I want know what am I doing wrong in this code:
#!/bin/sh
SERVICE_NAME=neocloud
PATH_TO_JAR=/etc/neocloud/cloud.jar
PID_PATH_NAME=/tmp/neocloud-pid
case $1 in
start)
echo "Starting $SERVICE_NAME ..."
if [ ! -f $PID_PATH_NAME ]; then
nohup java -jar $PATH_TO_JAR /tmp 2>> /dev/null >> /dev/null &
echo $! > $PID_PATH_NAME
echo "$SERVICE_NAME started ..."
else
echo "$SERVICE_NAME is already running ..."
fi
;;
stop)
if [ -f $PID_PATH_NAME ]; then
PID=$(cat $PID_PATH_NAME);
echo "$SERVICE_NAME stoping ..."
kill $PID;
echo "$SERVICE_NAME stopped ..."
rm $PID_PATH_NAME
else
echo "$SERVICE_NAME is not running ..."
fi
;;
restart)
if [ -f $PID_PATH_NAME ]; then
PID=$(cat $PID_PATH_NAME);
echo "$SERVICE_NAME stopping ...";
kill $PID;
echo "$SERVICE_NAME stopped ...";
rm $PID_PATH_NAME
echo "$SERVICE_NAME starting ..."
nohup java -jar $PATH_TO_JAR /tmp 2>> /dev/null >> /dev/null &
echo $! > $PID_PATH_NAME
echo "$SERVICE_NAME started ..."
else
echo "$SERVICE_NAME is not running ..."
fi
;;
esac
executing via sh I get the following error message:
Syntax error: word unexpected (expecting "in")
But in the case command I've the in word.
Anyone know how to fix this bug?
Thanks a lot!
As posted here, your code is syntactically valid POSIX-like shell code, which you can verify at shellcheck.net (which will, however, warn you about potentially unwanted side effects due to not double-quoting your variable references (e.g., echo $PID_PATH_NAME rather than echo "$PID_PATH_NAME"), which however does NOT apply to $1 in the case statement[1]
).
Similarly, copying the code in your question and pasting it into a new file and using sh on Ubuntu (which is Dash) to execute it, works fine too.
Thus - unless your sh is not what it should be - I suspect that you have "weird" characters in your shell file, such as Unicode whitespace outside the standard ASCII range, which may look like normal whitespace, but isn't; the Unicode no-break space character (U_00A0, UTF8-encoded as 0xC2 0xA0) is an example.
To look for such characters, run the following (where script represents your script):
LC_ALL=C cat -e script
and look for M- and ^<letter> sequences in the output; for instance, the aforementioned no-break space shows up as M-BM-.
[1] Double-quoting the argument given to the case statement doesn't hurt, but is not necessary.
While unquoted parameter/variable references are word-split and pathname-expanded in most places in POSIX-like shells, case is a curious exception.
The following demonstrates this, and works with all major POSIX-like shells (dash, bash, ksh, zsh):
$ sh -c 'case $1 in "foo *") echo "match";; *) echo "nomatch"; esac' - 'foo *'
match
Literal argument foo * matches the case branch, even though $1 is unquoted.
(Contrast this with the typical situation (e.g., echo $1), where the value of $1 would be subject to both word-splitting and pathname expansion (globbing).)
Have you tried quoting the argument?
case "$1" in
...
Related
I am working on a script that take 1 string argument and a file. I want it so that if a file is put in that doesn't exist, then it will display the "filename cannot be read" message.
That part does work however it also displays a "grep: grep.txt: No such file or directory" message. Is there any way to stop the grep message from posting and ending the script if the first if statement is true?
#! /bin/sh
if [ ! -f "$2" ]
then
echo "$0" cannot be read 1>&2
fi
if [ $# -eq 2 ]
then
grep "$1" $2
else
echo there is more or less than 2 arguments 1>&2
fi
Exit the script with a non-zero exit code to indicate failure and stop it from continuing on to the grep.
if [ ! -f "$2" ]
then
echo "$0" cannot be read 1>&2
exit 1
fi
You can add /dev/null in grep command it will suppress the error part.
grep "$1" $2 2>/dev/null
The > operator redirects the output usually to a file but it can be to a device. You can also use >> to append.
2> file redirects stderr to file
/dev/null is the null device it takes any input you want and throws it away. It can be used to suppress any output.
You could redirect all errors from grep, for example:
grep "$1" $2 2>/dev/null
(the 2> means redirect standard error, as opposed to standard output with > or 1>).
That introduces a race condition, however: if the file disappears while your script as running, it might still exist when you check that it exists, but be gone by the time grep runs.
You could handle that by checking the exit status...
grep "$1" $2 2>/dev/null
if [[ $? -gt 1 ]]; then
echo "grep failed unexpectedly" >&2
fi
IMHO, in this example it would be better to just let grep print the error.
I have one shell script.
#
. ./shprofile
if [ "$1" != "INSERT_ALL_ITEMS" -a "$1" != "INSERT_ONE_ITEM" ]
then
echo "Usage: $0 [INSERT_ALL_ITEMS|INSERT_ONE_ITEM]"
exit
fi
#
echo "Start Date & Time is.." `date`
case "$1" in
INSERT_ALL_ITEMS)
echo "Executing all Items"
;;
INSERT_ONE_ITEM)
#
echo "Executing one Item"
;;
*)
echo "Invalid Options!!"
;;
esac
#
echo "Stop Date & Time is.." `date`
If I execute the above script with below command
./runItemsBatch.sh INSERT_ALL_ITEMS
Then it is showing echo message, instead of executing the script
Usage: ./runItemsBatch.sh [INSERT_ALL_ITEMS|INSERT_ONE_ITEM]
What is the wrong in script? Why it is not executing the script, even though I give correct option.
Need help.
Given that your script ran fine on my CentOS 7 virtual machine (commenting the reference to the other file), I think it may be that you didn't set up the interpreter properly.
The first line (#) should be a shabang (#!) followed by the interpreter. e.g.: #!/bin/sh.
You could also try to remove the quotes for the arguments in your script because the interpreter could ...interpret them as literals.
I mean that you should try to replace the line
if [ "$1" != "INSERT_ALL_ITEMS" -a "$1" != "INSERT_ONE_ITEM" ]
with
if [ $1 != "INSERT_ALL_ITEMS" -a $1 != "INSERT_ONE_ITEM" ]
Try with either this two things. In my case however, the script worked with every combination of "shabang / no shabang and quotes / no quotes"
Hope I helped.
EDIT: Try also to delete the spaces between variables and operators in the if (Again, these are rules depending by the interpreter you are using, or defaulted to use)
I have BBB based custom Embedded Linux based board with busybox shell(ash)
I have a situation where my script must run in background with following condition
There must only one instance of the script.
wrapper script need to know if script started successfully in background or not.
There is another wrapper script which starts and stops my script, wrapper script is as mentioned below.
#!/bin/sh
export PATH=/bin:/sbin:/usr/bin:/usr/sbin
readonly TEST_SCRIPT_PATH="/home/testscript.sh"
readonly TEST_SCRIPT_LOCK_PATH="/var/run/${TEST_SCRIPT_PATH##*/}.lock"
start_test_script()
{
local pid_of_testscript=0
local status=0
#Run test script in background
"${TEST_SCRIPT_PATH}" &
#---------Now When this point is hit, lock file must be created.-----
if [ -f "${TEST_SCRIPT_LOCK_PATH}" ];then
pid_of_testscript=$(head -n1 ${TEST_SCRIPT_LOCK_PATH})
if [ -n "${pid_of_testscript}" ];then
kill -0 ${pid_of_testscript} &> /dev/null || status="${?}"
if [ ${status} -ne 0 ];then
echo "Error starting testscript"
else
echo "testscript start successfully"
fi
else
echo "Error starting testscript.sh"
fi
fi
}
stop_test_script()
{
local pid_of_testscript=0
local status=0
if [ -f "${TEST_SCRIPT_LOCK_PATH}" ];then
pid_of_testscript=$(head -n1 ${TEST_SCRIPT_LOCK_PATH})
if [ -n "${pid_of_testscript}" ];then
kill -0 ${pid_of_testscript} &> /dev/null || status="${?}"
if [ ${status} -ne 0 ];then
echo "testscript not running"
rm "${TEST_SCRIPT_LOCK_PATH}"
else
#send SIGTERM signal
kill -SIGTERM "${pid_of_testscript}"
fi
fi
fi
}
#Script starts from here.
case ${1} in
'start')
start_test_script
;;
'stop')
stop_test_script
;;
*)
echo "Usage: ${0} [start|stop]"
exit 1
;;
esac
Now actual script "testscript.sh" looks something like this,
#!/bin/sh
#Filename : testscript.sh
export PATH=/bin:/sbin:/usr/bin:/usr/sbin
set -eu
LOCK_FILE="/var/run/${0##*/}.lock"
FLOCK_CMD="/bin/flock"
FLOCK_ID=200
eval "exec ${FLOCK_ID}>>${LOCK_FILE}"
"${FLOCK_CMD}" -n "${FLOCK_ID}" || exit 0
echo "${$}" > "${LOCK_FILE}"
# >>>>>>>>>>-----Now run the code in background---<<<<<<
handle_sigterm()
{
# cleanup
"${FLOCK_CMD}" -u "${FLOCK_ID}"
if [ -f "${LOCK_FILE}" ];then
rm "${LOCK_FILE}"
fi
}
trap handle_sigterm SIGTERM
while true
do
echo "do something"
sleep 10
done
Now in above script you can see "---Now run the code in background--" at that point I am sure that either lock file is successfully created or instance of this script is already running. So Then I can safely run other code in background and wrapper script can check for lockfile and find out if the process mentioned in the lock file is running or not.
can shellscript itself make it to run in background ?
if not is there a better way to meet all the conditions ?
I think you can look into job control built-in, specifically bg.
Job Control Commands
When processes say they background themselves, what they actually do is fork and exit the parent. You can do the same by running whichever commands, functions or statements you want with & and then exiting.
#!/bin/sh
echo "This runs in the foreground"
sleep 3
while true
do
sleep 10
echo "doing background things"
done &
I'm trying to run programs (for example mv file1.txt file2.txt) in my .sh script and I need to hide errors, but handle it with my script.
Currently I'm trying to do something like
EXECUTE="mv -v $VOL $BACKUP_YESTERDAY_CRYPT"
{
EXEC_ERROR=$($EXECUTE)
} &2>> $LOG_FILE
if [[ -n $EXEC_ERROR ]]; then
echo "There is an error!"
fi
But it doesn't work at all - it shows an error (for example mv: cannot stat 'file1.txt': No such file or directory) and $EXEC_ERROR variable is empty.
Is there any way to get output to variable + to log file?
How about something like:
mv -v $VOL $BACKUP_YESTERDAY_CRYPT 2>> $LOG_FILE
if [[ ! ( $? -eq 0 ) ]] ; then
echo "There is an error\!"
fi
Though $? is good for saving and processing exit codes, the if statement is designed to take any command, not just [ or [[:
if ! mv -v "$VOL" "$BACKUP_YESTERDAY_CRYPT" 2>> $LOG_FILE; then
echo "There is an error!"
fi
This includes saving variables:
if OUTPUT=$(mv -v "$VOL" "$BACKUP_YESTERDAY_CRYPT" 2>> $LOG_FILE); then
echo ">>> $OUTPUT <<<"
fi
In fact, if can take more than one command, as its man page describes. Documentation on boolean operators such as !, &&, and || is hidden within the description of shell commands, where they form pipelines (!) and lists (&&, ||).
Try this:
mv sourcefile destfile 2> /dev/null 1>logfile
returnstatus=`echo $?`
if [[ $returnstatus -ne 0 ]]; then
echo "There was an error!"
fi
I need a simple busybox sh wrapper which will do:
IF "-Q" PARAMETER IS PROVIDED THEN
acommand ALL PARAMETERS BUT "-Q" 2>&1 1>/dev/null
ELSE
acommand ALL PARAMETERS
FI
Parameters may include spaces.
BTW I want to run the script with busybox sh and it doesn't support arrays.
It's possible to do it all in busybox's ash shell:
#!/bin/sh
for i in "${#}"
do
if [ "$i" = "-Q" ]
then
flagQ=1
else
args="$args \"$i\""
fi
done
if [ "$flagQ" = "1" ]
then
eval acommand "$args" 2>&1 1>/dev/null
else
eval acommand "$args"
fi
This uses bash arrays - but I see from the comments to another answer that the code isn't supposed to run under bash (despite the bash tag originally applied to the question); it is meant to run under the busybox shell.
I'm almost certain it doesn't answer the question because the question is substantially unanswerable given the limitations of busybox. In times past, I have used a custom program I called 'escape' to build up an argument string that can be eval'd to get the original arguments - spaces and all. But that requires support from outside the shell.
This solution only uses 'bash'. I'm not sure it is fully idiomatic bash code, but it works.
#!/bin/bash
i=0
Qflag=0
for arg in "$#"
do
if [ "X$arg" = "X-Q" ]
then Qflag=1
else args[$((i++))]=$arg
fi
done
if [ $Qflag = 1 ]
then exec acommand "${args[#]}" 2>&1 >/dev/null
else exec acommand "${args[#]}"
fi
The first loops builds up an array, args, with the arguments to the script, except it doesn't add '-Q' to the list and records its presence in variable Qflag.
The if statement at the end notes whether Qflag was set to 1, and if so, sends the errors from 'acommand' to standard output and sends regular standard output to /dev/null (which is different from the effect if the I/O redirections are reversed - that would send standard output to /dev/null and send standard error to the same place, forcing silence on 'acommand').
The use of 'exec' is a trivial optimization that simplifies exit status handling in this case.
Tested with 'acommand' that prints its arguments on separate lines:
#!/bin/sh
for arg in "$#"
do echo "$arg"
done
and with command lines such as:
bash wrapper.sh -c -d 'arg with spaces'
which produces the output:
-c
-d
arg with spaces
Obviously, with the I/O redirection in place, there is no output from:
bash wrapper.sh -c -Q -d 'arg with spaces'
However, if you omit the I/O redirection, you get to see the same output.
It's a pity that you need to handle spaces in the arguments otherwise this might work:
#!/bin/sh
Q=0
ARGS=
while [ $# -ge 1 ]; do
case $1 in
-Q)
Q=1
;;
*)
ARGS="$ARGS $1"
;;
esac
shift
done
if [ $Q -eq 1 ] ; then
acommand $ARGS 2>&1 1>/dev/null
else
acommand $ARGS
fi
EDIT:
So this version handles spaces, at the expense of interpreting back-ticks.
#!/bin/busybox ash
Q=0
ARGS=
while [ $# -ge 1 ]; do
case $1 in
-Q)
Q=1
;;
*)
ARGS="$ARGS \"$1\""
;;
esac
shift
done
if [ "$Q" -eq 1 ] ; then
eval acommand $ARGS 2>&1 1>/dev/null
else
eval acommand $ARGS
fi
I think to have a complete solution you are going to have to code it in C, which will be a bit ugly.