Tera-scale Open-source Resource and QUEue manager
Go to file
2018-10-16 09:10:35 +01:00
.gitignore Merge branch 'master' into el6 2015-04-07 20:55:09 -07:00
config forgot to add config on EL-6 branch. 2010-07-14 21:18:58 +00:00
CVE-2013-4319.patch Fix CVE-2013-4319 (RHBZ #1005918, #1005919) 2014-10-02 09:42:45 +02:00
CVE-2013-4495.patch Fix CVE-2013-4495 (RHBZ #1029752) 2014-09-05 04:54:28 +02:00
mom.layout Fix requires for systemd and add sample mom.layout 2016-04-09 05:18:00 -07:00
pbs_mom.service lets make the service names the old names since its been that 2015-05-18 20:31:12 -07:00
pbs_sched.service lets make the service names the old names since its been that 2015-05-18 20:31:12 -07:00
pbs_server.service lets make the service names the old names since its been that 2015-05-18 20:31:12 -07:00
pbs-config-multilib - Remove own copy of pbs-config. #657027 2010-12-09 22:54:47 +01:00
README.Fedora - Enable or rather force munge support. 2010-12-10 20:36:38 +01:00
sources updated upstream source 2015-04-06 18:55:39 -07:00
torque-2.5.7-rhbz#759141-r5167-pbs_server-crash.patch Add torque-2.5.7-rhbz#759141-r5167-pbs_server-crash.patch 2011-12-03 19:38:21 +01:00
torque-4.2.10-fix-bad-crypto-check.patch Cleanup spec file conditionals 2018-01-11 16:13:07 -06:00
torque-4.2.10-port-args.patch get the right args to the momctl bits 2016-02-20 21:21:26 -08:00
torque-4.2.10-remove-unused-header.patch Cleanup spec file conditionals 2018-01-11 16:13:07 -06:00
torque-buffer-overrun-2.5.5.patch Add torque-buffer-overrun.patch rhbz#711463 2011-06-08 19:10:05 +02:00
torque-create-request.patch Add patch torque-initd-hangs-rhbz-744138.patch 2011-10-09 00:32:05 +02:00
torque-fix-munge-rhbz#752079-PTII.patch Add torque-fix-munge-rhbz#752079-PTII.patch 2011-11-21 19:46:54 +01:00
torque-fix-munge-rhbz#752079.patch Add torque-fix-munge-rhbz#752079.patch 2011-11-17 21:15:32 +01:00
torque-initd-hangs-rhbz-744138.patch Add patch torque-initd-hangs-rhbz-744138.patch 2011-10-09 00:32:05 +02:00
torque-munge-size.patch fix the torque munge patch 2014-10-14 22:57:24 -07:00
torque-rhbz#758740-r5258-dis-close.patch Add torque-rhbz#758740-r5258-dis-close.patch and 2011-11-30 19:23:52 +01:00
torque-rhbz#758740-r5270-dis-array.patch Add torque-rhbz#758740-r5258-dis-close.patch and 2011-11-30 19:23:52 +01:00
torque.spec Adjust dependencies for install requirements, few cleanups, fix up some conditionals for obsolete deps on systemd systems, kill off bits < el6 era 2018-10-16 09:10:35 +01:00
trqauthd.service here's some service scripts for systemd 2015-05-02 07:45:49 -07:00
xpbs.desktop - correct errors in desktop entry files 2007-08-16 05:03:04 +00:00
xpbs.png auto-import torque-2.1.0p0-0.7.200604171430cvs on branch devel from 2006-04-18 02:10:40 +00:00
xpbsmon.desktop - correct errors in desktop entry files 2007-08-16 05:03:04 +00:00
xpbsmon.png auto-import torque-2.1.0p0-0.7.200604171430cvs on branch devel from 2006-04-18 02:10:40 +00:00

This README describes how to get the most basic working
torque service on a single host.

To setup a basic single-node localhost-only batch system, install the
torque-server, torque-mom, and torque-scheduler packages, and do something like
this:

0) If torque is built with munge support then this 
   must be enabled first on all nodes. The munge
   package should allready be installed.

   Create a munge key with 

   /usr/sbin/create-munge-key

   Copy resulting key /etc/munge/munge.key to 
   all torque nodes in your cluster including
   pbs_server, pbs_mom and client (qstat,qsub) nodes.

1) Get your full hostname with

# /bin/hostname --long

e.g myhost.example.org

2) Edit /etc/torque/server_name 
to contain the single line

myhost.example.org

3) Edit /etc/torque/mom/config 
to contain the single line

$pbsserver myhost.example.org

4) Create a torque serverdb file.
# /usr/sbin/pbs_server -D -t create

Warning this will remove any existing serverdb 
file located at /var/lib/torque/server_priv/serverdb

You will have to Ctrl^C the pbs_server command, it will
only take a moment to create this file.

5) Start the pbs_server and configure it.
service pbs_server start
# qmgr -c "s s scheduling=true"
# qmgr -c "c q batch queue_type=execution"
# qmgr -c "s q batch started=true"
# qmgr -c "s q batch enabled=true"
# qmgr -c "s q batch resources_default.nodes=1"
# qmgr -c "s q batch resources_default.walltime=3600"
# qmgr -c "s s default_queue=batch"

6) Add one batch worker to your pbs_server.

# qmgr -c "c n myhost.example.org"

7) Start the pbs_mom and pbs_sched deamons.

# service pbs_mom start
# service pbs_sched start

8) Use chkconfig to start the services at boot time.

# /sbin/chkconfig pbs_mom on
# /sbin/chkconfig pbs_server on
# /sbin/chkconfig pbs_sched on
# /sbin/chkconfig munge on

9) Submit a test job.
As a user not as root run the following

$ qsub <<EOF
hostname 
echo "Hi I am a batch job running in torque"
EOF

10 ) Monitor the state of that job with qstat.

In case of problems first of all look in /var/log/torque