The shell script run by crond keeps stopping after five minutes

Question

As titled, I am using CentOS 6.10

Here is part of my shell script, it runs find manually.

#!/bin/sh
backup_dir="/mnt/backup/website"
all_web="$(</mnt/backup/list.txt)"

for web in $all_web
do
    IFS=',' read -ra arraydata <<< "$web"
    tar zcvf $backup_dir/${arraydata[0]}\_$(date +%Y%m%d).tar.gz ${arraydata[1]}
done
exit 0;

The settings in /etc/crontab

0 22 * * * root sh /mnt/backup/backup.sh

I have tried to print the script's logs, but nothing was found, the log just shows that it was stopped.

/mnt/website/corey/public/files/111.pdf
/mnt/website/corey/public/files/222.pdf
/mnt/website/corey/public/files/333.pdf
/mnt/website/corey/pub

Looked through the limits, don't know which one would cause the issue.

core file size          (blocks, -c) 0
data seg size           (kbytes, -d) unlimited
scheduling priority             (-e) 0
file size               (blocks, -f) unlimited
pending signals                 (-i) 65536
max locked memory       (kbytes, -l) 64
max memory size         (kbytes, -m) unlimited
open files                      (-n) 65535
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) 819200
real-time priority              (-r) 0
stack size              (kbytes, -s) 10240
cpu time               (seconds, -t) unlimited
max user processes              (-u) 4096
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

Inspected the system using these commands, all of them showed nothing.

dmesg | grep -i "killed process"
dmesg | grep -i "out of memory"
dmesg | grep -i "corn"
dmesg | grep -i "backup"
grep -i "sigkill" /var/log/messages
grep -i "limit" /var/log/messages

Didn't find any error message in /var/log/cron either.

2025/03/31 Updated
I tried to execute the tar command line natively in cornd, still got stopped after five minutes, below is part of my /etc/crontab file

50 3 * * * root /bin/tar zcvf /mnt/backup/website/corey_20250331.tar.gz /mnt/webdisk/corey/

ls -lh

-rw-r--r-- 1 root root 6.1G Mar 31 03:55 corey_20250331.tar.gz

Workaround?
I moved the script from /etc/crontab to folder /etc/cron.daily, the script doesn't get killed. I have no idea what is the difference.

2025/04/07 Updated
As mentioned above, the workaround that moves the script to /etc/cron.daily worked well, but when I tried to specify the execution time in /etc/crontab like this

30 21 * * * root run-parts /etc/cron.daily

The script GET KILLED AFTER FIVE MINUTES AGAIN!

Any idea would really appreciated.

Answer 1

While you did check for OOM killer, may I suggest you are using a lot of unnecessary resources. I'd write this loop as

#!/bin/bash

backup_dir=/mnt/backup/website
all_web=/mnt/backup/list.txt

printf -v today '%(%Y%m%d)T' -2
ret=0
while IFS=, read -ru3 -a arraydata
do
    tar -zcvf "$backup_dir/${arraydata[0]}_$today.tar.gz" -- "${arraydata[1]}" ||
      ret=$?
done 3<"$all_web" || ret=$?
exit "$ret"

This reads the files line by line and processes one at a time. Previously you were loading the entire file, which I assume could be large, into memory at once, which could use a lot of resources and isn't necessary (unless $all_web can change during the backup). I also quoted things to handle any space or tab or wildcard characters it encounters, added the missing -- to handle file names starting with -, retrieved today's date only once and with the printf builtin (getting the date at the time of shell invocation with -2), and report any failure of any of the tar commands in the script's exit status.