Run smartctl on every disk, read the five numbers that actually predict failure on SATA and NVMe drives, check software and hardware RAID, and set up smartd so the server warns you before a disk dies.
Publish one TXT record at _dmarc, read the daily reports, fix the services that fail alignment, then move from p=none to quarantine and reject without blocking your own invoices and newsletters.
Find which of the four upload gates refused the file, raise client_max_body_size in the domain’s nginx directives, match the PHP limits, fix ModSecurity if it’s involved, and prove it with a test upload.
Register ns1 and ns2 as glue records at your registrar, add matching A records, set them as your reseller nameservers in WHM, move existing clients over and prove every step with dig.
IMAP keeps mail on the server and syncs every device; POP3 moves it to one computer. When each makes sense, the ports to use, and how to switch an old POP3 setup to IMAP without losing years of mail.
Unlock the domain, get the EPP or auth code, start the transfer at the new registrar and verify it with whois, without downtime for your website or email. Includes the 60-day rules and .ca specifics.
Run the cPanel Virus Scanner and ImunifyAV, check the folders scanners miss, quarantine safely and find how the attacker got in, so the infection doesn’t come back next week.
A 503 on a Plesk for Windows site means its IIS application pool has stopped. Restart it in Plesk, read the System log entry for the real cause, then fix Rapid-Fail Protection or a broken pool identity.
Chrome says the certificate doesn’t match the site name. One openssl command shows which names it really covers; then add the missing www or subdomain, bind the right certificate, or rule out your own device.
A 508 error means your account hit a CloudLinux limit. Read the fault counts in cPanel Resource Usage, name the limit (entry processes, CPU, memory or I/O) and fix that one instead of guessing.