A very simple script to list all available scram architectures for a specified CMSSW release number. Requires cvmfs.
Parse the output of the CMSSW SimpleMemoryCheck service to determine the peak VSIZE and RSS values from a given process log.
Tracks the progress of compilation with scram, assuming the scram printouts are redirected into a log file. Output looks like:
scram status:
in progress: 10 / 42
finished: 32 / 42
The denominator counts all packages in the release area as directories matching */*, e.g. "foo/bar".
"finished" counts all packages which scram has noted as "built".
"in progress" counts all packages which scram has noted as "Compiling" (but not yet "built").
Efficient and safe disk usage script for the EOS filesystem.
eosdu [options] <LFN>
Options:
-s xrootd server name (default = root://cmseos.fnal.gov)
-f count number of files instead of disk usage
-g search for files matching specified string within directory
-h print human readable sizes
-r run eosdu for each file/directory in <LFN>
('recursive'/'wildcard' option, like 'du *')
Script to copy a file over xrootd to specified directory (local area by default), keeping the entire LFN as the local filename ('/' replaced with '_'). Arguments:
-f: usexrdcp -f-q: usexrdcp -q-x [redir]: xrootd redirector (required)-L [lfn]: logical filename (LFN) (required)-o [dir]: output directory (default =./)
A very simple script to sort the output of condor_userprio -allusers by the usage column.
Sums up the total number of jobs (with each status) over all accessible schedulers, for specified user:
./condorq_all.sh [username]
Monitors overall status of running, idle, and held jobs on the LPC batch farm.
PAGER='less -S' jobmon.py
Date + status = datus, i.e. git status -s output sorted by date modified.
Supports all untracked file display options of git status.
git datus [options]
Options:
-u[mode] show untracked files (mode = no, normal, all)
-l,--long show long timestamp
-h,--help show this message and exit
-- separator; anything after this will be passed to git status
Squash the last N commits, i.e. git squash HEAD~N.
git squash [options] <REF>
Options:
-k, --keep keep commit messages from squashed commits
Overwrites the current branch and working area with the branch version from a specified remote. Can optionally stash and pop/apply uncommitted changes in the working area. Useful when developing on one server and testing on another.
git sync [options] <remote>
Options:
-s, --stash stash any uncommitted changes before syncing
-p, --pop pop stashed changes after syncing
-a, --apply apply stashed changes after syncing
Wrapper to run a command that creates a new file, and then edit the new file. Usage:
cmde COMMAND [cmde options] [--] [COMMAND options] FILE1 FILE2Options:
-e [editor]: text editor (default:nano)-f: overwrite FILE2 (if it already exists)-h: display this message and exit--: optional separator (in case COMMAND uses same flags ascmde)
cmde does not work with all forms of all commands, but just those that end with FILE1 FILE2 (i.e. SRC DEST) syntax.
It is most typically used with cp (but could also be used with mv, etc.). Useful alias:
alias cpe='cmde cp'Fast deletion of a directory containing many files, using rsync.
Script to use diff side-by-side view (diff -y) with automatic width (max line length from input files), piped to less -S (no word wrap). Options:
-M [num]- specify max width-L- shorthand for--left-columndiff option
All other flags are passed directly to the diff command. (-t is enabled by default, so passing it explicitly instead disables it.).
Script to run igprof and produce a report. Currently only supports performance profiling and ASCII reports. If a name is specified with no associated command but with sorting options, it will try to locate the corresponding report and produce sorted reports. Options:
-e [command]- command to profile (enclose in "" if contains spaces)-n [name]- name for output files (default = test)-t [exe]- limit profiling to specified target executable-s [modules]- produce sorted reports of contributions, one for each module (comma-separated list)-d [modules]- produce sorted reports of contributions, one for each module's descendants (comma-separated list)-x [args]- any extra arguments to igprof (quoted)-r- profile a ROOT macro, prependsroot.exe -b -l -qto specified command (avoid quote nesting)-c- special settings forcmsRun(use-t cmsRun, sort descendants ofdoEvent)-h- print help message and exit
Script to set up a grid certificate for use on a remote server. Usage:
setupCert.sh [options] [user@server] [p12 file]
Options:
-c set up for cern-get-sso-cookie
-d dry run (just print commands, don't execute)
A simple utility to create and manage SSH tunnels.
tunn [operation] [options] [arguments]
Default settings are obtained from the config file at /home/pedrok/.tunnconfig.
To override the config file location, put this in your .bashrc or other login file:
export TUNN_CONFIG=/my/preferred/file
The available config variables are: TUNN_PREFIX, TUNN_PORT, TUNN_VERBOSE.
Their values should be specified in the config file using bash syntax, e.g.:
TUNN_PORT=8XXX
(If TUNN_PORT is not specified in the config file or via the command line option,
the default value is taken from the last three digits of your UID.)
Operations:
make make new tunnel
-n [name] tunnel socket name prefix (default: /home/pedrok/.tsock)
-p [port] tunnel port (default: 8864)
[destination] ssh destination for tunnel (required)
list list open tunnels
kill kill specified tunnel
[index] index of tunnel (required)
Common options:
-u (unclean) do not auto-remove closed tunnels from list
-v toggle verbosity (default: false)
-h print this message and exit
Useful aliases:
alias tmake='tunn make'
alias tlist='tunn list'
alias tkill='tunn kill'
Script to use wget with cern-get-sso-cookie (ref),
to download files protected by CERN single sign on.
CERN SSO authentication can proceed via kerberos (default) or via user certificates
with the arguments --cert and --key (see above link for more info).