At 10:31 04/10/00 -0400, Jacob Haller wrote: >I don't see where either piece of mail was delayed in the messages >whose headers you included. I don't know if that's relevent or not. Hello. No, this headers aren´t relevant. Are an example. I can send you the delayed mail >One thing that comes to mind is that I know that some versions of >Netscape for Unix had a bug which resulted in an invalid Sender: The user have netscape in window 95 >address being inserted into mail headers generated by the Netscape >mail client. If your users are using Netscape for Unix and your >mailing list is set to Send=Public that might have something to do >with the problem, as LISTSERV will take the Sender: address to be the >"real" address of the person sending the message and if that address >is not subscribed will delay distribution of the message for ten >minutes. (This is referred to as a "spam quarantine" and is to allow >LISTSERV time to see if, for instance, the same email address sends >the same message to a bunch of other mailing lists.) This wouldn't >explain the four-day delays you mention, though, so even if this is >part of the problem it probably isn't the whole story. > >I haven't used Netscape for Unix myself so the following information >is culled from various sources, mainly mailing lists such as this >one. I can't vouch for its total accuracy, therefore. > >To reiterate, affected versions of Netscape Messenger have the >following bug: If the server login ID doesn't match the LHS of the >e-mail address, Netscape Messenger generates a Sender line to try to >authenticate who the "real sender" is. Unfortunately, this line is >not correctly generated, resulting in a bad Sender line. > >A workaround is to set the mail.suppress_sender_header variable in >Netscape's preferences file (unfortunately not accessible by the user >interface). > >Close Netscape, open `~/.netscape/preferences.js' with a text editor, >and add a line that says: > >user_pref("mail.suppress_sender_header", true); > > From then on, Netscape won't generate the "Sender:" header, and your >message will be RFC822 compliant. > >I hope that you succeed in determining the cause of your problem. Thanks you very much, but if user is using W95, the problem is another one, isn´t it? Best regards, >Thanks, >-- >Jacob Haller, Technical Support >L-Soft international, Inc >http://www.lsoft.com/ > -- ---------------------------------------------------------------- Mercedes Giménez Barrio e-mail: [log in to unmask] Unidad Técnica de Comunicaciones Phone: +34 91 397 51 53 Tecnologías de la Información Fax : +34 91 397 39 14 Facultad de Económicas, E-VI-110 Universidad Autónoma de Madrid Crta. de Colmenar, Km. 15 E-28049 Cantoblanco.España ----------------------------------------------------------------