some senders on this maillist use a version of thunderbird that adds more then one references header and this imho breaks threading :( -- xpoint
On 8/7/2010 6:24 PM, Benny Pedersen wrote:> some senders on this maillist use a version of thunderbird that adds > more then one references header and this imho breaks threading :(Maybe it would help if you actually said *who* was doing this? I don't see any obviously broken threads, at least in the last few days...
Benny Pedersen wrote:> >some senders on this maillist use a version of thunderbird that adds >more then one references header and this imho breaks threading :(Whether or not this would break threading depends on the MUA or other software doing the threading. For example, Mailman's pipermail archiver prefers In-Reply-To: over References: for threading. Multiple References: headers is a clear violation of RFCs 2822 and 5322 (RFC 822 is not clear on this point). It would be helpful to those who may be using this broken Thunderbird and perhaps others to know which Thunderbird version(s) has/have this bug. -- Mark Sapiro <mark at msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
Benny Pedersen wrote:> maybe not thunderbirds fault, but in all cases thunderbird was used in > the quarantined mails hereThunderbird has nothing to do with amavisd-new... it may be what you are using to view the quarantined messages, but it is just an MUA... If you are seeing these extra headers only in messages that have been quarantined, the problem is with amavisd-new. Otherwise, please point to a specific Message-ID that exhibits the problem. -- Best regards, Charles
On 11:59 AM, Benny Pedersen wrote:> > maybe not thunderbirds fault, but in all cases thunderbird was used in > the quarantined mails hereI looked at the last several list messages from Timo, and they mostly are composed with "X-Mailer: Evolution 2.28.3" and some with "X-Mailer: Apple Mail (2.1081)". I saw none from any version of Thunderbird, and none with more than one References: header. Perhaps you could give a Message-ID or other identifying information about one of the offending messages. -- Mark Sapiro <mark at msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan
On man 09 aug 2010 16:53:58 CEST, Mark Sapiro wrote> Perhaps you could give a Message-ID or other identifying information > about one of the offending messages.Message-ID: <4C5D97DF.3060106 at list.ims.co.at> Message-ID: <4C5D7D90.3030903 at list.ims.co.at> Message-ID: <4C5C7358.2090608 at test.ims.co.at> Message-ID: <4C5C575B.3010808 at test.ims.co.at> Message-ID: <4C5C4B98.3050103 at test.ims.co.at> Message-ID: <4C5C215F.9050304 at list.ims.co.at> Message-ID: <4C5C0486.8020503 at list.ims.co.at> Message-ID: <4C5BD654.10208 at list.ims.co.at> Message-ID: <4C5BBC0F.8050807 at list.ims.co.at> Message-ID: <4C5B1B87.8090705 at list.ims.co.at> -- xpoint
Timo, oh man, I see these are really my messages. Lets try if this new one has got only one reference... Seems to be a problem with my postfix virtual alias going into test domain with own dovecot-imap. Thunderbird tries to insert 2 references, but no sure why. Btw, did you look at the LSUB/listescape bug ? I've got quite lost in your pointer jungle... On 08/09/2010 05:23 PM, Timo Sirainen wrote:> On Mon, 2010-08-09 at 17:19 +0200, Benny Pedersen wrote: >> On man 09 aug 2010 16:53:58 CEST, Mark Sapiro wrote >> >>> Perhaps you could give a Message-ID or other identifying information >>> about one of the offending messages. >> Message-ID: <4C5D97DF.3060106 at list.ims.co.at> >> Message-ID: <4C5D7D90.3030903 at list.ims.co.at> >> Message-ID: <4C5C7358.2090608 at test.ims.co.at> > .. > > Yeah, those are buggy and have multiple References: headers. But they're > not from me like you said.. > > Samuel: Fix your client. :) > >
On man 09 aug 2010 17:45:20 CEST, Samuel Kvasnica wrote> Timo, oh man, I see these are really my messages. Lets try if this new > one has got only one reference...X-Amavis-Alert: BAD HEADER SECTION, Header field occurs more than once: "References" occurs 8 times -- xpoint