This page describes how to have your environment configured to run ATLAS jobs at SLAC. It also tries to explain where you may need to do something different.
For first time SLAC computing users, you need to first register as a SLAC user and obtain a SLAC computing account. The initial account setup related info can be found on the US Western Tier-2 page under the link "SLAC UNIX users", which also contains links to info on batch system and printing etc. Other general computing services and information provided by the SLAC Scientific Computing and Computing Services (SCCS) can be found at the general "SLAC computing environment" page. There is also some information on configuring ssh to avoid using passwords multiple times. Among the various Public Machines at SLAC, the recommended interactive machines to login are yakut's (yakut@slac.stanford.edu), which are Linux nodes running SL3. If you use tcsh then you should email unix-admin@slac.stanford.edu to request /usr/etc/tcsh as your shell.
Some essential background to the ATLAS software setup at SLAC:
If you have any questions please post them to the Non-Grid Jobs at SLAC Forum in the ATLAS Hypernews system, where all local ATLAS computing issues at SLAC are discussed.
To turn on the ATLAS login you can do it one of two ways, below. Next time you login the ATLAS commands should be in your path. You need to use the HEPiX login to get the general SLAC setup (so things like LSF can be used).
In this case there are three steps required to get it working correctly. If you've already set your account up for HEPiX login (because you use to work on BaBar for instance) then you only need to do that last step.
#####################################################
# Do HEPiX login
if ( -r /usr/local/lib/hepix/central_env.csh ) then
source /usr/local/lib/hepix/central_env.csh
endif
#####################################################
or (if you use a bourne based shell) the following to
your .bashrc;
#####################################################
# Do HEPiX login
if [ -r /usr/local/lib/hepix/central_env.sh ]; then
source /usr/local/lib/hepix/central_env.sh
fi
#####################################################
###########################################################
# Do HEPiX login
if ( -r /usr/local/lib/hepix/central_login.csh ) then
source /usr/local/lib/hepix/central_login.csh
endif
###########################################################
or (if you use a bourne based shell) the following to
your .profile;
###########################################################
# Do HEPiX login
if [ -r /usr/local/lib/hepix/central_login.sh ]; then
source /usr/local/lib/hepix/central_login.sh
fi
###########################################################
[noric06] ~ > cat .hepix/preferred-group atlas
For this to work you add the following line to your .login file if you use a c based shell:
source /afs/slac.stanford.edu/g/atlas/etc/atlas_env.csh
or the following for a bourne based shell;
. /afs/slac.stanford.edu/g/atlas/etc/atlas_env.sh
This command can be used to switch releases. The only argument it takes is the release you want to use (eg "atlsetup 12.2.0").
Amir Farbin wrote a script he called addpkg (to mimic a BaBar script of the same name) to allow you to check out a package without knowing the tag if you wanted the same version as was in the release. I've created a new version that works using a cmt command (suggested by Zach Marshal).
The default release is the "current" released (defined by a symlink
in $ATLROOT/b/AtlasOffline). If someone wants to choose a different release
by default they should create a file in their home directory called ATLCURRENT with the release number in it (or "latest" to use the
latest symlink in $ATLROOT/b/AtlasOffline). You can in principle also do "atlsetup
Information on this is available on another page now.
If your user name is different between SLAC and CERN you'll need to set CVSROOT to include your CERN user name. By default it is set to :ext:atlas-sw.cern.ch:/atlascvs. You should set it with a command like;
setenv CVSROOT :ext:<CERNusername>@atlas-sw.cern.ch:/atlascvs
if you use a c-shell or;
CVSROOT=:ext:<CERNusername>@atlas-sw.cern.ch:/atlascvs export CVSROOTif your use a bourne based shell. You should of course replace <CERNusername> with your CERN username.
As not all machines have grid certificates installed so you need to edit your .srmconfig/config.xml file after it gets created. In there you'll find an entry for x509_user_trusted_certificates which will point to a non-existent directory in /etc. You should instead change it to point to /afs/slac.stanford.edu/package/globustools/vdtclient/curr/globus/TRUSTED_CA.