[Prev][Next][Index][Thread]

SUMMARY: MH aliases and NIS



My original query:

some questions about mail:
> 1) is it possible to handle an entry such as :
> mylist : ":include:/home/anyuser/user-correspondants.list"
> through NIS (in file /var/yp/src/mail.aliases for example) ? The purpose of
> this is to let users handle "well-known" lists without giving them root
> privileges on our NIS master.
> 2) is it possible to send a message to an alias list, without any single
> receiver of this message having the list of all the member displayed ? Some of 
> the messages are made 95% of mail addresses and 5% of real stuff !
> 3) according to CERT advisory CA-94:12, OSF/1 version 2.0 sendmail is unsafe 
> (the -oE and the -debug flaws), what about 2.1 and 3.0 versions ?
> 4) for performance reasons, I want to enforce another mail relay than the
> default one for my domain. I was not able to achieve this through the
> mailsetup utility (I performed the steps required but the relay did not
> change even after an rcinet restart). Any suggestions ?
> 5) it seems that NIS aliases do not allow the form Ronald.Doe but do accept
> ronald.doe ? Any clues ? (thank you Ronald 7:^]

I received one answer from 
Bonfield James <jkb@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>

which resolved point 3) :
> This is the only one I'm qualified to comment on. The -d problem has indeed
> been fixed in 3.0 - I spoke to the chap at Digital about it myself. I think it
> maybe fixed in 2.1 too, but I can't remember. Try something like:
> "sendmail -d3294967296.66".
> This will usually crash on buggy systems. although it's hardly a proof of when
> the bug doesn't exist.
> The -oE bug I thought was only in the BSD sendmail v8; I was unaware of DECs
> sendmail supporting this. It's not listed in the manual page anyway.

Note: James is right, only the debug flaw is relevant for OSF.

and point 4) :
> I'd guess changing the definition for 'R' would do this. See the "DR" line
> near the top of your sendmail.cf. No guarantees mind.
> As for NIS, I wouldn't touch it with a barge pole. In fact I'd feel happier
> writing my own from scratch :)

Note: Right again, even if I supposed Cambridge to have faster boats than barges
(not to be repeated to some of my friends in JET, Oxford anyway ;-)

I tested out James's tip and found :
OSF/1 version 3.0 and 2.1 : no crash, version 1.3 : crash (Segmentation fault)

Thanks,
+--------------------------------------------------------------+
| Herve DEMARTHE   %^)    E-Mail:   demarthe@xxxxxxxxxxxxxxxx  |
| CEA/DSM/DRFC/STEP       Tel: +33 42257527  Fax: +33 42252661 |
| CEN Cadarache Bt 506    13108 St Paul Lez Durance     FRANCE |
|        <<< Apprentiz de todo, Maestro de nada ... >>>        |
| All opinions expressed herein are mine and not those of CEA. |
+--------------------------------------------------------------+