RCF XRootD Coniguration Details

From Computing@RNC

Revision as of 23:13, 2 February 2010 by Aarose (Talk | contribs)
(diff) ← Older revision | Current revision (diff) | Newer revision → (diff)
Jump to: navigation, search
  1.  !!!Removing this file is punished by death !!! #
  2. #
  3. This is a configuration file for XROOTD System #
  4. #

# # # XRD Daemon section # # #

  1. it specifies xrd port (1094 is rootd)

xrd.port 1095

  1. debug level for xrd and xrootd module

xrd.trace conn net xrootd.trace login emsg 

       #						#

# XROOTD Section # # #

  1. here is loading the extended file system support for xrootd

xrootd.fslib /common/star/starlib/ROOT/xrootd/lib/libXrdOfs.so

  1. here is loading support of authentication

if named dataserver xrootd.seclib /star/u/starlib/ROOT/xrootd/lib/libXrdSec.so fi

  1. enable security only one nodes that holds data

if named dataserver sec.protocol /star/u/starlib/ROOT/xrootd/lib pwd -a:0 fi

  1. specify a valid path prefix to file requests (export all of them, since we still have PFN)

xrootd.export /data0 xrootd.export /data1 xrootd.export /data2 xrootd.export /data3 xrootd.export /home

# # # OLB Section # # #

  1. Restriction for hosts that subscribe to the olbd manager

if named redirector olb.allow host rcas*.rcf.bnl.gov fi

  1. Designate the port number to use for incoming requests from client (this is mandatory directive)

olb.port 3121


  1. Specify how load is computed and reported.

if named dataserver olb.perf int 5m pgm /star/u/starlib/ROOT/xrootd/utils/XrdOlbMonPerf.pl 300 fi

  1. Specify the parameters for the load balancing scheduling algorithm (tresholds set up by monitoring)

if named redirector supervisor olb.sched cpu 20 io 10 mem 40 pag 20 runq 10 fuzz 20 refreset 3600 fi

  1. Specify how server are selected for file creation (important for purging tresholds)

olb.space linger 0 250m 500m

  1. specify ping directive (request each 60s usage from servers- usage*15s and log none of them)

if named redirector supervisor olb.ping 15s log 0 usage 4 fi

  1. Designate paths handle by server (r - read, s will be staging) (specify also for /data* because we still have some files out of XROOTD namespace)

if named dataserver supervisor olb.path rs /data0 olb.path rs /data1 olb.path rs /data2 olb.path rs /data3 olb.path rs /home fi

  1. Setup Manage Procesing latency (main for decreasing lookup delay)
  2. @see documentation

if named redirector olb.delay discard 5 drop 10s full * lookup 3 overload * olb.delay servers 80% service 60 startup 90 suspend 30 fi if named supervisor olb.delay discard 5 drop 10s full * lookup 1 overload * olb.delay servers 80% service 60 startup 90 suspend 30 fi

  1. Tell server olbd's the location of the manager

if named dataserver supervisor olb.subscribe rcas6132.rcf.bnl.gov 3121 fi

  1. Indecate the role this server will have based on host name (default is server)

olb.role manager if named redirector olb.role server if named dataserver olb.role supervisor if named supervisor

  1. Tell the odc component the location of the manager

if named redirector supervisor odc.manager rcas6132.rcf.bnl.gov 3121 fi

  1. Tell level of event loging

odc.trace -all

  1. for meassuring how many of requests are comming to xrootd

if named redirector olb.trace all else olb.trace -all fi

# # # OFS and OSS Section # # #

  2. Set redirect option using instance name (Supervisor must have both options)

ofs.redirect target if named dataserver ofs.redirect remote if named redirector ofs.redirect remote if named supervisor ofs.redirect target if named supervisor

  1. Restrict access to files only for operation read

oss.readonly

  1. Tell level of loging

ofs.trace -all oss.trace -all


  1. Control the way a disk cache partition is selected for file placement

oss.alloc 250m 10 50

  1. Control file descriptor's allocating

oss.fdlimit * max

  1. Specify whether or not to check that a file exists in HPSS

oss.check

  1. Specify command to obtain data from HPSS

oss.mssgwcmd /star/u/starlib/ROOT/xrootd/utils/hpss_talk.pl

  1. use mps stage command for dynamic staging and inside call our hpss_talk which knows how to communicate with DataCarousel

oss.stagecmd /star/u/starlib/ROOT/xrootd/utils/mps_Stage -F -c /star/u/starlib/ROOT/xrootd/config/configuration.cfg

  1. Restrict access to files only for operation read

oss.readonly

  1. Specify processing options for any entry matching the specified path prefix

oss.path /data0 oss.path /data1 oss.path /data2 oss.path /data3 oss.path /home

  1. prefix requests with /home/starlib (we don't have any other chance)

oss.localroot /home/starlib

  1. Specify the location of a disk cache partition
  2. PJ: Only wildcard would work here. Individual drive would not and crash
  3. xrootd.
  4. /data0 is linked to /tmp, must have more free space

oss.cache rhstar /data*

  1. number of threads, for beginning take defaults

oss.xfr 20 * 30 *


# # # MPS section # # #

  1. Where error message go via mail, choose some user

mps.adminuser starlib

  1. Specify path where will be placed symlinks

mps.scan /home/starlib

  1. set level of debuging for all components (4 -additionally records internal debugging information)

mps.debug 2

  1. is the command to be used to mail files to "adminuser" specified above

mps.mailcmd /bin/mail

  1. is the name of the file that is to hold error information

mps.errlogfn /star/data08/XROOTD/logs/logerr

  1. specify command for bringing the files from hpss

mps.stage.xfrcmd /star/u/starlib/ROOT/xrootd/utils/hpss_stage.pl %sfn %tfn

  1. specify minimum size of free space that must be mantained before trying to stage a file

mps.stage.minsize 786432000

  1. configuration before will execute purging

mps.stage.purgecmd /star/u/starlib/ROOT/xrootd/utils/mps_MigrPurg -c /star/u/starlib/ROOT/xrootd/config/configuration.cfg

  1. specify names of log files for every components of MPS separately

mps.stage.logfn /star/data08/XROOTD/logs/stage_log mps.purg.logfn /star/data08/XROOTD/logs/purge_log

  1. Set the number of times a failing data transfer operation may be retried.

mps.stage.max_retry 1 mps.max_retry 1

  1. is the name or number of the Unix group that is to be used for newly created files

mps.stage.cpgroup rhstar

  1. is the command to be used to communicate with the Mass Storage System.

mps.stage.msscmd /star/u/starlib/ROOT/xrootd/utils/hpss_talk.pl

  1. the command to issue in order to obtain free space information about a file system.

mps.stage._fs_stat /star/u/starlib/ROOT/xrootd/utils/fs_stat

Retrieved from "https://rnc.lbl.gov/Computing/wiki/index.php/RCF_XRootD_Coniguration_Details"
Debug data:
Personal tools