A single line saying "my site is down" often starts the longest conversation a support team has all day. The engineer first has to ask which site, since when, what error appears and where it happens, and every round trip costs time. When the first message already contains the essentials, many problems are solved in one reply. Here is how to write that message and which simple checks to run before you send it.
What the first message should contain
The key is to identify the exact service. One account may hold several domains, so write the full domain name instead of "my site". Then describe when the problem started, what you see and what changed beforehand.
- the full domain or subdomain, for example
shop.example.uz; - the approximate start time and whether it is constant or intermittent;
- the exact error text or a screenshot;
- recent changes: a plugin install, DNS edit or PHP version switch;
- whether it happens everywhere or only on your computer or network.
The last point matters a lot. If the site fails at home but works on mobile data, the problem is probably your network or a blocked IP address, not the server.
Checks you can run yourself
A few minutes of testing often shows where the problem is. Paste the output as text into your ticket so nobody has to ask for it again.
nslookup example.uz
nslookup -type=mx example.uz
These show the domain's IP address and mail server. If the IP differs from the one shown in your hosting cabinet, the issue is DNS. After changing nameservers, remember that new values take time to propagate.
curl -I https://example.uz/
This fetches only the response headers. 200 means the site works, 403 points to permissions, 500 to an error in site code, and 502 or 504 to a backend process not responding. A certificate error from curl points to SSL.
tracert example.uz (Windows)
traceroute example.uz (Linux, macOS)
If the site does not load at all, this shows where packets stop. Send the full output, not just the last line.
A real-world comparison
The owner of a small building materials shop in Tashkent writes in the morning: "Site not opening, check urgently." Support asks for the domain, the reply comes an hour later, then they ask for the error text, then the start time. A simple issue takes half a day.
Now the same issue written differently: "example.uz shows error 500 since 09:20 today. Last night I updated a caching plugin in WordPress. Same on mobile data. curl -I output attached." The engineer opens the error log, checks the plugin and finds the cause in minutes, probably fixed by disabling that plugin temporarily.
Be honest about severity
If every ticket is marked "very urgent", the label stops meaning anything. When the whole site or mail is down, say so plainly: "the site does not load at all, orders have stopped". A shifted image or a request for a new mailbox is a normal question. Honest priorities let truly critical issues move up the queue, and your urgent tickets will be taken seriously.
What you should never send
Never share passwords, cabinet login codes, bank card details or SMS confirmation codes in a ticket or messenger. Genuine support staff will not ask, because they already have the access they need. If someone claiming to be support asks for a password by phone or Telegram, treat it as fraud. Before sending logs or config files, replace database passwords and API keys with asterisks.
Keep the conversation in one place
Writing about one problem through a ticket, phone, Telegram and email at once does not speed things up; it scatters information. For sayt.uz customers the ticket section of the cabinet is the best channel, because the history is kept and engineers can see earlier messages. Open a new ticket for a new problem instead of reusing an old, resolved one. Once it is fixed, note down what helped, which saves time next time for both sides.