I am running my shell script on multiple machines and all those machines can be in different datacenters.
If machine is in abc datacenter, then I don't want to sleep at all and move forward to next line in the shell script.
If machine is in def datacenter, then I want to sleep for 30 minutes and after that I will move to the next line in the shell script.
If machine is in pqr datacenter, then I want to sleep for 60 minutes and after that I will move to the next line in the shell script.
My machine name is like this and it will be always and as you can see, datacenter name is always before .host.com and it will be like this only.
machineA.abc.host.com
machineB.def.host.com
machineC.pqr.host.com
machinef-12341.testra.abc.host.com
.....
In my below shell script, I already have machine name stored in HOSTNAME variable so how can I extract the datacenter name from that in shell script and apply above conditions? I need to extract datacenter name which is just before .host.com so I need to do start from at the end?
#!/usr/bin/env bash
HOSTNAME=$hostname
.....
// I want to execute this line after the above if/else if logic
echo "Hello World"
What is the best way to do this? I can split the lines into variables but how to get relevant portion which I need and then apply if/elseif logic here?
Two different solutions in a testloop:
hosts="machineA.abc.host.com machineB.def.host.com machineC.pqr.host.com machinef-12341.testra.abc.host.com"
for testhost in ${hosts}; do
echo "sed ${testhost}: $(sed 's/.*\.\([^.]*\).host.com$/\1/' <<< "${testhost}")"
echo "cut ${testhost}: $(rev <<< "${testhost}"|cut -d"." -f3 | rev)"
done
Related
We work with Red Hat Linux (version 3.10.0-1160.59.1.el7.x86_6) and we are running into a problem with a process we have that reads data from a text file and then uses that data to pass parameters into a script file.
We have a control shell file that looks as follows.
year=$(date +"%Y%m%d")
xargs -L1 --arg-file=/u01/data.txt /u01/process.sh $year $1 $2 $3 $4
exit
As you can see, in the control file we simply get the current system date and then we read the data coming from a text file using xargs. The text file looks as follows.
John McEnroe Tennist US
Willie Nelson Singer US
"Jose Maria" Napoleon Singer Spain
Luciano Pavarotti Singer Italy
Finally, the process file that looks as follows (I have added the echos just to see the data coming into the process file).
. /u01/app/env/bt.env
echo "TK Total parms: $#"
echo "Parm1 $1"
echo "Parm2 $2"
echo "Parm3 $3"
echo "Parm4 $4"
echo "Parm5 $5"
/u01/app/task /task:localhost:CREATION /user:$LINUX_USER:$LINUX_PASSWORD /params:DATE="$1",NAME="$2",LAST_NAME="$3",JOB="$4",COUNTRY="$5" /debug
sleep 2
exit
The problem we have is with the row where the Name is composed by two words ("Jose" and "Maria"). I thought that if we added quotes around the entire value, that would work well, but it does not: the process file is getting the two words and treating them as if "Jose" was parameter #3, and "Maria" as if it was parameter #4 (and this causes the process to crash).
I have tried to modify the data file to use apostrophes instead of quotes (i.e., 'Jose Maria'); I also tried to wrap the quote around apostrophes and viceversa (i.e. "'Jose Maria'", '"Jose Maria"'), but none of these methods fixes the problem.
Could anyone suggest me how to write the data file to make it work as we need?
Thanks in advance!
I have a task to monitor the system with a quota, if the monitored result is over the quota, send a warning email. But this monitor program should be called once in half an hour, after one warning email is sent out, the next time if the monitored state is still the same as last time, there is no need to send the same warning email again.
In order to do this, I would like to make use of environment variable to store the state of the last monitored result, so that the next time it can be checked and duplicate email would not be sent. One of my solution is to add or update the export syntax in .bashrc, but in order to activate the updated export syntax, I have to run bash, which might be unnecessary.
So I would like ask is there any way to update the environment variable so that every time when the monitor program Bash script is called, it gets the fresh updated value?
This is a self contained solution using a heredoc. At first glance it may seem an elaborate and inperfect solution, it does have its uses in that it's resilient and it works well when deploying across more than one machine, requires no special monitoring or permissions of external files, and most importantly, there are no unwanted surprises with environment.
This example uses bash, but it will work with sh if the $thisfile variable is set manually, or another way.
This example assumes that 20 is already in the script file as mymonitorvalue, and uses argument $1 as a proof of concept. You would obviously change newval="$1" to whatever calculates the quota:
Example usage:
#bash $>./script 10
Value from previous run was 20
Value from this test was 10
Updating value ...
#bash $>./script 10
not doing anything ... result is the same as last time
#bash $>
Script:
#!/bin/bash
thisfile="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )" ; thisfile="${DIR}${0}"
read -d '' currentvalue <<'EOF'
mymonitorval=20
EOF
eval "$currentvalue"
function changeval () {
sed -E -i "s/(^mymonitorval\=)(.*)/mymonitorval\="$1"/g" "$thisfile"
}
newvalue="$1"
if [[ "$newvalue" != "$mymonitorval" ]]; then
echo "Value from previous run was $mymonitorval"
echo "Value from this test was "$1""
echo "Updating value ..."
changeval "$newvalue"
else
echo "not doing anything ... result is the same as last time"
fi
Explanation:
thisfile= can be set manually for script location. This example uses the automated solution from here: https://stackoverflow.com/a/246128
read -d...EOF is the heredoc which is saved into variable $currentvalue
eval "$currentvalue" in this case is the equivalent of typing mymonitorval=10 into a terminal
function changeval...} updates the contents of the heredoc in place (it changes the physical .sh script file)
newvalue="$1" is for the purpose of testing. $newvalue would be determined by whatever your script is that is calculating quota
if.... block is to perform two alternate sets of actions depending on whether $newvalue is the same as it was last time or not.
Store environment variable in different .file and then source <.file>
for my first bash project I am developing a simple bash script that shows basic information about my system:
#!/bash/sh
UPTIME=$(w)
MHZ=$(cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq)
TEMP=$(cat /sys/class/thermal/thermal_zone0/temp)
#UPTIME shows the uptime of the device
#MHZ shows the overclocked specs
#TEMP shows the current CPU Temperature
echo "$UPTIME" #displays uptime
echo "$MHZ" #displays overclocked specs
echo "$TEMP" #displays CPU Temperature
MY QUESTION: How can I code this so that the uptime and CPU temperature refresh every 2seconds without re-generating the code new every time (I just want these two variables to update without having to enter the file path again and re-running the whole script).
This code is already working fine on my system but after it executes in the command line, the information isn't updating because it executed the command and is standing by for the next command instead of updating the variables such as UPTIME in real time.
I hope someone understands what I am trying to achieve, sorry about my bad wordings of this idea.
Thank you in advance...
I think it will help you. You can use the watch command for updating that for every two seconds without the loop.
watch ./filename.sh
It will give you the update of that command for every two second.
watch - execute a program periodically, showing output fullscreen
Not sure to really understand the main goal, but here's an answer to the basic question "How can I code this so that the uptime and CPU temperature refresh every two seconds ?" :
#!/bash/sh
while :; do
UPTIME=$(w)
MHZ=$(cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq)
TEMP=$(cat /sys/class/thermal/thermal_zone0/temp)
#UPTIME shows the uptime of the device
#MHZ shows the overclocked specs
#TEMP shows the current CPU Temperature
echo "$UPTIME" #displays uptime
echo "$MHZ" #displays overclocked specs
echo "$TEMP" #displays CPU Temperature
sleep 2
done
I may suggest some modifications.
For such simple job I may recommend no to use external utilities. So instead of $(cat file) you could use $(<file). This is a cheaper method as bash does not have to launch cat.
On the other hand if reading those devices returns only one line, you can use the bash built-in read like: read ENV_VAR <single_line_file. It is even cheaper. If there are more lines and for example you want to read the 2nd line, you could use sg like this: { read line_1; read line2;} <file.
As I see w provides much more information and as I assume you need only the header line. This is exactly what uptime prints. The external utility uptime reads the /proc/uptime pseudo file. So to avoid to call externals, you can read this pseudo file directly.
The looping part also uses the external sleep(1) utility. For this the timeout feature of the read internal could be used.
So in short the script would look like this:
while :; do
# /proc/uptime has two fields, uptime and idle time
read UPTIME IDLE </proc/uptime
# Not having these pseudo files on my system, the whole line is read
# Maybe some formatting is needed. For MHZ /proc/cpuinfo may be used
read MHZ </sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq
read TEMP </sys/class/thermal/thermal_zone0/temp
# Bash supports only integer arithmetic, so chomp off float
UPTIME_SEC=${UPTIME%.*}
UPTIME_HOURS=$((UPTIME_SEC/3600))
echo "Uptime: $UPTIME_HOURS hours"
echo $MHZ
echo $TEMP
# It reads stdin, so pressing an ENTER it returns immediately
read -t 2
done
This does not call any external utility and does not make any fork. So instead of executing 3 external utilities (using the expensive fork and execve system calls) in every 2 seconds this executes none. Much less system resources are used.
you could use while [ : ] and sleep 2
You need the awesome power of loops! Something like this should be a good starting point:
while true ; do
echo 'Uptime:'
w 2>&1 | sed 's/^/ /'
echo 'Clocking:'
sed 's/^/ /' /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq
echo 'Temperature:'
sed 's/^/ /' /sys/class/thermal/thermal_zone0/temp
echo '=========='
sleep 2
done
That should give you your three sections, with the data of each nicely indented.
I'm currently working on a script (using BASH) which backups VM file to a remote server.
I want to try and make the script a bit more dynamic by being able to just looping though each VM from a "show VM command". my idea is to take the standard output of a command which show all the VM and break up and turn it to useful variables. possibly a multi-array.
the Output comes out like this is there anyway to break it all up? say by spaces and line breaks?
Vmid Name File Guest OS Version Annotation
10 FREEPBX [datastore2] FREEPBX/FREEPBX.vmx other26xLinux64Guest vmx-08
13 AdaptivNICE2Cloud [datastore2] AdaptivNICE2Cloud/AdaptivNICE2Cloud.vmx other26xLinux64Guest vmx-08
15 IVSTelManager [datastore2] IVSTelManager/IVSTelManager.vmx debian6Guest vmx-08
4 Neptune [datastore1] Neptune/Neptune.vmx winNetEnterprise64Guest vmx-08
9 Kayako [datastore2] Kayako/Kayako.vmx other26xLinux64Guest vmx-08
I guess you need this:
$ vim-cmd vmsvc/getallvms | sed -n 's|.*\[|/vmfs/volumes/|;s|\] *|/|;s|\.vmx .*|.vmx|p'
/vmfs/volumes/datastore2/FREEPBX/FREEPBX.vmx
/vmfs/volumes/datastore2/AdaptivNICE2Cloud/AdaptivNICE2Cloud.vmx
/vmfs/volumes/datastore2/IVSTelManager/IVSTelManager.vmx
/vmfs/volumes/datastore1/Neptune/Neptune.vmx
/vmfs/volumes/datastore2/Kayako/Kayako.vmx
# Prints all VMX files paths
OR
$ vim-cmd vmsvc/getallvms | sed -n 's|.*\[|/vmfs/volumes/|;s|\] *|/|;s|/[^/]*\.vmx .*||p'
/vmfs/volumes/datastore2/FREEPBX
/vmfs/volumes/datastore2/AdaptivNICE2Cloud
/vmfs/volumes/datastore2/IVSTelManager
/vmfs/volumes/datastore1/Neptune
/vmfs/volumes/datastore2/Kayako
# Prints all directories having VMX files. These directories also contain the virtual HDDs, which you would want to backup.
(Ignore the $ in the prompt; it is still root prompt. SO would interpret it as comment if I use # in place if $..)
I wanted to run a simple bash timer and found this online (user brent7890)
#!/usr/bin/bash
timer=60
until [ "$timer" = 0 ]
do
clear
echo "$timer"
timer=`expr $timer - 1`
sleep 1
done
echo "-------Time to go home--------"
I couldn't copy and paste this code because the server is on another network. I typed it like this (below) and got an error on the line that starts with "until".
#!/usr/bin/bash
timer=60
#Note I forgot the space between [ and "
until ["$timer" = 0 ]
do
clear
echo "$timer"
timer=`expr $timer - 1`
sleep 1
done
echo "-------Time to go home--------"
Where is spacing like this documented? It seems strange that it matters. Bash scripts can be confusing, I want to understand why the space is important.
There are several rules, two basic of that are these:
You must separate all arguments of a command with spaces.
You must separate a command and the argument, that follows after, with a space.
[ here is a command (test).
If you write ["$timer" that means that you start command [60,
and that is, of course, incorrect. The name of the command is [.
The name of the command is always separated from the rest of the command line with a space. (you can have a command with a space in it, but in this case you must write the name of the command in "" or '').