i need to redirect stout and stderr in bash each to separate file.
well i completed this command:
((/usr/bin/java -jar /opt/SEOC2/seoc2.jar 2>&1 1>&3 | tee --append /opt/SEOC2/log/err.log) 3>&1 1>&2 | tee --append /opt/SEOC2/log/app.log) >> /opt/SEOC2/log/combined.log 2>&1 &
which works fine running from a command line.
trying to put the very same command into bash script
...
12 cmd="(($run -jar $cmd 2>&1 1>&3 | tee --append $err) 3>&1 1>&2 | tee --append $log) >> $combined 2>&1"
...
30 echo -e "Starting servis..."
31 $cmd &
32 pid=`ps -eo pid,args | grep seoc2.jar | grep -v grep | cut -c1-6`
33 if [ ! -z $pid ]; then
...
leads to error like this:
root#operator:/opt/SEOC2# seoc2 start
Starting servis...
/usr/local/bin/seoc2: line 31: ((/usr/bin/java: dir or file doesn't exist
tried to cover this command by $( ), ` ` etc but with no effect at all :(
any suggestion or advice would be very appreciated, playing around for hours already :/
thanx a lot
Rene
If you store the whole command line in a variable you have to use eval to execute it:
cmd="(($run -jar $cmd 2>&1 1>&3 | tee --append $err) 3>&1 1>&2 | tee --append $log) >> $combined 2>&1"
...
eval $cmd &
Related
I try to run the following command (ref.) using nohup, which basically separates stdout and stderr into two processes.
{ foo 2>&1 1>&3 3>&- | sed -u 's/^/err: /'; } 3>&1 1>&2 | sed -u 's/^/out: /'
The foo script is like below.
#!/bin/bash
while true; do
echo a
echo b >&2
sleep 1
done
This is the test result.
$ nohup { foo 2>&1 1>&3 3>&- | sed -u 's/^/err: /'; } 3>&1 1>&2 | sed -u 's/^/out: /' >/dev/null 2>&1 &
-bash: syntax error near unexpected token `}'
That's syntatically impossible. But you can wrap your {} in a sh -c cmd:
nohup sh -c 'foo 2>&1 1>&3 3>&- | sed -u "s/^/err: /"'
Notice I change the single quote for sed to double quote.
I am writing a bash script to store the individual values of minutes and seconds.
However, I am unable to do so.
TIMEFORMAT='%3lR'
exec 3>&1 4>&2
cmd='ls -l / >/dev/null 2>&1'
var=$( { time $cmd 1>&3 2>&4; } 2>&1 ) # Captures time only.
exec 3>&- 4>&-
echo 'Get the var'
echo $var
min=$(echo $var | awk -F 'm|s' '{print $2}')
echo $min
The output shows ls -l / which I don't want.
Next, I don't understand what is exec 3>&- 4>&-. I copied this from other stack overflow answers. I don't understand what is the use of { inside the ( bracket.
Try this:
TIMEFORMAT='%3lR'
var=$({ time ls -l / 2>&1 > /dev/null; } 2>&1)
echo 'Get the var'
echo ${var}
min=$(echo ${var} | awk -F 'm|s' '{print $2}')
echo ${min}
How to print output from a loop which is piped to some other command:
for f in "${!myList[#]}"; do
echo $f > /dev/stdout # echoed to stdout, how to?
unzip -qqc $f # piped to awk script
done | awk -f script.awk
You can use /dev/stderr or second file descriptor:
echo something >&2 | grep nothing
echo something >/dev/stderr | grep nothing
You can use another file descriptor that will be connected to stdout:
# for a single command group
{ echo something >&3 | grep nothing; } 3>&1
# or for everywhere
exec 3>&1
echo something >&3 | grep nothing
# same as above with named file descriptor
exec {LOG}>&1
echo 123 >&$LOG | grep nothing
You can also redirect the output to current controlling terminal /dev/tty (if there is one):
echo something >/dev/tty | grep nothing
Consider the following example:
{ echo 1 | tee /dev/stderr 2> >(sed -e 's|1|err|' >&2) 1> >(sed -e 's|1|out|') ; }
which prints
out
err
Note that out out is printed on stdout and err on stderr.
Question: How to do this remotely via ssh?
More precisely, how to run
ssh host 'echo 1 | tee /dev/stderr SOME_MAGIC_HERE'
st. again out/err pops up on stdout/stderr (for an appropriate bash magic SOME_MAGIC_HERE).
Clearly, the following works:
ssh host 'echo 1 | tee /dev/stderr' 2> >(sed -e 's|1|err|' >&2) 1> >(sed -e 's|1|out|')
But that executes sed locally, and I'd rather want to do that remotely on host.
after the update:
ssh host 'echo 1 | tee >(cat - | sed -e "s|1|err|" >&2) | sed -e "s|1|out|"'
out
err
The idea is to use <pipes> | for processing /dev/stdout and use process substitution in combination with tee to create the /dev/stderr part.
Now it works as expected:
$ ssh host 'echo 1 | tee >(cat - | sed -e "s|1|err|" >&2) | sed -e "s|1|out|"' > /dev/null
err
$ ssh host 'echo 1 | tee >(cat - | sed -e "s|1|err|" >&2) | sed -e "s|1|out|"' 2> /dev/null
out
original answer:
The following command executes by changing your <single quotes> into <double quotes> :
ssh host 'echo 1 | tee /dev/stderr 2> >(sed -e "s|1|err|") 1> >(sed -e "s|1|out|")'
but this has everything in /dev/stdout. Example:
$ ssh host 'echo 1 | tee /dev/stderr 2> >(sed -e "s|1|err|") 1> >(sed -e "s|1|out|")' > /dev/null
$ ssh host 'echo 1 | tee /dev/stderr 2> >(sed -e "s|1|err|") 1> >(sed -e "s|1|out|")' 2> /dev/null
out
err
and this is exactly what your original command does on the host system:
{ echo 1 | tee /dev/stderr 2> >(sed -e "s|1|err|") 1> >(sed -e "s|1|out|") ; } >/dev/null
{ echo 1 | tee /dev/stderr 2> >(sed -e "s|1|err|") 1> >(sed -e "s|1|out|") ; } 2>/dev/null
out
err
The ssh program normally handles the passing of /dev/stdout,/dev/stderr and /dev/stdin correctly:
$ ssh host "echo 1; echo 2 > /dev/stderr" > /dev/null
2
$ ssh host "echo 1; echo 2 > /dev/stderr" 2> /dev/null
1
Looking for an intercepting proxy made with netcat I found this script:
#!/bin/sh -e
if [ $# != 3 ]
then
echo "usage: $0 <src-port> <dst-host> <dst-port>"
exit 0
fi
TMP=`mktemp -d`
BACK=$TMP/pipe.back
SENT=$TMP/pipe.sent
RCVD=$TMP/pipe.rcvd
trap 'rm -rf "$TMP"' EXIT
mkfifo -m 0600 "$BACK" "$SENT" "$RCVD"
sed 's/^/ => /' <"$SENT" &
sed 's/^/<= /' <"$RCVD" &
nc -l -p "$1" <"$BACK" | tee "$SENT" | nc "$2" "$3" | tee "$RCVD" >"$BACK"
Which work nicely, as expected.
Since I need to look closely to the encoding used, hence the actual bytes passing, I tried to change some lines to use hexdump -vC:
#!/bin/sh -e
if [ $# != 3 ]
then
echo "usage: $0 <src-port> <dst-host> <dst-port>"
exit 0
fi
TMP=`mktemp -d`
BACK=$TMP/pipe.back
SENT=$TMP/pipe.sent
RCVD=$TMP/pipe.rcvd
trap 'rm -rf "$TMP"' EXIT
mkfifo -m 0600 "$BACK" "$SENT" "$RCVD"
( hexdump -vC | sed 's/^/ => /' ) <"$SENT" &
( hexdump -vC | sed 's/^/<= /' ) <"$RCVD" &
nc -l -p "$1" <"$BACK" | tee "$SENT" | nc "$2" "$3" | tee "$RCVD" >"$BACK"
Now it's not working anymore. Actually, I've lost the "realtime" feature of the previous script. Every byte sent is dumped in a single batch; then every byte received in another batch; and this all only after the connection is closed.
I'm suspecting some sort of buffering occurs in the pipe (|), but I'm not sure how to:
test this hypotesis;
fix the script to make it work in realtime again.
PS1. I'm using cygwin.
PS2. sh --version outputs:
GNU bash, version 4.1.10(4)-release (i686-pc-cygwin)
Edit:
Removind the | sed ... part (that is, leaving only hexdump -vC <"$SENT" and hexdump -vC <"$RCVD") the realtime feature is back, increasing my suspicion over the pipeline operator. But the output turns out to be confusing since sent and received bytes are mixed.
Still I couldn't manage to resolve the buffering (?) issue, but I could change the hexdump invocation to render the sed unnecessary:
#!/bin/sh -e
if [ $# != 3 ]
then
echo "usage: $0 <src-port> <dst-host> <dst-port>"
exit 0
fi
TMP=`mktemp -d`
BACK=$TMP/pipe.back
SENT=$TMP/pipe.sent
RCVD=$TMP/pipe.rcvd
trap 'rm -rf "$TMP"' EXIT
mkfifo -m 0600 "$BACK" "$SENT" "$RCVD"
hexdump -v -e '" => %08.8_Ax\n"' -e '" => %08.8_ax " 8/1 "%02x " " " 8/1 "%02x "' -e '" |" 16/1 "%_p" "|\n"' <"$SENT" &
hexdump -v -e '"<= %08.8_Ax\n"' -e '"<= %08.8_ax " 8/1 "%02x " " " 8/1 "%02x "' -e '" |" 16/1 "%_p" "|\n"' <"$RCVD" &
nc -l "$1" <"$BACK" | tee "$SENT" | nc "$2" "$3" | tee "$RCVD" >"$BACK"
Yes, the new hexdump looks ugly, but works.
This question for me is now open just for the sake of curiosity. I'm still willing to give the "correct answer" points to the one who explains (and fixes) the buffering (?) behavior.