[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: amd + "traditional" NIS: Unable to automount (i386)
- To: misc_(_at_)_openbsd_(_dot_)_org
- Subject: Re: amd + "traditional" NIS: Unable to automount (i386)
- From: "C. Bensend" <benny_(_at_)_bennyvision_(_dot_)_com>
- Date: Tue, 18 May 2004 18:26:16 -0500 (CDT)
> Yes amd can grok maps through NIS, it's just that the format conflicts
> with whatever server you're using (is it is a different amd on that
The NIS server is actually on an IRIX machine. Don't ask. No, I can't
change that right now. Yes, I hate it.
> There is a CAVEAT in the man page, "Some care may be required when
> creating a mount map."
I took close notice of that.
> From this it seems very obvious that the format of your amd.master map
> is different format and amd does not like it, but then you seem to have
> figured that out already. You need to create a proper map on the NIS
> server, perhaps under a different name for different implementations.
> type:=nfs;rhost:=hostname;rfs:=/home/user ...
Yes, it's very obvious the format is wrong. :)
However, the format of the auto.* maps we're using are pretty old-school
and "standard". I'm defining "standard" as the fact that all of our
AIX 4.1.5 - 5.1, HP-UX 10.20, 11, 11i, IRIX 6.2 - 6.5.16, Solaris 6, 7
and 8, and every linux box in-house can grok everything they need from
the NIS maps automagically. No tweaking needed. Plug in the NIS domain
and reboot, and It Just Works(tm).
Feel free to replace "standard" above with "stupid" or "disgusting" or
whatever. I'd probably agree.
I'm not going to get into NIS history to see where the different methods
parted ways (if indeed they ever shared a common path), I guess I was
just curious to know if OpenBSD's amd could grok these NIS map formats
without redoing them. That's not worth it for this box.
I guess amd cannot hook into this NIS domain without some work?
"Oh, the Jedis are going to feel this one!" -- Professor Farnsworth,