PremiumComes with the add-on reqad-account. Installing add-ons

Metrics

Who visits your sites, what they open and the traffic it makes, counted every hour from the web server's log; the log itself, request by request; and what your sites, mail and sessions use of the server's processors and memory. In the account panel they are under Metrics > Visitors and bandwidth, Metrics > Logs and Metrics > Resource usage; your Home page shows the visitors and the bandwidth of this month. Metrics > PHP error log opens the same page as PHP > PHP error log.

Visitors and bandwidth

• Site and period: one site or all of them together; the last 30 days, the last 12 months, or one month. The address of the page keeps your choice, so a reload or a link shows the same.
• Totals: visitors, page views, requests, bandwidth and server errors, compared with the 30 days or the month before. Click one to see it in the chart.
• The chart: per day, or per month for 12 months. Hover a bar for its numbers. Today is lighter, days that were not counted are dashed, and a dot marks a day with more than 1% server errors.
• Lists: pages, the sites your visits came from (search engines and AI assistants are marked), countries, browsers / systems / devices, missing pages (404), the answers (2xx to 5xx), what uses the most bandwidth, bots and the busiest hours. Show all opens the top 50.
• CSV downloads the numbers per day (or per month).

What the numbers mean

Bots (search engines, SEO tools, AI crawlers, scanners) are left out of visitors and page views; their traffic still counts in the bandwidth. Missing pages marked scanner are robots looking for software your site does not have; the others may be old links, which a redirect sends to the right page.

No visitor's address is kept: visitors are counted with a hash that changes every day. Countries use IP Geolocation by DB-IP. Your server's admin can turn visitor stats off; then Visitors and bandwidth is not in your menu (Logs stays). More on how it is counted in Visitor stats & bandwidth.

Logs

One day of a site's web server log at a time: today, yesterday and every older day the server still keeps (usually 52). Two tabs:

• Requests (access_log): every request, newest first, 50 at a time and Show older, with the time, the answer code in colour, the address asked for, and the visitor's IP address, country and browser, or the bot's name. Filter by All, Errors (4xx and 5xx) or Server errors (5xx), Hide bots, or search the address, IP, browser or referring page. Click a request for everything in its line, what the answer means, the raw line to copy, and Only this address; a missing page has Redirect it.
• Web server errors (error_log): what the web server could not do for a request, grouped by kind with how often, when and what it was about: a file the web server may not read, a folder without an index page, PHP not running, PHP too slow, an upload over the size limit, a redirect loop, a request blocked by a security rule... Each kind says what it means and links the page that fixes it. All lines shows the raw lines. PHP's own errors are in the PHP error log.

Live (today) shows new requests as they come, every 5 seconds, while the page is open (for 30 minutes at a time). Download gives the day's log as a .gz file. Visitors' IP addresses are in the log, as on any web server: treat a downloaded log as personal data. On servers where Apache answers the sites directly, the web server's errors go to one log for all sites and the second tab says so.

Resource usage

CPU, memory and processes of your account over the last 24 hours, 7 days or 30 days, checked every minute, against your limits.

• Tiles: CPU, memory and processes now and at their peak, your sites' PHP workers running now and how often they were all busy, and your database connections. Click one to see it in the chart.
• The chart: CPU, memory and processes are stacked by what uses them: SSH, cron and the terminal at the bottom (so the dashed line of their limit compares directly), then each site's PHP, email logins and background jobs (WordPress updates, deployments, apps). PHP workers are drawn per site and database connections per database user, at their highest check. Hover it for the numbers; a dot marks a limit that was reached.
• Limits reached: what happened, when, how often, what it means and the page that helps: all of a site's PHP workers were busy (visitors waited and may have got 502 or 504), a request was stopped after the site's time limit (with the script), a database user used all its connections, the memory limit stopped a process of SSH, cron or the terminal, a new process was refused at the process limit, or they were held at their CPU limit.
• Running now: your account's processes, grouped the same way, with CPU over the last minute, memory and how long each one runs. Stop asks a process to stop and ends it after 5 seconds if it does not; it can only reach your own processes. Refresh reads them again.
• Your limits: what SSH, cron and the terminal may use together (CPU, memory, processes), each site's PHP workers and time limit, and the connections per database user. Your hosting provider sets them.

A task shorter than a minute may not show (except in SSH, cron and the terminal, which have their own counters). Database queries run in the database server and are not counted here, and a site that runs PHP inside the web server is not counted per account. When a site runs out of PHP workers 3 or more times in 24 hours, or the memory limit stops a process, it also shows in Needs attention on your Home page, and you can be told by mail (Profile, Resource limits).

Need a hand? Create a free account or contact us.