Learning About Bash Scripts
Getting Started 📖
First of all you need to have knowledge and some concepts in Linux, at least the basic level.
Why should I use it?
If you have ever been a SysAdmin, you will understand me. Sometimes it’s necessary to create a workaround for you to have a weekend in peace. Below I’ll write some reasons for you to create a Bash Script.
If you:
- want to do something for several times;
- need to monitor a simple action that is running on server;
- need to schedule an action to run from time to time;
- maybe a workaround.
Right! So let’s do it.
Curious about the first line ❓❓❓
The first line is called Shebang. In summary, we are going to tell the operating system which interpreter to use when executing the file.
We’ll use the "/usr/bin/env bash" because we don’t worry what operating system we are running the script on.
Hands on 🙌
In this post, we’ll see about the RSync script and then, follow me step-by-step:
(1) Creating the script and putting it into execution mode:
touch rsync_script.sh
chmod +x $_
(2) Adding the first line and variables like arguments:
#!/usr/bin/env bash
SOURCE=$1
DESTINATION=$2
LOG_FILE=/tmp/rsync.log
(3) What I think important about this script is that you need to create lockfile to prevent overhead. In that case, we need to create two functions to do this:
function log {
echo [$(date "+%d/%m/%Y %H:%M:%S")] - $@
}
function create_lock {
basename=$(basename $0)
file_lock="/tmp/$basename.lock"
pid=$$
log Starting Script
if [ -f $file_lock ]; then
if ps -p $(cat $file_lock) > /dev/null 2>&1; then
log Lock file found. Exiting...
exit 0
else
log Lock file found but no process running. Removing lock file: $file_lock
rm -f $file_lock
log Creating lockfile: $file_lock
echo $pid > $file_lock
fi
else
log Creating lock file: $file_lock
echo $pid > $file_lock
fi
}
function remove_lock {
log Removing lock file: $file_lock
rm -f $file_lock 2> /dev/null
}
(4) Now, we need to create a function for RSYNC:
function rsyncExec {
rsync --log-file="${LOG_FILE}" \
-av \
--numeric-ids \
--progress \
$SOURCE \
$DESTINATION
}
However, what will happen if the source and destination aren’t valid? Therefore, we need create the testing phase. In this case, we can create a test for source and another for destination in just one line for each. For example:
[ ! -d $SOURCE ] && echo "Not found directory: $SOURCE" && exit 1
[ ! -d $DESTINATION ] && echo "Not found directory: $DESTINATION" && exit 1
(5) Finally, we need to assemble the script and put each part together:
$ vim rsync_script.sh
#!/usr/bin/env bash
# STEP 1: ----------------------------------------------------------------
SOURCE=$1
DESTINATION=$2
LOG_FILE=/tmp/rsync.log
# STEP 2: TESTS ----------------------------------------------------------
[ ! -d $SOURCE ] && echo "Not found directory: $SOURCE" && exit 1
[ ! -d $DESTINATION ] && echo "Not found directory: $DESTINATION" && exit 1
# STEP 3: ----------------------------------------------------------------
function log {
echo [$(date "+%d/%m/%Y %H:%M:%S")] - $@
}
function create_lock {
basename=$(basename $0)
file_lock="/tmp/$basename.lock"
pid=$$
log Starting Script
if [ -f $file_lock ]; then
if ps -p $(cat $file_lock) > /dev/null 2>&1; then
log Lock file found. Exiting...
exit 0
else
log Lock file found but no process running. Removing lock file: $file_lock
rm -f $file_lock
log Creating lockfile: $file_lock
echo $pid > $file_lock
fi
else
log Creating lock file: $file_lock
echo $pid > $file_lock
fi
}
function remove_lock {
log Removing lock file: $file_lock
rm -f $file_lock 2> /dev/null
}
# STEP 4: ----------------------------------------------------------------
function rsyncExec {
rsync --log-file="${LOG_FILE}" \
-av \
--numeric-ids \
--progress \
$SOURCE \
$DESTINATION
}
# STEP 5: ----------------------------------------------------------------
create_lock;
rsyncExec $SOURCE $DESTINATION;
remove_lock;
And now? Let’s test!
We need to create a source directory to do our testing:
mkdir -p class/01/{1,2,3,4}
mkdir -p class/02
touch class/01/{1,2,3,4}/main.yaml
Look this tree directory:
tree class
class
├── 01
│ ├── 1
│ │ └── main.yaml
│ ├── 2
│ │ └── main.yaml
│ ├── 3
│ │ └── main.yaml
│ └── 4
│ └── main.yaml
└── 02
Now, let’s run our script! 👀
$ ./rscript.sh class/01 class/02
[19/01/2021 17:39:45] - Starting Script
[19/01/2021 17:39:45] - Creating lock file: /tmp/rscript.sh.lock
building file list ...
11 files to consider
01/
01/1/
01/1/main.yaml
0 100% 0.00kB/s 0:00:00 (xfer#1, to-check=8/11)
01/2/
01/2/main.yaml
0 100% 0.00kB/s 0:00:00 (xfer#2, to-check=6/11)
01/2/1/
01/2/1/main.yaml
0 100% 0.00kB/s 0:00:00 (xfer#3, to-check=4/11)
01/3/
01/3/main.yaml
0 100% 0.00kB/s 0:00:00 (xfer#4, to-check=2/11)
01/4/
01/4/main.yaml
0 100% 0.00kB/s 0:00:00 (xfer#5, to-check=0/11)
sent 446 bytes received 166 bytes 1224.00 bytes/sec
total size is 0 speedup is 0.00
[19/01/2021 17:39:45] - Removing lock file: /tmp/rscript.sh.lock
Current directory:
tree class
class
├── 01
│ ├── 1
│ │ └── main.yaml
│ ├── 2
│ │ ├── 1
│ │ │ └── main.yaml
│ │ └── main.yaml
│ ├── 3
│ │ └── main.yaml
│ └── 4
│ └── main.yaml
└── 02
└── 01
├── 1
│ └── main.yaml
├── 2
│ └── main.yaml
├── 3
│ └── main.yaml
└── 4
└── main.yaml
12 directories, 9 files
You can see more examples and ideas in my repository:
https://github.com/wsalles/scripts-shell