I have switched the encoding of sent emails, with the patch related in this post:
It works great from one of our server, (in Outlook and gmail)
To: coil@domain.com
Subject: =?UTF-8?B?U3VqZXQgYXZlYyBkw6lzIGFjY8Oow6jDqMOow6jDqMOobnRzIOKCrCAhIQ==?=
Date: Wed, 31 Aug 2011 16:37:38 +0200
The subject and body are OK.
But on the second server, there is a weird behaviour, I get this email header for the subject:
To: coil@domain.com
Subject: =?US-ASCII?B?PT9VVEYtOD9CP1UzVnFaWFFnWVhabFl5?=
=?US-ASCII?B?Qmt3Nmx6SUdGalk4T293NmpEcU1Pb3c2akRxTU9vYm5SeklPS0NyQ0FoSVE9?=
=?US-ASCII?B?PT89?=
It seems that the subject is re-encoded and forced to US-ASCII, it leads to an unreadable subject and the UTF8 characters are lost in the body.
The same test script and email class is used on both servers.
Any idea where this bug could comes from ?