In <[EMAIL PROTECTED]>, on 04/01/03 at 09:23 AM, "Bryan Ragon" <[EMAIL PROTECTED]> said:
>forgive my confusion, but isn't the aliases file used to route incoming >mail? not deliver mail to a MUA? Or am I thinking about it wrong? Both. "incoming" is to the MTA, not just the box it is on, hence can be 'seen' by outbound messages from shell users, or processes which create such messages, such as a system's daily, wekly, and security runs *as well as* what the MTA does w/r handing a message received via (E)SMTP from a distant box. IOW - it is possible to run traffic that has no Maildir or .courier files and doesn't have a shell account either. Look at the commonly used aliase entries for 'root' and 'postmaster'. 'man sendmail' (system original or courier's version - which ver is installed) explains 'include:' One uses 'aliases', for example, with Ecartis for complex, dynamic, mailing lists, and 'includes' to suck in simpler dynamic lists for smaller mailing lists (tech support folk) so that one need not do makealiases/newaliases or 'make all' every time the list file is changed. Sam has also provided an incredibly powerful 'dynamic delivery mechanism' in courier-mta which can pipe to nested procedures and still keep track of return codes. See the docs and man pages. That one can probably also do laundry and cook you a meal, but has *slightly* more arcane setup needs and learning curve than an alias entry or include. Bottom line: If an entry exists either directly as an alias entry (high maintenance as script has to call for a rebuild of the aliases db after any change) *OR* in a flat-file called by an 'include' (no maintenance so long as you don't change the fully-qualifed file name), *OR* in a file in the path of a mailing-list or such function - addresses so recorded should get different handling than those in the rest of the mailusers area (whatever that is). Does it actually work that way? Well... I have a shell account <name> who can send mail out. I also have a mailing list of the same <name>. Traffic in or out of the mailing list does NOT end up in the shell account's mailbox. YMMV Bill Hacker -- ----------------------------------------------------------- [EMAIL PROTECTED] While a computer virus can be aggravating to the point of illness, it is not *otherwise* transmissable to or from humans or animals. - This oversight will be corrected in the next Windows release. - (we in Hong Kong are wondering if there is an Alpha out already?) -------------------------------------------------------------- >> -----Original Message----- >> From: [EMAIL PROTECTED] >> [mailto:[EMAIL PROTECTED] Behalf Of >> [EMAIL PROTECTED] >> Sent: Monday, March 31, 2003 12:30 PM >> To: [EMAIL PROTECTED] >> Subject: RE: [courier-users] Re: LDAP backend / password expiration >> >> >> KISS method. May not do what you want, BUT.... >> >> How about an 'include:" in your aliases file that refers to a list. >> >> The (dynamic) list is of expired/overdue accounts. Generated by >> ....whatever.. (billing software, even) >> >> It points their traffic to a 'reminder' message mailing list. Or pipes to >> *whatever*... ... instead of their (temporarily/permanently inaccessible) >> maildir.... >> >> That way, your billing system - or *whatever* needn't know much about the >> mta, nor the reverse, nor is there a lot of coding.... >> >> HTH >> >> Bill Hacker >> >> In <[EMAIL PROTECTED]>, on 03/31/03 >> at 08:50 AM, "Bryan Ragon" <[EMAIL PROTECTED]> said: >> >> >hehe..... straight and to the point. From looking over the code, it >> >looks like I could add the functionality by editing authldaplib.c. >> >Probably in auth_ldap_do3. I could use copy_value to retrieve the >> >expiration date from ldap. I figure the code could go either just before >> >or after: >> >> > if (rc == 0 && callback) >> > rc= (*callback)(&auth, arg); >> >> >What function does the *callback run? and does it have any bearing on if >> >I put my code before or after. If after I basically planned on something >> >like: >> >> >// pseudo-pseudo code obviously :) >> >if (rc == 0) >> >{ >> > copy_value(my_ldap_fp, expire_date, "expire", &p, user) ; >> >> > // would need to find c functions /librarys for modifying unix >> >timestamps. >> > if (//compare dates) >> > rc = -1 ; >> > else >> > rc = 0 ; >> >} >> >> >This would return a bad password (permanent error) correct? Is there >> >anyway that I could return a custom negative return code, and have the >> >calling function report back a custom error message to the user? i.e. >> >"Password has expired" (I know, very original, thoughtful and catchy >> >error message :-) (I would need to add the "custom error message" >> >functionality to both imapd and pop3d. >> >> >Thanks for the help, >> >Bryan >> >> >> -----Original Message----- >> >> Bryan Ragon writes: >> >> >> >> > Does courier-imapd and pop3d support a password expiration >> >> field in an ldap >> >> > backend? I have the ldap setup to allow imap & pop access >> >> already, but I >> >> >> >> Nope. >> >> >> >> ------------------------------------------------------- >> This SF.net email is sponsored by: ValueWeb: >> Dedicated Hosting for just $79/mo with 500 GB of bandwidth! >> No other company gives more support or power for your dedicated server >> http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ >> _______________________________________________ >> courier-users mailing list >> [EMAIL PROTECTED] >> Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users >> ------------------------------------------------------- This SF.net email is sponsored by: ValueWeb: Dedicated Hosting for just $79/mo with 500 GB of bandwidth! No other company gives more support or power for your dedicated server http://click.atdmt.com/AFF/go/sdnxxaff00300020aff/direct/01/ _______________________________________________ courier-users mailing list [EMAIL PROTECTED] Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users
