I’ve recently added other account to my miab instance (some of the same domain and some from other domains). I’ve noticed that emails going in and out to email addresses not managed on my miab instance work just fine (gmail), but if an email is sent to an account that is managed by my instance, it causes my server to hang.
Once the server is rebooted the email appears in the other account’s inbox, but if I don’t reboot the server remains unresponsive.
I recorded the following from the nginx error logs:
2018/12/07 23:56:06 [error] 1407#0: *952 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 127.0.0.1, server: my.server.domain, request: "OPTIONS /firstname.lastname@example.org/ HTTP/1.1", upstream: "https://127.0.0.1:email@example.com/", host: "127.0.0.1"
This error repeats itself for all the accounts on the server.
I get the following from the access log:
<server ip> - firstname.lastname@example.org [08/Dec/2018:00:02:08 -0500] "POST /Microsoft-Server-ActiveSync?jBAJBBBOaW5lfy78dvzg4MkIwNTc0BAgb0iAHQW5kcm9pZAcBAA== HTTP/1.1" 401 5 "-" "Nine-Onyx_00WW/PPR1.180610.011"
This appears for each account on the server. “Nine-Onyx” is the Android client I use to send/receive my emails.
Any ideas what could cause the server to hang like this?