DOCSEARCHbyAlgolia
Submit Docs

ApisCP Notes and Cheats

website·Indexed 33 days ago

https://notes.apiscp.com/

A collection of administration notes, cheatsheets, and guides for ApisCP. Covers topics including PHP memory and PECL module configuration, MariaDB and PostgreSQL upgrades, Podman and Docker container routing, Apache .htaccess rules, and general system administration tasks using cpcmd and Ansible.

Tokens
7.4K
Snippets
54
Records
60
Agent score
59%

What's inside ApisCP

  1. Analyze Xdebug profiling data

    Xdebug generates profiling data compatible with the Cachegrind format. This data can be analyzed using tools such as KCacheGrind (Linux/KDE), QCacheGrind (Windows), or WebGrind.

  2. Configure WordPress Redis Object Cache plugin for Unix sockets

    After creating a Redis instance, install and configure the redis-cache plugin via WP-CLI. To use a Unix socket instead of a TCP connection, set the WP_REDIS_SCHEME to unix and provide the absolute path to the socket file in WP_REDIS_PATH.
    # Switch to the site account and directory
su benchmark.test
cd /var/www/html

# Install and activate the plugin
wp-cli plugin install --activate redis-cache

# Configure the plugin to use the Unix socket
wp-cli config set WP_REDIS_SCHEME unix
wp-cli config set WP_REDIS_PATH /var/www/redis.sock

# Enable and verify the Redis cache
wp-cli redis enable
wp-cli redis status

  3. Analyze site storage and inode usage

    You can analyze storage usage using either the system repquota tool or the ApisCP admin:get-usage('storage') command.

    When using admin:get-usage('storage'), the following keys are used in the output:

    • qused: Storage blocks used, measured in 4 KB units.
    • fused: Inode usage.
    # Option 1: Using repquota for virtual home directory
repquota -g $(findmnt -no SOURCE --target /home/virtual) | sort -n --key=3

# Option 2: Using cpcmd to sort by storage used (qused)
cpcmd -o json admin:get-usage storage | jq 'to_entries | sort_by(.value.qused)'

  4. Monitor resource usage by control group with systemd-cgtop

    Use systemd-cgtop to view a top-like interface for all control groups. This is particularly useful for monitoring processes spawned within the context of a site, as they are governed by resource limits assigned in the cgroup service class. Use the following hotkeys to sort the view:

    • m: Sort by Memory
    • c: Sort by CPU
    • t: Sort by Tasks (PID count)

  5. Update user preferences without jq using PHP serialization

    If jq is not installed and cannot be added to the system, you can manipulate preferences by switching the input/output format from json to serialize. This allows you to use a PHP one-liner to unserialize the preferences, modify the array, and re-serialize the data before saving.
    # Example: set nexus.notify-create value to 1 using PHP serialization
cpcmd -i serialize common:save-preferences "$(cpcmd -o serialize common:load-preferences | php -r '$prefs = unserialize(file_get_contents("php://stdin")); $prefs["nexus"]["notify-create"] = 1; echo serialize($prefs);')"

  6. Install Podman on ApisCP

    Podman is the supported drop-in replacement for RHEL 8+ on ApisCP. Note that Podman may interfere with typical ApisCP behavior by changing firewall rules to accommodate container routing; these rules may be refreshed by the panel, potentially making containers unresponsive when accessing exposed ports from the host or external network.
    dnf install -y podman
/usr/local/apnscp/bin/scripts/yum-post.php install -d podman siteinfo
systemctl reload fsmount

  7. Update user preferences using cpcmd and jq

    To modify user preferences, use cpcmd to load preferences in JSON format, manipulate the JSON using the jq utility, and save the modified JSON back. You can masquerade as a specific site using -d domain and/or a specific user using -u user to target the correct preferences.
    # Install jq if not present
yum install -y jq

# Example: set nexus.notify-create value to 1
cpcmd -i json common:save-preferences "$(cpcmd -o json common:load-preferences | jq -r '.nexus["notify-create"]=1')"

  8. Install Yarn on apiSCP using the apiscp-yarn addin

    Yarn can be installed on apiSCP by cloning the apiscp-yarn playbook addin into the resources directory and executing the addin.yml Ansible playbook with the appropriate extra variable.
    cd /usr/local/apnscp/resources/playbooks
git clone https://github.com/thundersquared/apiscp-yarn.git addins/apiscp-yarn
ansible-playbook addin.yml --extra-vars=addin=apiscp-yarn

  9. Create a shell shortcut for cross-node command execution

    You can create a shell function (e.g., in ~/.bashrc, ~/.zshrc, or a separate ~/.functions file) to wrap the Ansible shell module. This allows you to run a command across all nodes in your inventory by simply calling the function name.

    Note: The command passed to the function must be a string; any quotes within the command must be escaped to ensure correct execution.

    function apiscps() {
  cd /path/to/panel-management # Replace with your absolute project path
  echo "ansible all -m shell -a $@"
  ansible all -m shell -a "$@"
}

# Usage examples:
# Get all panels' version
apiscps 'cpcmd misc:cp_version'

# Query borg status
apiscps 'borgmatic info'

# Escape quoted values
apiscps "cpcmd scope:set cp.nightly-updates 'Mon..Fri 20:00'"

# Shell interpolation
apiscps 'dnf remove $(rpm -qa kernel-ml)'

  10. Redirect all hostnames to the ApisCP panel hostname

    To ensure all hostnames redirect to the primary panel hostname (defined by cpcmd scope:get net.hostname), configure a mod_rewrite rule in the ApisCP custom HTTPD configuration. This prevents access via incorrect hostnames and enforces the use of the primary panel address.
    # Add to /usr/local/apnscp/config/httpd-custom.conf
<Directory ${APNSCP_ROOT}/public>
RewriteEngine on
RewriteOptions inherit
RewriteCond %{HTTP_HOST} ^([^:]+)
RewriteCond "%1 %{ENV:HOSTNAME}" "!(^[^ ]*) \1"
RewriteRule ^(.*)$ %{REQUEST_SCHEME}://%{ENV:HOSTNAME}:%{SERVER_PORT}/$1 [R,L]
</Directory>

    Restart ApisCP to apply changes

    systemctl restart apiscp

  11. Update plugin paths after cloning WordPress

    After cloning a WordPress installation to a new location, manually update the absolute paths to the new WordPress root directory for any plugins that store the root path as a constant in their configuration.

  12. Switch from Podman to Docker CE on ApisCP

    If Podman's firewall behavior causes issues, you can switch to Docker CE. This process involves removing Podman, installing Docker CE, and applying a systemd unit override to fix the ExecStart definition. Note that systemd-backed container definitions must be recreated using docker run or docker-compose after the switch.
    # Remove podman or other docker-related packages
dnf remove docker docker-client docker-client-latest docker-common docker-latest docker-latest-logrotate docker-logrotate docker-engine podman runc

# Add Docker CE repo
dnf config-manager --add-repo https://download.docker.com/linux/rhel/docker-ce.repo

# Install docker-ce package
dnf install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin

# Stop unit for reconfiguration
systemctl stop docker

# Fix ExecStart definition with a unit override
mkdir -p /etc/systemd/system/docker.service.d
cat << EOF > /etc/systemd/system/docker.service.d/override.conf
[Service]
# This clears any ExecStart= inherited from docker.service
ExecStart=
ExecStart=/usr/bin/dockerd --containerd=/run/containerd/containerd.sock
EOF

# Reload systemd daemon
systemctl daemon-reload

# Docker group should be created by default, but just in case
groupadd --system docker

# Allow docker to be accessible under /.socket
echo -e '{\n\t"hosts": ["unix:///var/run/docker.sock", "unix:///.socket/docker.sock"],\n\t"group": "docker"\n}' > /etc/docker/daemon.json

# Enable and start the unit
systemctl enable --now docker

# Verify installation
docker ps