> Therefore we should concentrate on finding the real cause(s) of the > problem and try to solve them. Full acknowledge. That's why I wrote "until you have found the error". But in the meanwhile, Ram possibly does not want to loose mails. And it's easier to set a well documented option in postfix than hacking a dangerous quick and dirty patch and recompile cyrus instead to give 4xx responses. I just fear that I cannot be of any help with debugging/fixing this bug. Sorry. Best, Daniel On 12.10.2006 13:46, Farzad FARID wrote: > Hi, > > Daniel Eckl wrote: >> Hmmm, well, a non existent mailbox is no temprary failure, it's very >> permanent. okay, in this case it's a false error, but to make another >> error to circumvent that is not a smooth solution. >> >> Perhaps you might want to set "soft_bounce = yes" in postfix' main.cf >> until you have found the error. >> See: http://www.postfix.org/postconf.5.html#soft_bounce >> > But in fact there are cases in a Cyrus Murder configuration when lmtpd > returns a failure with a 5xx code because "mupdate" is simply not ready > or temporarily unavailable. So simply setting soft bounces in Postfix is > IMHO overkill and hides a "misfeature" in the Cyrus Imapd code. > > Therefore we should concentrate on finding the real cause(s) of the > problem and try to solve them. I'll try to file a bug report by > describing a reproducible test case for my setup. > > Ram, do you also use mupdate or any external source of information that > could be the cause of the error lmtp gives you? > > Regards > ---- Cyrus Home Page: http://cyrusimap.web.cmu.edu/ Cyrus Wiki/FAQ: http://cyrusimap.web.cmu.edu/twiki List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
