Hi, I have some problems with UTF-8 mail subjects and otrs. The first comes from otrs logging. I use Kernel::System::Log::File logging method. It has some problems with message subjects containing non-english characters. I have log lines like this: [Sat Sep 6 23:18:03 2008][Notice][Kernel::System::Ticket::TicketCreate] New Ticket [2008090610000038/Rossz sz<E1>mla] created (TicketID=1192,Queue=Raw,Priority=3 normal,State=new) Where the <E1> should be an 'á'. In addition to that there is a message, which i get from root@localhost (Cron Daemon): "Wide character in print at /usr/share/otrs/Kernel/System/Log/File.pm line 66" I searched this problem on google and it seems to be some kind of perl related issue. The perl code waits for utf-8 data to write to the file, and the data in fact is not utf-8. The original subject is encoded as it should be that way and looks like that: Subject: Rossz =?UTF-8?B?c3rDoW1sYQ==?= So i think there is a base64 decoding problem somewhere. I have an other issue about this topic. Its simply that Thunderbird wraps the subject line if the subject is too long. And otrs tries to merge the wraps. But because of that after some replys the UTF-8 encoded subjects can become garbled in mail clients, because sometimes otrs merge wraps wrong, like this: =?UTF-8?Q?=K=C3?= =?UTF-8?Q?=B6r?= Where the C3 and B6 should be in same line beacause the two are one character together and that way this subject would be 'Kör'. But when wrapped like that it becomes a garbled character. So i think the possible solution would be to change the outgoing mail subject encoding to base64 in otrs. Is it possible? Thanks, Pal Gergely