On Sat, 20 Oct 2001, Kenneth E. Lussier wrote: > I am using Amanda as a backup system, running on an old SPARCstation 20, > SunOS 5.5.1 (sendmail 8.11.3). The reports that amanda sends out on a > daily basis are rejected by my mail server (Courier mail server running > on Debian 2.2r3) because they contain 8-bit encoded text, but they don't > have the correct MIME headers.
Too many variables. Time for some "fault isolation", i.e., keep removing parts until it starts working. First, try to determine definitively whether or not the message is pure 7-bit ASCII, or if it contains 8-bit data, regardless of any encoding. If there isn't a trace of 8-bit data, then Courier is giving a false positive. If there is 8-bit data, you might want to jump right to Amanda, and look at the output it is generating. Is it unencoded 8-bit data that is being encoded incorrectly? If so, is the encoding being done by Amanda, Sendmail, or something between the two? Or perhaps Amanda is outputting 8-bit data, and it is never being encoded at all? Or is Amanda generating 7-bit data, and something else is guming up the works? > I could fix the Courier code to make it accept mail without checking for > the missing headers or the rfc violations, but I don't think that that > is really the problem. Depends on how you define "problem". A lot Internet software follows the RFC robustness principle: Be strict in what you generate; be lenient it what you accept. Courier, on the other hand, seems to be a stickler for the rules, judging from its FAQ. Which is attitude is best is largely personal preference. -- Ben Scott <[EMAIL PROTECTED]> | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ********************************************************** To unsubscribe from this list, send mail to [EMAIL PROTECTED] with the following text in the *body* (*not* the subject line) of the letter: unsubscribe gnhlug **********************************************************
