Path: utzoo!mnetor!uunet!lll-winken!lll-tis!ames!ncar!noao!mcdsun!mcdchg!clyde!wtr@moss.ATT.COM
From: wtr@moss.ATT.COM
Newsgroups: comp.sys.att
Subject: Re: Changing ulimit on 3b2/600
Message-ID: <24951@clyde.ATT.COM>
Date: 16 Apr 88 15:56:41 GMT
References: <1603@lll-lcc.aRpA> <211@anumb.UUCP> <1122@cadre.dsl.PITTSBURGH.EDU>
Sender: nuucp@clyde.ATT.COM
Reply-To: wtr@moss.UUCP (Bill Rankin)
Organization: AT&T Bell Laboratories, Whippany NJ
Lines: 31
Keywords: 3b2/600, ulimit
Summary: you dont have to patch the kernal

In article <1122@cadre.dsl.PITTSBURGH.EDU> cgw@cadre.dsl.pittsburgh.edu.UUCP (Gray Watson) writes:

>	I'm trying to change the sh's ulimit from 2048 blocks (1 Meg) to
>some large value.  Sure I can do it for root but that doesn't help our
>big database users.
>

We have a routine here on our vaxen called ulimit++,
which is basically a short C program which checks the 
current ulimit, and if it is less than a specific value
(8 Meg in our case), it creates a user shell calling ulimit() 
to raise the ulimit in the new shell.

The program is suid'd to root ( naturally, in order to invoke
ulimit() ), but I've never heard of any security problems with it.

Its about 40 lines of source, but I'll have to check on it's 
'proprietaryness' before posting it.  You can probably write
your own, its a fairly simple to implement.  And it has the 
advantage of raising the ulimit only when you need it, and
you dont have to patch the kernal.

Good luck,

=====================================================================
Bill Rankin
Bell Labs, Whippany NJ
(201) 386-4154 (cornet 232)

email address:		...![ ihnp4 ulysses cbosgd allegra ]!moss!wtr
			...![ ihnp4 cbosgd akgua watmath  ]!clyde!wtr
=====================================================================