Path: utzoo!attcan!uunet!lll-winken!lll-lcc!ames!mailrus!cornell!uw-beaver!rice!sun-spots-request
From: auspex!guy@uunet.uu.net (Guy Harris)
Newsgroups: comp.sys.sun
Subject: Re: inodes table
Message-ID: <695@auspex.UUCP>
Date: 20 Dec 88 23:24:22 GMT
References: <8811300754.AA11396@groucho>
Sender: usenet@rice.edu
Organization: Sun-Spots
Lines: 16
Approved: Sun-Spots@rice.edu
Original-Date: 13 Dec 88 18:50:18 GMT
X-Sun-Spots-Digest: Volume 7, Issue 66, message 6 of 10

>With all of the discussion about the text table and the ntext parameter in
>the kernel...Is pstat inquiring of this table correctly?

Yes.

>Should I worry and reconfigure param.c with ninodes calculated with
>fudge factors like ntext?

No.

>Why is this not acting like any other kernel table?

Because entries in the directory name lookup cache require the inode for
the inode referred to by the entry to hang around.  If the system really
needs an inode table slot, it will purge DNLC entries until it either gets
one or runs out of entries to purge, at which point it has *really* run
out of inodes and complains.