shlock - Create lock files for use in shell scripts
shlock [-b|-c|-u] -f name -p
pid
shlock tries to create a lock file named name and write the
process ID pid into it. If the file already exists, shlock will
read the process ID from the file and test to see whether the process is
currently running. If the process exists, then the file will not be created.
shlock exits with a zero status if it could create the lock file, or
non-zero if the file refers to a currently active process.
A Perl wrapper around shlock can be used via the
"INN::Utils::Shlock" module.
- -b
- Process IDs are normally read and written in ASCII. If the -b flag
is used, then they will be written as a binary int.
- -c
- If the -c flag is used, then shlock will not create a lock
file, but will instead use the file to see if the lock is held by another
program. If the lock is valid, the program will exit with a non-zero
status; if the lock is not valid (i.e. invoking shlock without the
flag would have succeeded), then the program will exit with a zero
status.
- -f name
- name is the name of the lock file shlock attempts to create.
If the file already exists, it will read the process ID from the file and
exit with a non-zero status if this process is currently active.
- -p pid
- pid is the process ID to write into the file name.
- -u
- For compatibility with other systems, the -u flag is accepted as a
synonym for -b since binary locks are used by many UUCP
packages.
The following example shows how shlock would be used within a shell
script:
LOCK=<pathrun in inn.conf>/LOCK.send
trap 'rm -f ${LOCK} ; exit 1' 1 2 3 15
if shlock -p $$ -f ${LOCK} ; then
# Do appropriate work.
else
echo "Locked by `cat ${LOCK}`"
fi
Written by Rich $alz <rsalz@uunet.uu.net> for
InterNetNews after a description of HDB UUCP locking given by Peter Honeyman,
and improved by Berend Reitsma to solve a race condition. Converted to POD by
Julien Elie.