Administrators Guide
 ParaStation5 Administrators Guide
Info@par-tec.com
 Table of Contents
 Problem different groups of nodes are seen as up or down
 Introduction
History of ParaStation
 About this document
 Technical overview
Runtime daemon
Libraries
Kernel modules
 License
 Hardware
Installation
Prerequisites
 Kernel version
Directory structure
Software
 Installation via RPM packages
Getting the ParaStation5 RPM packages
Man
Mpi2, mpi2-intel, mpi2-pgi, mpi2-psc
 File Version
Installing the RPMs
Compiling the ParaStation5 packages from source
 ParaStation entries
Installing the documentation
Etc/init.d/xinetd reload
 Installing MPI
Further steps
# rpm -Uv psdoc-5.0.0-1.noarch.rpm
# rpm -Uv psmpi2.5.0.0-1.i586.rpm
 Uninstalling ParaStation5
 ParaStation5 Administrators Guide
 Configuration
Configuration of the ParaStation system
Copy template
Define Number of nodes
 Hostname id HWType runJob starter accounter
Enable optimized network drivers
# /opt/parastation/bin/testconfig
 Testing the installation
 # /opt/parastation/bin/psiadmin -s -c list
# /opt/parastation/bin/testnodes -np nodes
 Insight ParaStation5
ParaStation5 pscom communication library
 # cat /proc/sys/ps4/state/connections
# echo 10 /proc/sys/ps4/state/ResendTimeout
Directory /proc/sys/ps4/state
 Controlling process placement
Directory /proc/sys/ps4/local
 Exporting environment variables for a task
Using the ParaStation5 queuing facility
Using non-ParaStationapplications
 Controlling ParaStation5 communication paths
Export LDPRELOAD=/opt/parastation/lib64/libp4tcp.so
 Authentication within ParaStation5
Pspshm or Pspsharedmem
PSPP4S or PSPP4SOCK
Export PSPLIB=/opt/parastation/lib64/libpscomopenib.so
 Homogeneous user ID space
Single system view
Parallel shell tool
Nodes and CPUs
 Integration with AFS
Integrating external queuing systems
Tok2env
PSIRARGPRE0=/some/path/env2tok
 Multicasts
 Using ParaStation accounting
Copying files in parallel
# UseMCast
Route add -net 224.0.0.0 netmask 240.0.0.0 dev ethX
 Using ParaStation process pinning
Using memory binding
Changing the default ports for psid8
Spawning processes belonging to all groups
 Port
 Problem node shown as down
Troubleshooting
Problem psiadmin returns error
 Problem cannot start parallel task
Problem bad performance
Problem different groups of nodes are seen as up or down
Problem cannot start process on frontend
 Problem pssh fails
Problem psid does not startup, reports port in use
 Problem processes cannot access files on remote nodes
 Reference Pages
 ParaStation5 Administrators Guide
 InstallDir inst-dir , InstallationDir inst-dir
Parastation.conf
Description
Parameters
 Setupscript
Startscript
Stopscript
Statusscript
 P4sock
Openib
Mvapi
Elan
 Accounter
NrOfNodes num
 Node node17 16 HWType ethernet p4sock starter yes runJobs no
$GENERATE 1-96 node$0,2 $0
 SelectTime time
DeadInterval num
LogLevel num
MCastGroup group-num
 Core size
CPUTime time
DataSize size
MemLock size
 Proc
 Processes maxprocs
CPUmap map
 RdpTimeout ms
StatusTimeout ms
RdpClosedTimeout ms
RdpResendTimeout ms
 Errors
See also
 ParaStation5 Administrators Guide
 Options
Psiadmin
Synopsis
 Standard Error
Standard Input
Standard Output
Extended description
 Allproc cnt count
Exit
All
 Count hw hw
Down
Hardware
Load
 Quit
Rdp
Summary max max
 Accounters nodes
User nodes
Group nodes
Maxproc nodes
 Master nodes
FreeOnSuspend nodes
HandleOldBins nodes
NodesSort nodes
 Cpumap nodes
Rlrss nodes
 RdpTimeout nodes
StatusTimeout nodes
RdpClosedTimeout nodes
RdpResendTimeout nodes
 Resolve nodes
Restart nodes
 Selecttime time nodes
Psiddebug mask nodes
 Pattern Name Description
 HandleOldBins 0 1 nodes
 Rdpmaxretrans val nodes
 RdpTimeout ms nodes
StatusTimeout ms nodes
RdpClosedTimeout ms nodes
RdpResendTimeout ms nodes
 Files
Quiet
Normal
Verbose
 Psid
 Logfile=file
Configfile=file
Debug=level
 Testconfig
Filename
Num
? , --usage Show a help message
 ParaStation5 Administrators Guide
 Testnodes
Np num
Cnt num
Map
 ParaStation5 Administrators Guide
 Testpse
Testpse -npnum
 ParaStation5 Administrators Guide
 P4stat
Sock
Net
?,--help
 ParaStation5 Administrators Guide
 Delete
P4tcp
Add
 ParaStation5 Administrators Guide
 Psaccounter
Pattern Description
 Dumpcore
Coredir=dir
?, --help
Var/account/yyyymmdd Accounting files, one per day
 Psaccview
 Lj,--ljobs
Lu,--ltotuser
Lg,--ltotgroup
Ls,--ltotsum
 Cpuweight
Aqtime
Cputime
End
 Initialization file
 Mlisten
 ParaStation5 Administrators Guide
 Appendix A. Quick Installation Guide
 Testing
# /opt/parastation/bin/psiadmin psiadmin add
# chkconfig -a /etc/init.d/parastation
 Appendix B. ParaStation license
Page
Page
Page
 Building and installing ParaStation5 packages
# psiadmin -s
Appendix C. Upgrading ParaStation4 to ParaStation5
Changes to the runtime environment
Page
 Glossary
ARP
 See ParaStation Logger
 To share a common address space within a node
 ParaStation5 Administrators Guide