need to install the package first:
yum -y install ImageMagick-devel ImageMagick-c++-devel
then we can install it per PHP version separately
/opt/cpanel/ea-php73/root/usr/bin/pecl install imagick

To find the missed packages we can use the following command:

awk '$4=="E"' /var/cpanel/updatelogs/last

It will allow us to see the missed RPMs and install them after via yum.

The original article is stored here

To fix the issue, you need to execute the following command:

grub2-mkconfig -o /boot/efi/EFI/centos/grub.cfg;grub2-mkconfig -o /boot/efi/EFI/centos/grub.cfg;reboot

in the file /etc/selinux/config
need replace the value for option SELINUX from:

SELINUX=enforcing

to:

SELINUX=disabled

And then reboot the server.

mkdir -pv /reports;cxs --qcreate --quarantine /home/quarantine;cxs -U;cxs --logfile /var/log/cxs.log --exploitscan --virusscan --sversionscan --bayes -I /etc/cxs/cxs.ignore -Q /home/quarantine --options mMOLfSGchexdnwZRD --voptions mfuhexT --qoptions Mv -Z --www --summary --html --ssl -C /var/clamd --nofallback -T 5 --allusers --quiet --report /reports/full_malware_scan_`date +%d.%m.%Y`.txt

we can do it in 2 simple steps:

1. need to install additional cPanel repository:

yum install -y ea4-experimental

2. now we can install memcache and memcached for all php versions:
for i in `ls -1 /opt/cpanel/|grep php`;do yum install -y "$i"-php-memcached "$i"-php-memcache;done

find the issue:
named-checkzone somedomain.com /var/named/somedomain.com.db
zone somedomain.com/IN: NS 'ns1.somedomain.com' has no address records (A or AAAA)
zone somedomain.com/IN: NS 'ns2.somedomain.com' has no address records (A or AAAA)
zone somedomain.com/IN: not loaded due to errors.

when it fixed:
named-checkzone somedomain.com /var/named/somedomain.com.db
zone somedomain.com/IN: loaded serial 2020021308
OK

I have faced the issue on one of my servers where I am using software RAID5.
One of the drives has died, so we replace it with a new one.

However, after that, we need to put it to a raid and recover the normal work, so:

1. We need to create exactly the same structure on the new drive that we are using for the rest of them:

sfdisk -d /dev/sda | sfdisk /dev/sdc
Where SDA is an old one inside the RAID and SDC is a new one (we are copy structure to it)

2. Also, the second issue was discovered – one of the arrays (MD0) is going to an inactive state. To fix it need to do the following:

2.1 stop it
mdadm --stop /dev/md0

2.2 assemble them once again
mdadm --assemble --force /dev/md0 /dev/sd[abd]1

3. readd the new HDD to a RAID array:
mdadm --manage /dev/md0 --add /dev/sdc1
mdadm --manage /dev/md1 --add /dev/sdc2
mdadm --manage /dev/md2 --add /dev/sdc3

Note: Better to start from the small arrays because it can be synced only in a step by step mode. So we can sync small arrays quickly.

To fix it:
cd /var/spool/exim/db;rm -fv retry retry.lockfile retry.lockfile wait-remote_smtp.lockfile;service exim restart; cd ~

A lot of the issue and issues with the spam can happen if the user can send the emails without authorization. In order to prevent the issue, we can block them:

In the end of acl_check_not_smtp, before accept section need to add the following code:
#############################
accept condition = ${if eq{$originator_uid}{0}}
drop message = Local users can't send mail
#############################

It will prevent all emails except the emails from the root.

The second way, you can limit it, via the following lines:
#############################
deny message = Sender rate overlimit - $sender_rate / $sender_rate_period
ratelimit = 10 / 1h / strict
#############################

It means that only 10 emails can be sent via users or web scripts without authorization in 1 hour.

But only 1 way can be used, it is not possible to mix it.