The EPIKH Project Exchange Programme to advance e-Infrastructure

The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) g. Lite SE(DPM) Installation ZHU Wei CC-IHEP [email protected] ac. cn 1 September 2010 www. epikh. eu

Outline u OVERVIEW u INSTALLATION & CONFIGURATION u TESTING 2

OVERVIEW • The Grid DM Challenge • Heterogeneity ―Data are stored on different storage systems using different access technologies Need common interface to storage resources Storage Resource Manager (SRM) • Distribution ―Data are stored in different locations – in most cases there is no shared file system or common namespace ―Data need to be moved between different locations Need to keep track where data is stored File and Replica Catalogs Need scheduled, reliable file transfer File transfer service 3

OVERVIEW • Assumptions: – Users and programs produce and require data – the lowest granularity of the data is on the file level (we deal with files rather than data objects or tables) – Data = files • Files: – – Mostly, write once, read many Located in Storage Elements (SEs) Several replicas of one file in different sites Accessible by Grid users and applications from “anywhere” • Also… – WMS can send (small amounts of) data to/from jobs: Input and Output Sandbox – Files may be copied from/to local filesystems (WNs, UIs) to the Grid (SEs) 4

OVERVIEW • Storage Element is the service which saves/loads files to/from local storages. These local storages can be both, a disk or large storage systems. • Functions: – File storage. – Storage resources administration interface. – Storage space administration. • g. Lite 3. 1 data access protocols: – File Transfer: GSIFTP (Grid. FTP) – File I/O (Remote File access): gsidcap insecure RFIO secured RFIO (gsirfio) 5

OVERVIEW • Classic SE: – Grid. FTP server – Insecure RFIO daemon (rfiod) – only LAN limited file access – Single disk or disk array – No quota management – Will not be supported anymore • Mass Storage Systems (Castor) – Files migrated between front-end disk and back-end tape storage hierarchies – Grid. FTP server – Insecure RFIO (Castor) – Provide a SRM interface with all the benefits • Disk pool managers (d. Cache and g. Lite DPM) – manage distributed storage servers in a centralized way – Physical disks or arrays are combined into a common (virtual) file system – Disks can be dynamically added to the pool – Grid. FTP server – Secure remote access protocols (gsidcap for d. Cache, gsirfio for DPM) – SRM interface 6

OVERVIEW • The Disk Pool Manager (DPM) is a lightweight solution for disk storage management, which offers the SRM (Storage Resource Manager) interfaces. • Each DPM–type Storage Element (SE), is composed by a head node and a disk server on the same machine. • The DPM head node has to have at least one files system in the pool, and then, an arbitrary number of disk servers can be added by YAIM. • The DPM handles the storage on Disk Servers. It handles pools: a pool is a group of file systems, located in one or more disk servers. The DPM Disk Servers can have multiple files system in the pool. 7

OVERVIEW • Usually the DPM head node hosts: – SRM server (srmv 1 and/or srmv 2) : receives the SRM requests and passes them to the DPM server. – DPM server : keeps track of all the requests. – DPM name server (DPNS) : handles the namespace for all the files under the DPM control. – DPM RFIO server : handles the transfers for the RFIO protocol. – DPM Gridftp server : handles the transfer for the Gridftp protocol. 8

OVERVIEW • File Catalog Motivations • • Users and programs produce and require data Data may be stored in Grid datasets (files) – Located in Storage Elements (SEs) – Several replicas of one file in different sites – Accessible by Grid users and applications from “anywhere” – Locatable by the WMS (data requirements in JDL) – Data may be copied from/to local filesystems (WNs, UIs) to the Grid • Why File Catalogs are used? – They keep track of the location of copies (replicas) of Grid files – They provide DM tools and APIs through which users and grid services can deal with files 9

OVERVIEW • Logical File Name (LFN) – An alias created by a user to refer to some item of data, e. g. “lfn: /grid/gilda/tutorials/zhuw/text_file. txt” • Globally Unique Identifier (GUID) – A non-human-readable unique identifier for an item of data, e. g. “guid: f 81 d 4 fae-7 dec-11 d 0 -a 765 -00 a 0 c 91 e 6 bf 6” • Site URL (SURL) (or Physical File Name (PFN) or Site FN) – The location of an actual piece of data on a storage system, e. g. “srm: //gilda 03. ihep. ac. cn/dpm/ihep. ac. cn/home/gilda/generated/2010 -08 -29/ filef 7 c 237 f 9 -15 ff-4 cd 1 -80 d 8 -f 2 ae 33199 b 51” (lcg-lr they can be many if the file has replicas somewhere) • Transport URL (TURL) – Temporary locator of a replica + access protocol: understood by a SE, e. g. “rfio: //gilda 03. ihep. ac. cn//storage/gilda/2010 -08 -29/filef 7 c 237 f 9 -15 ff-4 cd 1 -80 d 8 -f 2 ae 33199 b 51. 31. 0” 10

INSTALLATION & CONFIGURATION • INSTALLATION & CONFIGURATION 11

Preliminaries • The instructions provided in this presentation assume that you are going to use a machine with Scientific Linux 5. 4 installed. – we are using the Scientific Linux 5. 4 x 86_64. 12

Repository settings • Specify the mrepo host: export MREPO=http: //grid-it. cnaf. infn. it/mrepo/repos/sl 5/x 86_64/ • Configure the repository as follows: REPOS=“dag ig glite-generic lcg-ca glite-se_dpm_mysql” • Get repositories with: for name in $REPOS; do wget $MREPO/$name. repo -O /etc/yum. repos. d/$name. repo; done • Copy the gilda utils: wget http: //grid 018. ct. infn. it/mrepo/repos/gilda. repo -O /etc/yum. repos. d/gilda. repo 13

System requirements • Please check: – If you have a valid hostname – To verify, type: hostname -f – To edit the hostname: § vim /etc/hosts § vim /etc/sysconfig/network § hostname 14

YUM install • Then update your host: yum clean all • Install CAs on ALL profiles: yum install lcg-CA • Metapackage installation: yum install ig_SE_dpm_mysql • Gilda utils: yum install gilda_utils 15

site-info. def • Copy the ig-site-info. def script template cp /opt/glite/yaim/examples/siteinfo/ig-siteinfo. def /opt/glite/yaim/examples/siteinfo/mysiteinfo-se-dpm. def • Open your own site-info. def and customize it: vi /opt/glite/yaim/examples/siteinfo/mysite-info-sedpm. def 16

site-info. def • USERS_CONF=${INSTALL_ROOT}/glite/yaim/examples/igusers. conf • GROUPS_CONF=${INSTALL_ROOT}/glite/yaim/examples/iggroups. conf • MYSQL_PASSWORD=123456 • JAVA_LOCATION="/usr/lib/jvm/java-1. 6. 0 -openjdk 1. 6. 0. 0. x 86_64/jre“ • SITE_NAME=my-site-name (haven’t defined it yet) • SITE_EMAIL=“your email address“ • PX_HOST=myproxy. ct. infn. it • DPM_HOST=“your DPM hostname“ • SE_LIST=“your DPM hostname“ • BDII_HOST=top. BDII hostname • Add # before SITE_BDII_HOST=my-bdii. $MY_DOMAIN 17

site-info. def • Maybe you need add these line below： • • • ##################### # Sto. RM backend configuration variables # ##################### • # More information here: http: //igrelease. forge. cnaf. infn. it/doku. php? id=doc: guides: site-infovariables#se_storm_backend • • # Sto. RM backend host STORM_BACKEND_HOST=my-storm-backend. $MY_DOMAIN # Sto. RM storage path STORM_DEFAULT_ROOT=/storage 18

CONFIGURATION • In the machine of gilda**. ihep. ac. cn , the host certifcates locate at : /root/gilda 1228 certs. tar – – • #tar -xvf gilda 12 -28 certs. tar Then your host certificates are in the directory named your hostname Install host certificate: – – Download your certificates in /etc/grid-security: # cd gilda 12 -28 certs #mv gilda**/hostcert. pem /etc/grid-security/hostcert. pem #mv gilda**/hostkey. pem /etc/grid-security/hostkey. pem – and set proper permissions: – #chmod 644 /etc/grid-security/hostcert. pem – #chmod 400 /etc/grid-security/hostkey. pem

CONFIGURATION • • • #cd /opt/glite/yaim/examples #vi ig-groups. conf Add "/gilda/ROLE=Software. Manager": : gm: "/gilda/grelc/das/*": gilda: : : "/gilda": : • • • [[email protected] 03 ~]# vi /etc/hosts. allow Add one line: slapd: all • • #cd /opt/glite/yaim/examples/siteinfo/services #vi glite-se_dpm_mysql • • DPMPOOL="Permanent“ DPM_FILESYSTEMS="$DPM_HOST: /storage“ DPM_DB_USER=dpmdbuser DPM_DB_PASSWORD=123456 DPM_DB_HOST=$DPM_HOST DPM_INFO_USER="dpminfo” DPM_INFO_PASS=secret

Configuring the SE for GILDA Update the date /etc/init. d/ntpd status /etc/init. d/ntpd restart • Configure the UI with YAIM /opt/glite/yaim/bin/ig_yaim -c -s /opt/glite/yaim/examples/siteinfo/mysite-info-se-dpm. def -n ig_SE_dpm_mysql • If everithing is ok, the SE is now READY for testing 21

• TESTING

TESTING • Login UI with your account • [[email protected] 08 ~]$ dpm-qryconf POOL Permanent DEFSIZE 200. 00 M GC_START_THRESH 0 GC_STOP_THRESH 0 DEF_LIFETIME 7. 0 d DEFPINTIME 2. 0 h MAX_LIFETIME 1. 0 m MAXPINTIME 12. 0 h FSS_POLICY maxfreespace GC_POLICY lru RS_POLICY fifo GIDS 0 S_TYPE - MIG_POLICY none RET_POLICY R CAPACITY 32. 99 G FREE 26. 98 G ( 81. 8%) gilda 03. ihep. ac. cn /storage CAPACITY 32. 99 G FREE 26. 98 G ( 81. 8%) [[email protected] 08 ~]$ dpns-ls / dpm [[email protected] 08 ~]$ dpns-ls /dpm/ ihep. ac. cn [[email protected] 08 ~]$ dpns-ls /dpm/ihep. ac. cn/ home [[email protected] 08 ~]$ dpns-ls /dpm/ihep. ac. cn/home gilda …………

TESTING • rfio • • • [[email protected] 08 ~]$ rfdir /dpm/ihep. ac. cn/home/gilda drwxrwxr-x 4 103 101 0 Aug 29 16: 07 generated -rw-rw-r-- 1 103 101 115 Aug 23 00: 28 malx-1. jdl -rw-rw-r-- 1 102 101 115 Aug 25 23: 28 malx-21. jdl -rw-rw-r-- 1 102 101 115 Aug 20 18: 34 malx. jdl -rw-rw-r-- 1 101 12 Sep 01 00: 46 tutorials [[email protected] 08 ~]$ rfcp /dpm/ihep. ac. cn/home/gilda/malx-1. jdl /tmp/zhuw. jdl 115 bytes in 0 seconds through eth 0 (in) and local (out) [[email protected] 08 ~]$ ll /tmp/zhuw. jdl -rw-rw-r-- 1 zhuw 115 Sep 1 01: 04 /tmp/zhuw. jdl [[email protected] 08 ~]$

TESTING • • • [[email protected] 08 ~]$ lcg-infosites --vo gilda se Avail Space(Kb) Used Space(Kb) Type SEs -----------------------------38243284 20169054 n. a vega-se. ct. infn. it 97173177 6523022 n. a dgt 02. ui. savba. sk ………… 299672322 94166511 n. a aliserv 6. ct. infn. it 28968762 6459208 n. a gilda 03. ihep. ac. cn 16243037 3462310 n. a se. hpc. iit. bme. hu 2391755671 95958513 n. a se 1 -egee. srce. hr 299672322 94166511 n. a aliserv 6. ct. infn. it [[email protected] 08 ~]$ • [[email protected] 08 ~]$ ldapsearch -x -h gilda 03. ihep. ac. cn -p 2170 -b mds-voname=resource, o=grid | grep Available. Space Glue. SAState. Available. Space: 28968762 •