mirror of
https://codeberg.org/forgejo/forgejo.git
synced 2024-12-23 17:14:24 +00:00
Update documentation for Implicit TLS (#16220)
As per RFC 8314, it is now recommended to prefer TLS over STARTTLS. Fix #16160 Signed-off-by: Andrew Thornton <art27@cantab.net>
This commit is contained in:
parent
681e81babd
commit
6a083a7234
|
@ -1387,8 +1387,8 @@ PATH =
|
|||
;; Mail server
|
||||
;; Gmail: smtp.gmail.com:587
|
||||
;; QQ: smtp.qq.com:465
|
||||
;; Using STARTTLS on port 587 is recommended per RFC 6409.
|
||||
;; Note, if the port ends with "465", SMTPS will be used.
|
||||
;; As per RFC 8314 using Implicit TLS/SMTPS on port 465 (if supported) is recommended,
|
||||
;; otherwise STARTTLS on port 587 should be used.
|
||||
;HOST =
|
||||
;;
|
||||
;; Disable HELO operation when hostnames are different.
|
||||
|
|
|
@ -550,9 +550,9 @@ Define allowed algorithms and their minimum key length (use -1 to disable a type
|
|||
- `DISABLE_HELO`: **\<empty\>**: Disable HELO operation.
|
||||
- `HELO_HOSTNAME`: **\<empty\>**: Custom hostname for HELO operation.
|
||||
- `HOST`: **\<empty\>**: SMTP mail host address and port (example: smtp.gitea.io:587).
|
||||
- Using opportunistic TLS via STARTTLS on port 587 is recommended per RFC 6409.
|
||||
- As per RFC 8314, if supported, Implicit TLS/SMTPS on port 465 is recommended, otherwise opportunistic TLS via STARTTLS on port 587 should be used.
|
||||
- `IS_TLS_ENABLED` : **false** : Forcibly use TLS to connect even if not on a default SMTPS port.
|
||||
- Note, if the port ends with `465` SMTPS/SMTP over TLS will be used despite this setting.
|
||||
- Note, if the port ends with `465` Implicit TLS/SMTPS/SMTP over TLS will be used despite this setting.
|
||||
- Otherwise if `IS_TLS_ENABLED=false` and the server supports `STARTTLS` this will be used. Thus if `STARTTLS` is preferred you should set `IS_TLS_ENABLED=false`.
|
||||
- `FROM`: **\<empty\>**: Mail from address, RFC 5322. This can be just an email address, or
|
||||
the "Name" \<email@example.com\> format.
|
||||
|
|
Loading…
Reference in a new issue