--------------- NB, condor is now in umatlas repo, and it along with yum-priorites ------------------
-------------- are now installed via cf3 ---------------
------------  epel should be installed as well ---------------------

[root@gate02 ~]# cat > /etc/yum.repos.d/rocks-local.repo
[Rocks-6.1]
name=Rocks 6.1
baseurl=http://192.41.231.100/install/rocks-dist/x86_64
enabled = 0

[root@gate02 ~]# yum --enablerepo=Rocks-6.1 --nogpgcheck install condor

================================================================================
 Package              Arch        Version                  Repository      Size
================================================================================
Installing:
 condor               x86_64      7.8.8-110288             Rocks-6.1       27 M
Installing for dependencies:
 perl-Date-Manip      noarch      6.24-1.el6               Rocks-6.1      1.3 M
 perl-Time-HiRes      x86_64      4:1.9721-130.el6_4       Rocks-6.1       46 k
 perl-YAML-Syck       x86_64      1.07-4.el6               Rocks-6.1       74 k

cf-agent -K   (no output, did it do anything?)

[root@gate02 ~]# rpm -qa|grep epel
epel-release-6-8.noarch                   THIS IS UP TO DATE
[root@gate02 ~]# yum install yum-priorities

--------------- end of condor, epel and yum-priorities changes -------------------

rpm -Uvh http://repo.grid.iu.edu/osg-el6-release-latest.rpm
Edit repo to set enabled=0

[root@gate02 ~]# service cfengine3 stop

-------------- Note, this is now enshrined in cf3 ---------------------
Added rsv and cndrcron entries to /etc/passwd and /etc/group as on gate04
Copied rsv account files (/var/rsv) from gate04 to gate02

[gate01:config.d]# grep -e rsv -e cndr /etc/passwd
rsv:x:305:306:RSV monitoring:/var/rsv:/bin/sh
cndrcron:x:306:307:Condor-cron service:/var/lib/condor-cron:/sbin/nologin
[gate01:config.d]# grep -e rsv -e cndr /etc/group
rsv:x:306:
cndrcron:x:307:

---------------- End of cf3 enshrinement ------------------

[root@gate02 ~]# yum --enablerepo=osg install empty-ca-certs

[root@gate02 ~]# yum --enablerepo=osg install osg-ce-condor

[root@gate02 ~]# yum --enablerepo=osg install globus-gram-job-manager-managedfork

Fussed a lot with monit as wrong rpm was installed, and it would not cleanly monitor.
May now be resolved.

Sort of a Catch-22.  http certs won't correctly install because http is not installed until the
osg-ce-condor rpm is installed.  Then, the only way to get them updated, is to either 
change the httpcert and httpkey in the repo (nope, not gonna happen), or remove all of the 
/var/cfengine/stage/http*.pem files.

Then, "cf-agent -K" will (re)start httpd with the certs in place.

autofs did not start.  Started it.
Failing though, reboot machine.
OK now

-------------- NB, do this instead of reboot ------------
   45  service rpcbind restart
   46  service rpcidmapd restart
   47  service autofs restart
----------------------------------------------------------

[root@gate02 ball]# osg-configure -c
[root@gate02 ball]# osg-configure -v

[root@gate02 ball]# gums-host-cron

[root@gate02 grid-security]# service gums-client-cron start
Enabling periodic gums-host-cron:                          [  OK  ]
[root@gate02 grid-security]# chkconfig gums-client-cron on
[root@gate02 grid-security]# service globus-gatekeeper start
Started globus-gatekeeper                                  [  OK  ]
[root@gate02 grid-security]# chkconfig globus-gatekeeper on
[root@gate02 grid-security]# service globus-gridftp-server start
Starting globus-gridftp-server:                            [  OK  ]
[root@gate02 grid-security]# chkconfig globus-gridftp-server on
[root@gate02 grid-security]# service tomcat6 start
Starting tomcat6:                                          [  OK  ]
[root@gate02 grid-security]# chkconfig tomcat6 on
[root@gate02 grid-security]# service gratia-probes-cron start
Enabling gratia probes cron:                               [  OK  ]
[root@gate02 grid-security]# chkconfig gratia-probes-cron on
You have new mail in /var/spool/mail/root
[root@gate02 grid-security]# service osg-cleanup-cron start
Enabling periodic osg-cleanup:                             [  OK  ]
[root@gate02 grid-security]# chkconfig osg-cleanup-cron on
[root@gate02 html]# chkconfig httpd on
(httpd was started by the osg startup)

Problems with condor.pm on Fork jobs.  They are getting all the requirements 
of a standard job, and are therefore not running.

FIXED, changes in condor_submit

Running on gate04:  rsv-control --run --all-enabled
Failed a bunch, need to do:
[root@gate02 grid-security]# yum install sharutils

-- BobBall - 02 Jul 2013
Topic revision: r4 - 18 Sep 2013, BobBall
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback