I run Thunderbird on Windows 10 on a laptop in my office. The laptop connects to the internet over the central ethernet-wired intranet of the office landlord. My email addresses are based on 2 separate websites I own, both held at the same UK host, 34sp.com. Both email accounts use port 587, STARTTLS security and the SMTP server, all as specified by my webhost. Mostly emails work fine. But every couple of weeks, sending emails fails for a couple of hours due to various SMTP errors, SMTP timeout, etc. The stated error varies. Then it all spontaneously comes OK again without me fixing anything. This happens with both email addresses on both websites at once, suggesting the problem is either in the guts of my computer, at the host or in the office intranet. [ie I'm assuming the problem is not with the settings in TB] I've tried restarting TB, and restarting the computer, to no effect. ( I didn't try restarting in Safe mode.) Yesterday after restarting I tried telnetting the host. I am only a script kiddy with telnet, but I did this: telnet smtp.mywebsitename.co.uk 587 and got this: 220 smtpauth2.mailarray.34sp.com ESMTP Exim 4.92 Tue, 25 Feb 2020 18:51:03 +0000 followed about a minute later by: Connection to host lost. and this was identically the same, barring the timestamp, both when the fault was occurring and later on when it had spontaneously fixed itself. First question, is my use of telnet correct? Then, is the response correct and does it help diagnose anything? Is there anything else I can do to diagnose things? I've read the help guides in this forum and they don't appear to apply to this kind of intermittent problem. And overall, what kind of thing might make SMTP fail for a couple of hours, then come back OK on its own? Many thanks Andy