Core
ServersProvisioning in 5 minutes SitesWordPress, Laravel, PHP DeployOn push, with no downtime DatabasesMySQL, users, privileges DockerContainers alongside your sitesAI and automation
AI assistantManage your server in plain words MCP server80 tools for Claude and Cursor Operations centerEvery task step with logs Auto-tuningSettings tuned to your hardwareObservability
MonitoringMetrics, APM and alerts LogsNginx, PHP, MySQL, application BackupsTo your storage, encrypted ClustersServers and sites togetherSecurity and access
SecurityFirewall, Fail2Ban, isolation NetworkingInterfaces, DNS, diagnostics Web terminalCommands without an SSH client ConfigurationPHP, Nginx, MySQL from the panelManage servers from Claude and Cursor
Who it's for
WordPress2-click install, WP-CLI LaravelQueues, Horizon, Reverb For web agenciesAll clients in one panel For freelancersClient servers in one placeComparisons
Deploykin vs Laravel Forge Deploykin vs Ploi Deploykin vs RunCloud Deploykin vs ISPmanagerA running server connects as it is: the panel scans its packages, sites and databases. There is a separate wizard for Ploi.
How to connect a running serverWhat's tracked
The agent takes a sample every minute. The server dashboard shows current values and charts for an hour, 6 hours, a day or a week.
Current CPU load, load average 1/5/15 min, number of running processes.
RAM and swap partition usage. See when memory is running out before the OOM killer kills a process.
Device, mount point, FS type, size/used/free, usage percentage, unallocated space.
RX/TX bytes and packets per network interface. Traffic history.
Detects virtualization overhead. Shows how much CPU time the hypervisor takes from your VM.
Requests/second, active processes, total processes. See whether you have enough workers for the current load.
HTTP status, response time, online/offline detection for every site.
Nginx, MySQL, Redis and Supervisor processes are checked every 5 minutes. If a service fails, you get an alert.
Diagnostics
The process hogging the CPU, the partition running out of space and the site short on workers are found in a minute — without SSH and top.
Running processes sorted by CPU, memory and RSS. Filter by PID, user or command and send a signal right from the panel.
Devices, mount points, filesystem type, size and usage. See unallocated space and resize availability.
/33 / 80 GB · 41%
/mnt/data166 / 240 GB · 69%
Unallocated0 GB
Full in~20 days
Every 5 minutes the panel requests each site and records whether it responds and in how many milliseconds. Next to it are requests per second and the busy workers of its PHP-FPM pool.
Alerts
Every alert type can be routed to any number of channels. You set the thresholds for CPU, memory and disk.
CPU and memory are compared against a 15-minute average, so a short peak does not raise an alert. For disk, the panel warns at your threshold and again at 90% and 95%.
90%90%80%90%95%Telegram, email and the in-app inbox. Each channel has its own delivery hours: outside the window, alerts are not lost — they wait until it opens.
No response to monitoring for more than 15 minutes.
The 15-minute average is above the threshold — 90% by default.
The same rule for memory: a 15-minute average.
At your threshold, then again at 90% and 95%.
Nginx, MySQL, Redis or another systemd service entered the failed state.
Horizon or a queue worker went down and Supervisor could not bring it back.
14, 7 and 3 days before expiry, and on the day it expires.
A site deployment finished with an error.
A backup profile did not complete.
A successful login from an address not seen before.
Files changed since the last deploy, or a dependency vulnerability was found.
An AI client was denied: wrong scope, access level or a dangerous command.
APM and analytics
Latency, errors and traffic are computed from the Nginx access log: nothing is installed in your code and nothing is loaded on the page.
Latency percentiles and error rate for every endpoint. Not a line of code in your application — metrics are computed from the web server logs.
POST
GET
GET
GET
Requests over a second
5003.10 s2004.82 sRequests, unique visitors, traffic, HTTP statuses, top pages and referrers. Granularity — by day.
Top pages
Metric collection
Deploykin installs its own metric collection agent on the server. It reads the values on the spot and pushes them to the panel itself — continuously, with nothing for you to do.
Next sample in60 s
What the agent collects
In the panel 203.0.113.21 Connect done in 4 min 52 s
Connect a server — and monitoring and alerts start working automatically.
We use cookies and the Yandex Metrica web-analytics service (including Session Replay) to improve the site. Details are in the privacy policy.