Wednesday 25 June 2014

decode hpc cluster


$clienthost pbsserver
$clienthost paroo3
cpuset_create_flags 0
# Enforce memory limits
$enforce mem
# Use local disk for temporary files, job files and checkpoints
$jobdir_root /scratch
$tmpdir /scratch
##$checkpoint_path /scratch/checkpoint
# Restrict use of batch nodes to PBS, interactive logins not allowed
##$restrict_user on
##$restrict_user_maxsysid 500
# Use cp rather than scp to transfer/stage files i.e. use Panasas
$usecp barrine*.hpcu.uq.edu.au:/home/ /home/
$usecp *.barrine.hpcu.uq.edu.au:/home/ /home/
$usecp barrine*.hpcu.uq.edu.au:/home2/ /home2/
$usecp *.barrine.hpcu.uq.edu.au:/home2/ /home2/
$usecp barrine*.hpcu.uq.edu.au:/home3/ /home3/
$usecp *.barrine.hpcu.uq.edu.au:/home3/ /home3/
$usecp barrine*.hpcu.uq.edu.au:/home4/ /home4/
$usecp *.barrine.hpcu.uq.edu.au:/home4/ /home4/
$usecp barrine*.hpcu.uq.edu.au:/panfs/imb /panfs/imb
$usecp *.barrine.hpcu.uq.edu.au:/panfs/imb /panfs/imb
$usecp barrine*.hpcu.uq.edu.au:/work1 /work1/
$usecp *.barrine.hpcu.uq.edu.au:/work1/ /work1/
$usecp barrine*.hpcu.uq.edu.au:/work2/ /work2/
$usecp *.barrine.hpcu.uq.edu.au:/work2/ /work2/
$usecp barrine*.hpcu.uq.edu.au:/HPC/home /HPC/home/
$usecp *.barrine.hpcu.uq.edu.au:/HPC/home/ /HPC/home/
$usecp barrine.hpcu.uq.edu.au:/PROJ/ /PROJ/
$usecp *.barrine.hpcu.uq.edu.au:/PROJ/ /PROJ/
$usecp barrine.hpcu.uq.edu.au:/ebi/home /ebi/home
$usecp *.barrine.hpcu.uq.edu.au:/ebi/home /ebi/home
$usecp barrine.hpcu.uq.edu.au:/ebi/bscratch /ebi/bscratch
$usecp *.barrine.hpcu.uq.edu.au:/ebi/bscratch /ebi/bscratch

# Dynamic host-level resource definitions - added 13/02/10
# /tmp on tmpfs filesystem in memory (reports bytes free)
##localtmp !/opt/sw/sys/pbs/pbsres_diskspaceavail.bash /tmp
# /scratch on local disk (report bytes free)
##scratch !/opt/sw/sys/pbs/pbsres_diskspaceavail.bash /scratch
#scratch !/usr/local/bin/diskspace /scratch

ip address
10.120.12.50

# GigE Login Node Entries
10.120.12.50    barrine1.barrine.hpcu.uq.edu.au barrine1-ge     barrine1


uqczhan2@barrine1:~> df -Hh
df: `/home3/uqmmallo/.gvfs': Permission denied
df: `/home/uqdgree5/.gvfs': Permission denied
df: `/home/uqdgree5/MappedDrives': Permission denied
Filesystem                         Size  Used Avail Use% Mounted on
/dev/sda2                           25G   16G  9.1G  64% /
udev                                24G  232K   24G   1% /dev
tmpfs                               24G   12K   24G   1% /dev/shm
/dev/sda1                          504M   60M  420M  13% /boot
/dev/sda5                          1.6T   21G  1.5T   2% /scratch
panfs://10.150.250.193/acceptance   92T   71T   21T  78% /panfs/acceptance
panfs://10.150.250.193/sw          2.8T  2.1T  696G  76% /sw
panfs://10.150.250.193/home         28T   12T   17T  42% /home
panfs://10.150.250.193/home2        28T  6.5T   21T  24% /home2
panfs://10.150.250.193/home3        28T  6.5T   21T  24% /home3
panfs://10.150.250.193/home4        28T  6.5T   21T  24% /home4
panfs://10.150.250.193/work1        27T   23T  3.6T  87% /work1
panfs://10.150.250.193/work2        92T   71T   21T  78% /work2
ebiserver:/bscratch                200T  174T   26T  87% /ebi/bscratch
cirrus:/HPC/backup1                2.0T  1.3T  645G  68% /HPC/backup1
cirrus:/HPC/home                   2.0T  1.3T  720G  65% /HPC/home
paroo3:/var/spool/PBS/sched_logs    99G   55G   39G  59% /var/spool/PBS/sched_logs
cirrus:/PROJ/jarrah                1.2T 1002G  170G  86% /PROJ/jarrah

######panfs is panasas purchased from outside

#################################################################
this is machine for computing


Filesystem                         Size  Used Avail Use% Mounted on
/dev/sda2                           27G  8.4G   18G  32% /
udev                               509G  205k  509G   1% /dev
tmpfs                              509G     0  509G   0% /dev/shm
/dev/sda1                          529M   63M  440M  13% /boot
/dev/sda5                          555G   23G  532G   5% /scratch
panfs://10.150.250.193/acceptance  101T   78T   23T  78% /panfs/acceptance
panfs://10.150.250.193/sw          3.0T  2.3T  747G  76% /sw
panfs://10.150.250.193/home         30T   13T   18T  42% /home
panfs://10.150.250.193/home2        30T  7.2T   23T  24% /home2
panfs://10.150.250.193/home3        30T  7.2T   23T  24% /home3
panfs://10.150.250.193/home4        30T  7.2T   23T  24% /home4
panfs://10.150.250.193/work1        29T   26T  3.9T  87% /work1
panfs://10.150.250.193/work2       101T   78T   23T  78% /work2
ebiserver:/bscratch                220T  191T   29T  87% /ebi/bscratch


uqczhan2@b11a07:/var/spool/PBS/mom_priv> cat config
$clienthost paroo3
cpuset_create_flags 0
$restrict_user_maxsysid 499
$tmpdir /scratch
$jobdir_root /scratch
$usecp barrine*.hpcu.uq.edu.au:/home /home
$usecp *.barrine.hpcu.uq.edu.au:/home /home



1.Problem:
qsub job
qsub: Bad UID for job execution MSG=User szhang does not exist in server password file
   solve:
 echo "+::::::" >> /etc/passwd
reference, the setup of  the barrine
solution:
  Make sure /ect/hosts.equiv has the machine name on it. See torque manual on this.

2. Problem
add default shell as bash in phpLDAPadmin


3.

ldapsearch -x -b 'dc=macondo04,dc=eait,dc=uq,dc=edu,dc=au' 'objectclass=*'

4 can not change password


5. how to add


6. the phpLDAPadmin can not be accessed by all the users. (which propbably is a good idea)



7. maui libtorque is not working at the moment.

one library called libtorque or pbs-config is missing  15-01-04

8. who. check who is using the server.

No comments:

Post a Comment