Tags:
create new tag
,
view all tags
---+ Upgrade dos Servidores ---++Description ---+++ Certificados para a spraid01 e spraid02 Na spraid01 e spraid02 <pre> [mdias@spraid01 ~]$ cert-request -ou s -host spraid01.sprace.org.br -dir . -label spraid01 -agree -email xxxx@@ift.unesp.br -phone +55.11.45263XX -reason "Instaling a new Storage Element (gridftp) for SPRACE site" -name "Marco Dias [mdias@spraid02 ~]$ cert-request -ou s -host spraid02.sprace.org.br -dir . -label spraid02 -agree -email xxxx@ift.unesp.br -phone +55.11.45263xx -reason "Instaling a new Storage Element (gridftp) for SPRACE site" -name "Marco Dias </pre> <pre> [mdias@spraid01 ~]$ cert-retrieve -certnum XXXXX -label spraid01 -dir . -prefix spraid01.sprace.org.br [root@spraid01 ~]# mv /home/mdias/spraid01.sprace.org.brcert.pem /etc/grid-security/hostcert.pem [root@spraid01 ~]# mv /home/mdias/spraid01.sprace.org.brkey.pem /etc/grid-security/hostkey.pem [root@spraid01 ~]# chown root:root /etc/grid-security/host*.pem </pre> na antiga spraid (spraid02) <pre> [root@spraid02 ~]# mv /etc/grid-security/hostcert.pem /etc/grid-security/spraidcert.pem [root@spraid02 ~]# mv /etc/grid-security/hostkey.pem /etc/grid-security/spraidkey.pem [root@spraid02 ~]# su - mdias [mdias@spraid02 ~]$ . /OSG/setup.sh [mdias@spraid02 ~]$ cert-retrieve -certnum XXXX -label spraid02 -dir . -prefix spraid02.sprace.org.br [root@spraid02 ~]# mv /home/mdias/spraid02.sprace.org.brcert.pem /etc/grid-security/hostcert.pem [root@spraid02 ~]# mv /home/mdias/spraid02.sprace.org.brkey.pem /etc/grid-security/hostkey.pem [root@spraid02 ~]# chown root:root /etc/grid-security/host*.pem [root@spraid02 ~]# chmod 444 /etc/grid-security/hostcert.pem [root@spraid02 ~]# chmod 400 /etc/grid-security/hostkey.pem </pre> Para fazer o gridftp ser feito pela interface certa (ele da um erro "Expected /host/<nome da maquina-rede interna> target but get /org=doegrid//nome da maquina.sprace.org.br: <pre> [root@spraid02 ~]# vim /opt/d-cache/config/gridftpdoor.batch create dmg.cells.services.login.LoginManager GFTP-${thisHostname} \ "${gsiFtpPortNumber} \ -listen=200.136.80.7\ </pre> ---+++Update dos nós Já que iremos parar a farm, por que não fazer update dos pacotes instalados nas máquinas? Criando o mirror na sprace: <pre> [root@@sprace ~]# nohup nice rsync -avzlH --delete --exclude=sites/Fermi --exclude=errata/debuginfo --exclude=errata/obsolete rsync://rsync.scientificlinux.org/scientific/44/i386 /export/linux/SL_45_i386& [root@@sprace ~]# yum install createrepo [root@@sprace ~]# mkdir /var/www/html/linux [root@@sprace ~]# mkdir /var/www/html/linux/scientific [root@@sprace ~]# mkdir /var/www/html/linux/scientific/45 [root@@sprace ~]# mkdir /var/www/html/linux/scientific/45/errata [root@@sprace ~]# ln -s /export/linux/SL_45_i386/errata/SL/RPMS/ /var/www/html/linux/scientific/45/errata/SL [root@@sprace ~]# ln -s /export/linux/SL_45_i386/SL/RPMS/ /var/www/html/linux/scientific/45/SL [root@@sprace ~]# createrepo /var/www/html/linux/scientific/45/errata/SL [root@@sprace ~]# createrepo /var/www/html/linux/scientific/45/SL/ [root@@sprace ~]# vim /etc/httpd/conf/httpd.conf #_____________Repositorio do Scientific Linux__________ <Directory "/var/www/html/linux"> Options +Indexes </Directory> [root@@sprace ~]# /etc/init.d/httpd restart </pre> Agora devemos configurar os clientes. Desabilitando o repositorio sl padrao: <pre> [root@@sprace ~]# for ((i=1; i<19; i++)) ; do ssh 192.168.1.$i "sed -i 's/enabled=1/enabled=0/g' /etc/yum.repos.d/sl.repo /etc/yum.repos.d/sl-errata.repo"; done [root@@sprace ~]# for ((i=21; i<84; i++)) ; do ssh 192.168.1.$i "sed -i 's/enabled=1/enabled=0/g' /etc/yum.repos.d/sl.repo /etc/yum.repos.d/sl-errata.repo"; done </pre> agora inserimos nosso repositorio <pre> [root@@sprace ~]# for ((i=1; i<19; i++)) ; do scp -r /export/linux/SL_45_i386/sprace*.repo 192.168.1.$i:/etc/yum.repos.d/. ; done [root@@sprace ~]# for ((i=21; i<84; i++)) ; do scp -r /export/linux/SL_45_i386/sprace*.repo 192.168.1.$i:/etc/yum.repos.d/. ; done </pre> fazendo a atualizacao <pre> [root@@sprace ~]# for ((i=1; i<19; i++)) ; do ssh 192.168.1.$i 'yum -y clean all; yum -y update yum; yum -y update' ; done [root@@sprace ~]# for ((i=21; i<84; i++)) ; do ssh 192.168.1.$i 'yum -y clean all; yum -y update yum; yum -y update' ; done </pre> ---+++ Backup do database da spdc00 Parar o dcache, pnfs <pre> [root@@spraid ~]# /opt/d-cache/bin/dcache-pool stop [root@@spraid ~]# /opt/d-cache/bin/dcache-core stop </pre> na spdc00 <pre> [root@@spdc00 ~]# /opt/d-cache/bin/dcache-core stop [root@@spdc00 ~]# /opt/pnfs/bin/pnfs stop [root@@spdc00 ~]# pg_dump -U pnfsserver admin >admin.dump ;pg_dump -U pnfsserver data1 >data1.dump [root@@spdc00 ~]# scp /root/*.dump mdias@@osg-se.sprace.org.br:/home/mdias/. </pre> ---+++Backup da spgrid e spraid: <pre> [root@@spgrid ~]# cd /raid0/spgrid_backup/ [root@@spgrid spgrid_backup]# dd if=/dev/sda of=spgrid.img bs=512k conv=noerror </pre> o mesmo para spraid: <pre> [root@@spraid ~]# mkdir /raid0/spraid_backup [root@@spraid ~]# cd /raid0/spraid_backup/ [root@@spraid spraid_backup]# dd if=/dev/sda of=spraid.img bs=512k conv=noerror </pre> A atualização dos backups pode ser feita diretamente nas imagens iso: Neste exemplo iremos montar o /var. Determine o numero de cilindros da maquina: <pre> [root@@spgrid ~]# df -h Filesystem Size Used Avail Use% Mounted on /dev/sda3 2.0G 695M 1.2G 37% / /dev/sda1 248M 13M 223M 6% /boot none 1014M 0 1014M 0% /dev/shm /dev/sda7 2.0G 45M 1.9G 3% /tmp /dev/sda5 9.9G 5.7G 3.7G 61% /usr /dev/sda8 12G 6.2G 4.8G 57% /usr/local /dev/sda6 4.0G 1.5G 2.4G 38% /var [root@@spgrid ~]# fdisk -l 64 heads, 32 sectors/track, 34680 cylinders </pre> na spraid <pre> [root@@spraid ~]# df -h Filesystem Size Used Avail Use% Mounted on /dev/sda2 2.0G 902M 1012M 48% / none 1013M 0 1013M 0% /dev/shm /dev/sda7 1012M 34M 927M 4% /tmp /dev/sda5 9.9G 3.2G 6.2G 34% /usr /dev/sda8 15G 2.5G 12G 18% /usr/local /dev/sda6 2.0G 540M 1.4G 29% /var [root@@spraid ~]# fdisk -l /dev/sda 255 heads, 63 sectors/track, 4425 cylinders </pre> usando a saida em <pre> [root@@spgrid ~]# fdisk -l -u -C 34680 /raid0/spgrid_backup/spgrid.img /raid0/spgrid_backup/spgrid.img6 34078752 42467327 4194288 83 Linux </pre> multiplique o Start por 512 bytes e monte <pre> [root@@spgrid ~]# bc 34078752*512 17448321024 quit [root@@spgrid ~]# mount -o loop,offset=17448321024 -t ext3 /raid0/spgrid_backup/spgrid.img /root/test2/ </pre> sincronize a pasta /var com a imagem dd <pre> [root@@spgrid ~]# rsync -avx --delete /var/ /root/test2/ [root@@spgrid ~]# umount /root/test2 </pre> ---+++ Atualizando o backup Atualizando o backup da spgrid para a data antes do upgrade <pre> [root@@spgrid mdias]# mount -o loop,offset=16384 -t ext3 /raid0/spgrid_backup/spgrid.img /root/test [root@@spgrid mdias]# rsync -avx --delete /boot/ /root/test2/ [root@@spgrid mdias]# umount /root/test2 [root@@spgrid mdias]# mount -o loop,offset=4563402752 -t ext3 /raid0/spgrid_backup/spgrid.img /root/test2 [root@@spgrid mdias]# rsync -avx --delete / /root/test2/ [root@@spgrid mdias]# umount /root/test2 [root@@spgrid mdias]# mount -o loop,offset=6710902784 -t ext3 /raid0/spgrid_backup/spgrid.img /root/test2 [root@@spgrid mdias]# rsync -avx --delete /usr/ /root/test2/ [root@@spgrid mdias]# umount /root/teste [root@@spgrid mdias]# mount -o loop,offset=17448321024 -t ext3 /raid0/spgrid_backup/spgrid.img /root/teste [root@@spgrid mdias]# rsync -avx --delete /var/ /root/teste/ [root@@spgrid mdias]# umount /root/teste [root@@spgrid ~]# mount -o loop,offset=21743288320 -t ext3 /raid0/spgrid_backup/spgrid.img /root/teste2 [root@@spgrid ~]# rsync -avx --delete /tmp/ /root/teste2/ [root@@spgrid ~]# umount /root/teste2 [root@@spgrid mdias]# mount -o loop,offset=23890771968 -t ext3 /raid0/spgrid_backup/spgrid.img /root/teste [root@@spgrid mdias]# rsync -avx --delete /usr/local/ /root/teste/ [root@@spgrid ~]# umount /root/teste </pre> ---+++Problemas no condor Nos nodes, temos que modificar o /etc/fstab <pre> sed -i 's/spg00/osg-ce/g' /etc/fstab </pre> e lembrar de editar o /etc/hosts e /etc/resolv.conf, mudando de if.usp.br para sprace.org.br. Na osg-ce editar tudo sobre security em <pre> [root@@osg-ce ~]# vim /opt/osg-0.8.0/condor/local.osg-ce/condor_config.local # Security setup to use pool password </pre> fazer a configuracao do <pre> vim /OSG/condor/etc/condor_config </pre> como em <pre> mkdir /scratch/condor;mkdir /scratch/OSG;chmod a+rw /scratch/OSG; chown condor:condor /scratch/condor cd /scratch/condor;mkdir execute log spool chown condor:condor execute log spool ln -s /OSG/condor/etc/condor_config condor_config vim condor_config.local ############################################# # # Local Condor Configuration File for osg-ce # ############################################ DAEMON_LIST = MASTER, COLLECTOR, NEGOTIATOR, STARTD, SCHEDD MASTER = $(SBIN)/condor_master COLLECTOR = $(SBIN)/condor_collector NEGOTIATOR = $(SBIN)/condor_negotiator STARTD = $(SBIN)/condor_startd SCHEDD = $(SBIN)/condor_schedd NETWORK_INTERFACE = 192.168.1.150 START_LOCAL_UNIVERSE = TotalLocalJobsRunning < 20 || GridMonitorJob =?= TRUE chown condor:condor condor_config.local </pre> ---+++ Problemas com o GUMS Devido a problemas com o pacman tivemos que reinstalar o osg inteiro: <pre> cd /opt/pacman-3.21/ source setup.sh cd /OSG/ pacman -get OSG:ce source setup.sh pacman -get OSG:Globus-Condor-Setup pacman -get OSG:ManagedFork $VDT_LOCATION/vdt/setup/configure_globus_gatekeeper --managed-fork y --server y pacman -get OSG:gums cp /opt/osg-0.8.0/post-install/gsi-authz.conf /etc/grid-security/. cp /opt/osg-0.8.0/post-install/prima-authz.conf /etc/grid-security/. /opt/osg-0.8.0/vdt/sbin/vdt-register-service -name gums-host-cron --enable /opt/osg-0.8.0/vdt/sbin/vdt-control --on gums-host-cron vim $VDT_LOCATION/MonaLisa/Service/VDTFarm/ml.properties lia.Monitor.group=OSG vim /OSG/MonaLisa/Service/CMD/site_env CONDOR_LOCATION=/opt/osg-0.8.0/condor export CONDOR_LOCATION CONDOR_CONFIG=/opt/osg-0.8.0/condor/etc/condor_config export CONDOR_CONFIG CONDOR_LOCAL_DIR=/scratch/condor export CONDOR_LOCAL_DIR vim /OSG/MonaLisa/Service/CMD/ml_env FARM_NAME=SPRACE vdt-register-service --name MLD --enable /OSG/vdt/setup/configure_prima_gt4 --enable --gums-server osg-ce.sprace.org.br vim /OSG/condor/etc/condor_config chown globus: /etc/grid-security/containerkey.pem /etc/grid-security/containercert.pem /OSG/tomcat/v55/webapps/gums/WEB-INF/scripts/addMySQLAdmin "/DC=org/DC=doegrids/OU=People/CN=Marco Dias 280904" /opt/osg-0.8.0/vdt/setup/configure_mds -secure cd /OSG/tomcat/v55/webapps/gums/WEB-INF/scripts ./gums-create-config --osg-template </pre> Para que o /OSG/verify/site_verify.sh rodasse sem erro, devemos editar o /OSG/globus/etc/grid3-info.conf: <pre> /opt/osg-0.8.0/monitoring/configure-osg.sh </pre> Ainda é preciso fazer o Prima compativel com o GUMS <pre> > cd /etc/grid-security/http > mv httpkey.pem httpkey.pem.sav > mv httpcert.pem httpcert.pem.sav > cd .. > cp hostkey.pem http/httpkey.pem > cp hostcert.pem http/httpcert.pem > chown -R daemon:daemon http </pre> Alguns diretórios onde fica os software, tiveram as permissôes ajustadas <pre> chmod 1777 /raid0/OSG/data chmod 1777 /raid0/OSG/app chmod 1777 /raid0/OSG/pool </pre> o 1 significa que mesmo com prmissao 777 (read, write, exec) ele nao pode mover este diretorio. Ainda, as contas como a uscms00x tem que ser mapeadas e configuradas da seguinte forma. Esses caras que tem numero no final do user precisam ser mapeados com tipo um range de numeracao. Isso eh feito entrando em https://osg-ce.sprace.org.br:8443/gums/ (infelizmente precisa ser na minha conta, com um navegador com minha assinatura digital), ir na aba lateral em Manage Pool Accounts e colocar o range uscms000-009 ---+++ Problemas com o CEmon e o GIP <pre> [root@osg-ce ~]# $VDT_LOCATION/vdt/setup/configure_cemon --consumer https://osg-ress-1.fnal.gov:8443/ig/services/CEInfoCollector --topic OSG_CE [root@osg-ce ~]# /opt/osg-0.8.0/vdt/setup/configure_gip [root@osg-ce ~]# vdt-register-service --name gris --enable </pre> Mexendo nas permissões: <pre> root@osg-ce ~]# ls -al /OSG/apache/conf/extra/httpd-ssl.conf -rw-r--r-- 1 daemon daemon 10400 Mar 9 15:11 /OSG/apache/conf/extra/httpd-ssl.conf [root@osg-ce ~]# more /OSG/apache/conf/extra/httpd-ssl.conf|grep SSLCertificateFile SSLCertificateFile /etc/grid-security/http/httpcert.pem [root@osg-ce ~]# ls -al /etc/grid-security/http/httpcert.pem -r--r--r-- 1 daemon daemon 1477 Mar 11 16:54 /etc/grid-security/http/httpcert.pem [root@osg-ce ~]# ls -al /etc/grid-security/grid-mapfile -rw-r--r-- 1 root root 190469 Mar 8 12:54 /etc/grid-security/grid-mapfile </pre> Observar se o apache e o tomcat estão rodadndo e se temos acesso ao link https://osg-ce.sprace.org.br:8443/ce-monitor/services/CEMonitor?method=GetInfo Por último verificar se o gip está fornecendo as informacões para o CEMon <pre> [root@osg-ce ~]# /OSG/lcg/libexec/osg-info-wrapper >/tmp/out 2>/tmp/err [root@osg-ce ~]# more /tmp/err [root@osg-ce ~]# [root@osg-ce ~]# /OSG/verify-gip-for-cemon/verify-gip-for-cemon-wrapper </pre> Observando nos logs <pre> [root@osg-ce ~]# tail -f /OSG/tomcat/v5/logs/glite-ce-monitor.log 25 Mar 2008 23:29:13,546 org.glite.ce.monitor.holder.NotificationHolder - Sending notification to https://osg-ress-1.fnal.gov:8443/ig/services/CEInfoCollector ... 25 Mar 2008 23:29:14,014 org.glite.security.trustmanager.CRLFileTrustManager - Client CN=osg-ress-1.fnal.gov, OU=Services, DC=doegrids,DC=org accepted </pre> Foi publicado. Agora tem que dar a saida exata: <pre> [root@node80 ~]# condor_status -pool osg-ress-1.fnal.gov -l -constraint "GlueCEInfoHostName==\"osg-ce.sprace.org.br\"" </pre> ---+++ Configurações do DNS Para que o nosso DNS reverso funcione, devemos agora ser master para a maquina ns1.ansp.br. As modificações para isso são: * No arquivo named.conf (est em /var/named/chroot/etc) voc deve editar as zonas do SPRACE (direta e reversa) e incluir o "allow-transfer" como est aabaixo <pre> //Dominio sprace.org.br zone "sprace.org.br" IN{ type master; file "sprace.org.br.zone"; allow-transfer{ 143.108.30.90; }; }; //Dominio Reverso da sprace.org.br zone "80.136.200.in-addr.arpa" IN{ type master; file "80.136.200.in-addr.arpa.zone"; allow-transfer{ 143.108.30.90; }; </pre> * Nos arquivos "sprace.org.br.zone" e "80.136.200.in-addr.arpa.zone" (estâo no /var/named/chroot/var/named). Você deve incluir a linha "@ IN NS ns1.ansp.br." Como mostrado abaixo: <pre> @ IN NS osg-ce.sprace.org.br. @ IN NS ns1.ansp.br. </pre> * Ainda aproveitamos e criamos um alias para ns.sprace.org.br ao invés de osg-ce.sprace.org.br, no "sprace.org.br.zone": <pre> ;Nome do servidor e e-mail do administrador @ IN SOA ns.sprace.org.br. root.ns.sprace.org.br. ( @ IN NS ns.sprace.org.br. @ IN NS ns1.ansp.br. ;adicionando endereços ns IN A 200.136.80.4 osg-ce IN A 200.136.80.4 </pre> e no "80.136.200.in-addr.arpa.zone" <pre> @ IN SOA ns.sprace.org.br. root.ns.sprace.org.br. ( ;Especificando o servidor de nomes @ IN NS ns.sprace.org.br. @ IN NS ns1.ansp.br. ;Adicionando os enderecos reversos 3 IN PTR access.sprace.org.br. 4 IN PTR ns.sprace.org.br. 4 IN PTR osg-ce.sprace.org.br. </pre> ---+++ Configuracoes na Spraid01 Depois da instalacao, Complementando a instalação: <pre> [root@@spraid01 ~]# vim /etc/yp.conf domain grid server 192.168.1.150 ypserver 192.168.1.150 [root@@spraid01 ~]# vim /etc/nsswitch.conf passwd: files nis shadow: files nis group: files nis publickey: nisplus aliases: files nisplus [root@@spraid01 ~]# vim /etc/sysconfig/network NISDOMAIN=grid [root@@spraid01 ~]# mv /etc/ntp.conf /etc/ntp_conf.bck [root@@spraid01 ~]# vim /etc/ntp.conf server 192.168.1.150 authenticate no driftfile /var/lib/ntp/drift [root@@spraid01 ~]# vim /etc/ntp/step-tickers 192.168.1.150 </pre> Montagem do diretorio OSG. Primeiro na osg-ce <pre> [root@@osg-se /]# scp /etc/hosts 200.136.80.30:/etc/. </pre> <pre> [root@@spraid01 ~]# echo "/spraid01 /etc/auto.spraid01 --timeout=30" >> /etc/auto.master [root@@spraid01 ~]# echo "OSG -rw,soft,bg,rsize=8192,wsize=8192,tcp 192.168.1.150:/OSG" > /etc/auto.spraid01 [root@@spraid01 ~]# chkconfig autofs on [root@@spraid01 ~]# ln -s /spraid01/OSG /OSG </pre> Na osg-se <pre> [root@@osg-se /]# scp /root/dcache-server-1.8.0-8.noarch.rpm /root/dcache-client-1.8.0-0.noarch.rpm /root/ganglia-monitor-core-gmond-2.5.4-8.i386.rpm /root/jdk-6u2-ea-bin-b02-linux-i586-12_apr_2007-rpm.bin 200.136.80.30: </pre> Instalando o ganglia <pre> [root@@spraid01 ~]# groupadd -g 104 ganglia;useradd -d /var/lib/ganglia -s /bin/false -g ganglia -u 107 ganglia [root@@spraid01 ~]# rpm -ivh /root/ganglia-monitor-core-gmond-2.5.4-8.i386.rpm [root@@spraid01 ~]# mv /etc/gmond.conf /etc/gmond_conf.bck [root@@spraid01 ~]# vim /etc/gmond.conf name "SPGRID Cluster" owner "SPRACE-HEP" url "http://osg-ce.sprace.org.br" num_nodes 86 setuid ganglia location "0,5,0" mcast_if eth1 </pre> Instalando java <pre> [root@@spraid01 ~]# cd /root/ [root@@spraid01 ~]# chmod 755 jdk-6u2-ea-bin-b02-linux-i586-12_apr_2007-rpm.bin [root@@spraid01 ~]# ./jdk-6u2-ea-bin-b02-linux-i586-12_apr_2007-rpm.bin </pre> Passando os certificados da sprace para a ftp-01: <pre> [root@@sprace mdias]# scp -r /home/mdias/ftp-01/ 200.136.80.30: </pre> Instalando os certificados: <pre> [root@@spraid01 ~]# mkdir /etc/grid-security [root@@spraid01 ~]# cp /root/ftp-01/host* /etc/grid-security/. [root@@spraid01 ~]# chmod 444 /etc/grid-security/hostcert.pem [root@@spraid01 ~]# chmod 400 /etc/grid-security/hostkey.pem </pre> Instalacao do dcache: <pre> [root@@spraid01 ~]# vim /etc/profile export JAVA_HOME=/usr/java/jdk1.6.0_02 [root@@spraid01 ~]# rpm -ivh /usr/local/src/dcache-client-1.8.0-0.noarch.rpm /usr/local/src/dcache-server-1.8.0-8.noarch.rpm[root@@spraid01 ~]# cp /opt/d-cache/etc/dCacheSetup.template /opt/d-cache/config/dCacheSetup [root@@spraid01 ~]# vim /opt/d-cache/config/dCacheSetup serviceLocatorHost=osg-se.sprace.org.br java="/usr/java/jdk1.6.0_02/bin/java" useGPlazmaAuthorizationModule=true useGPlazmaAuthorizationCell=false [root@@spraid01 ~]# cp /opt/d-cache/etc/node_config.template /opt/d-cache/etc/node_config [root@@spraid01 ~]# vim /opt/d-cache/etc/node_config NODE_TYPE=pool SERVER_ID=sprace.org.br ADMIN_NODE=osg-se.sprace.org.br GRIDFTP=yes poolManager=yes [root@@spraid01 ~]# vim /opt/d-cache/etc/dcachesrm-gplazma.policy saml-vo-mapping="ON" #era OFF saml-vo-mapping-priority="1" kpwd-priority="2" #era 3 grid-mapfile-priority="3" #era 4 gplazmalite-vorole-mapping-priority="4" #era 2 mappingServiceUrl="https://osg-ce.sprace.org.br:8443/gums/services/GUMSAuthorizationServicePort" [root@@spraid01 ~]# scp osg-se.sprace.org.br:/opt/d-cache/etc/dcache.kpwd /opt/d-cache/etc/dcache.kpwd [root@@spraid01 ~]# more /opt/d-cache/etc/dcache.kpwd|grep sprace.org.br|sed 's/login/authorize/g' > /etc/grid-security/storage-authzdb [root@@spraid01 ~]# echo "/raid2 1649 no" > /opt/d-cache/etc/pool_path [root@@spraid01 ~]# echo "/raid3 1649 no" >> /opt/d-cache/etc/pool_path [root@@spraid01 ~]# echo "/raid4 1649 no" >> /opt/d-cache/etc/pool_path [root@@spraid01 ~]# echo "/raid5 1649 no" >> /opt/d-cache/etc/pool_path [root@@spraid01 ~]# /opt/d-cache/install/install.sh [root@@spraid01 ~]# cp /opt/d-cache/bin/dcache-core /etc/init.d/. [root@@spraid01 ~]# cp /opt/d-cache/bin/dcache-pool /etc/init.d/. [root@@spraid01 ~]# chkconfig --add dcache-core [root@@spraid01 ~]# chkconfig --add dcache-pool [root@@spraid01 ~]# chkconfig dcache-core on [root@@spraid01 ~]# chkconfig dcache-pool on </pre> ---++ Migracao dos databases <pre> [root@@osg-se ~]# /etc/init.d/dcache-core stop [root@@osg-se ~]# /etc/init.d/pnfs stop [root@@osg-se ~]# dropdb -U postgres admin [root@@osg-se ~]# dropdb -U postgres data1 [root@@osg-se ~]# createdb -U postgres admin [root@@osg-se ~]# createdb -U postgres data1 [root@@osg-se ~]# psql -U postgres admin<admin.dump [root@@osg-se ~]# psql -U postgres data1<data1.dump [root@@osg-se ~]# /etc/init.d/postgresql stop [root@@osg-se ~]# /etc/init.d/postgresql start [root@@osg-se ~]# /etc/init.d/dcache-core start [root@@osg-se etc]# echo "osg-se.sprace.org.br:22125" >/pnfs/fs/admin/etc/config/dCache/dcache.conf [root@@osg-se etc]# echo "sprace.org.br" > /pnfs/fs/admin/etc/config/serverId [root@@osg-se etc]# echo "osg-se.sprace.org.br" >/pnfs/fs/admin/etc/config/serverName </pre> ---++ Update: Marco em 19/06/2008 Feito a correcao do /etc/nsswitch.conf da osg-ce, tirando o nisplus e tirando o nis de servicos desnecessarios, evitando erros nos logs: <pre> passwd: files nis shadow: files group: files nis #hosts: db files ldap nis dns hosts: files dns # Example - obey only what ldap tells us... #services: ldap [NOTFOUND=return] files #networks: ldap [NOTFOUND=return] files #protocols: ldap [NOTFOUND=return] files #rpc: ldap [NOTFOUND=return] files #ethers: ldap [NOTFOUND=return] files bootparams: files ethers: files netmasks: files networks: files protocols: files rpc: files services: files netgroup: files publickey: files automount: files aliases: files </pre> Essa alteracao tambem foi feita nos nodes.
E
dit
|
A
ttach
|
P
rint version
|
H
istory
: r21
<
r20
<
r19
<
r18
<
r17
|
B
acklinks
|
V
iew topic
|
Ra
w
edit
|
M
ore topic actions
Topic revision: r21 - 2008-04-01
-
MarcoAndreFerreiraDias
Home
Site map
Main web
Sandbox web
TWiki web
Main Web
Users
Groups
Index
Search
Changes
Notifications
RSS Feed
Statistics
Preferences
P
View
Raw View
Print version
Find backlinks
History
More topic actions
Edit
Raw edit
Attach file or image
Edit topic preference settings
Set new parent
More topic actions
Account
Log In
Copyright © 2008-2025 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki?
Send feedback