January 12

DA exim whitelist

The location of the files that can be used to whitelist something in EXIM that need to be done:
/etc/virtual/whitelist_domains
/etc/virtual/whitelist_from
/etc/virtual/whitelist_hosts
/etc/virtual/whitelist_hosts_ip
/etc/virtual/whitelist_senders

Post Views: 2
Category: simple memo | LEAVE A COMMENT
December 9

log all bash shell activity to separate file

log location:
echo 'local6.* /var/log/commands.log' >> /etc/rsyslog.d/bash.conf
add the following rule in the end of /etc/bashrc file:
export PROMPT_COMMAND='RETRN_VAL=$?; logger -p local6.debug "$(whoami) [$$]: $(history 1 | sed "s/^[ ]*[0-9]\\+[ ]*//" ) [$RETRN_VAL]"'

restart the service:

service rsyslog restart

done 🙂

Post Views: 32
Category: simple memo | LEAVE A COMMENT
November 9

Cron memo

# ┌────────── minute (0 - 59)
# │ ┌──────── hour (0 - 23)
# │ │ ┌────── day of month (1 - 31)
# │ │ │ ┌──── month (1 - 12)
# │ │ │ │ ┌── day of week (0 - 6 => Sunday - Saturday, or
# │ │ │ │ │ 1 - 7 => Monday - Sunday)
# ↓ ↓ ↓ ↓ ↓
# * * * * * command to be executed
#
# 0 - Sun Sunday
# 1 - Mon Monday
# 2 - Tue Tuesday
# 3 - Wed Wednesday
# 4 - Thu Thursday
# 5 - Fri Friday
# 6 - Sat Saturday
# 7 - Sun Sunday
#

Post Views: 9
Category: simple memo | LEAVE A COMMENT
October 12

extremely stupid backup script (files+dbs)

#!/bin/bash

if [[ "$1" != "db_backup" && "$1" != "files_backup" && "$1" != "all" ]]; then
echo -e "Please run the script with following keys:\ndb_backup -> to do backup of all DBs\nfiles_backup -> to do backup of all users without DBs\nall -> to do users and their DBs backups at once"
exit 1
fi
datetime=$(date +%d-%m-%Y_%H-%M-%S)
db_backup_function(){
mkdir -pv /backup/dbs/"$datetime"
for i in `mysql -e "show databases;"|grep -wv "Database\|performance_schema\|information_schema\|sys\|mysql"`;do
mysqldump "$i" >> /backup/dbs/"$datetime"/"$i".sql
if [[ $? -ne 0 ]];then
echo "$datetime DB $i FAILED" >> /backup/failed_DBs.log
fi

zip --quiet /backup/dbs/"$datetime"/"$i".sql.zip /backup/dbs/"$datetime"/"$i".sql
if [[ $? -eq 0 ]];then
rm -f /backup/dbs/"$datetime"/"$i".sql
fi
done
}

files_backup_function(){
mkdir -pv /backup/files/"$datetime"
cd /home/
for i in `ls -1 /home/`;do
zip -r --quiet --symlinks /backup/files/"$datetime"/"$i".zip "$i"
if [[ $? -ne 0 ]];then
echo "$datetime files for $i account FAILED" >> /backup/failed_users.log
fi
done
}

if [[ "$1" = "db_backup" ]];then
db_backup_function
fi

if [[ "$1" = "files_backup" ]];then
files_backup_function
fi

if [[ "$1" = "all" ]];then
db_backup_function
files_backup_function
fi

Post Views: 12
Category: simple memo | LEAVE A COMMENT
October 12

Small reminder about commands

A few thread file transfer:

rclone copy :sftp:/disk_folder/disk_hash_id /the_way/to_target_folder/ --sftp-host=target_host_name_or_ip --sftp-user=your_username --sftp-ask-password --sftp-key-use-agent=false --sftp-shell-type=unix --sftp-md5sum-command=md5sum --progress --transfers=1 --multi-thread-streams=6 --checksum --bwlimit=300M

The single thread file transfer with additional validation:
rsync --progress --checksum --checksum-choice=md5 /disk_folder/disk_hash_id your_username@target_host_name_or_ip:/the_way/to_target_folder/

Aggregated itop stat (very helpful for a debug in some cases):

iotop -aoP

Manual checksums for the file:

md5sum /path/to/file

Look deeper at TCP timeouts and waits:

netstat -napo|grep TIME_WAIT

Check disk info:
qemu-img info disk_name

Create disk from command line:
qemu-img create -f qcow2 new_disk_name 15G

Converting/cloning the disk (It should also fix the dependency on the backing file, which means building a standalone VM)
qemu-img convert -O qcow2 old_with_backing_dependency.qcow2 new_standalone.qcow2

Add the backuing file to already existing disk (or rewrite this file):
qemu-img rebase -u -b /way_to_the_backing/file -F qcow2 /way_to_disk/new.qcow2

Creating disk WITH backing file:
qemu-img create -f qcow2 -F qcow2 -b /way_to_the_backing/file new.qcow2 500G

Resize disk:
qemu-img resize disk.qcow2 +10G

Attach the disk to VM:
virsh attach-disk --domain short_VM_ID /way_to_disk/file --target vdx --driver "qemu" --subdriver "qcow2"

Detach the disk from VM:
virsh detach-disk --domain short_VM_ID --target vdx

Check VM details:
virsh dumpxml short_VM_ID

Show all VMs:
virsh list --all

Post Views: 14
Category: simple memo | LEAVE A COMMENT
September 11

CXS mount via FUSE and scan

Short reminder

Mount:

sshfs root@server2scan:/var/www/customer_id/ /mnt/customer_id -o IdentityFile=/root/.ssh/scan.key -o allow_other

Scan:

cxs --force --deep --timemax 1000 --filemax 100000 --report /root/scan_results/customer_id.scan_report /mnt/customer_id/

 

Post Views: 17
Category: simple memo | LEAVE A COMMENT
August 14

Resize disk in Linux OS:

1. Use lsblk to locate partitions and their IDs:

NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
vda 8:0 0 40G 0 disk
├─vda1 8:1 0 1M 0 part
└─vda2 8:2 0 40G 0 part /

2. LVM case:

growpart /dev/vda 2
pvresize /dev/vda2
lvextend -l +100%FREE /dev/mapper/ubuntu--vg-ubuntu--lv
xfs_growfs /dev/mapper/ubuntu--vg-ubuntu--lv

OR, if we use ext4:

resize2fs /dev/mapper/ubuntu--vg-ubuntu--lv

3. non-LVM case:

growpart /dev/vda 2
xfs_growfs /

OR, if we use ext4:

resize2fs /

ps. growpart is provided by the following package in RHEL:

cloud-utils-growpart
Post Views: 21
Category: simple memo | LEAVE A COMMENT
August 11

Nagios script to check host uptime 

#!/bin/bash
#Simple script to check the server uptime in seconds
#written by Vasyl T.
minimum='86400' # 1 day (24 hours) in seconds
real=$(/usr/bin/awk -F'.' '{print $1}' /proc/uptime)

if [[ "$minimum" -gt "$real" ]];then
echo "CRITICAL. Server uptime is $real seconds, when expected - more than: $minimum"
exit 2;
else
days=$(($real / $minimum))
echo "OK. Server uptime is $real seconds or $days day(s)"
exit 0;
fi
Post Views: 19
Category: simple memo | LEAVE A COMMENT