scripts

CLI

CCM

Cache

Node

PERL

XML

RDBMSSQL

SOAP

HTTP

panc

CDB

quattor deployment:

quattor is used to manage most of the Linux nodes in the CERN Computer Centre (>2500 nodes)• Multiple functionality (batch nodes, disk servers, tape servers, DB, web, …)• Heterogeneous hardware (memory, HD size,..)Also: UAM Madrid, NIKHEF, IN2P3/LAL

quattor functionality:• Configuration Management

– Configuration Database– Configuration access and caching– Graphical and Command Line Interfaces

• Node and Cluster Management– Automated node installation– Node Configuration Management– Software distribution and management

NodeConfigurationManagement

NodeManagement

FaultTolerance

Monitoring

Piotr Poznański(1) , German Cancio(1), Rafael Garcia Leiva(2), Lionel Cons(1)

(1) CERN, Geneva, Switzerland(2) Universidad Autonoma de Madrid (UAM), Spain

www.quattor.org