Date: Thu, 11 Jun 1998 15:04:09 -0400 From: Garance A Drosihn <drosih@rpi.edu> To: freebsd-current@FreeBSD.ORG Subject: Re: ulimit... Message-ID: <v04011706b1a5d8932b5c@[128.113.24.47]> In-Reply-To: <357FDD32.F5DBD872@camtech.net.au> References: <199806102040.QAA07676@auchroisk.pdl.cs.cmu.edu> <357F25D4.524A6BC6@ver1.telmex.net.mx>
next in thread | previous in thread | raw e-mail | index | archive | help
At 11:05 PM +0930 6/11/98, Matthew Thyer wrote: > Administrators of large multiuser systems expect these kind of > limits to be on by default for 'ordinary users'. > > Dont most vendors UNIXes have limits on by default ? > > The single user workstation machine has the one power user that > can afford to get bitten once by this for the knowledge it gains > them. I'm new to running my own FreeBSD systems, and in fact I just got bit by these resource limits while doing something yesterday. While it was inconvenient, I do think it's better to have the limits set at "reasonable" levels and let people run into them, then to not have them set. Say you do have someone who writes a fork-bomb by mistake (that often happens in our systems programming class when students write their first program that calls fork()...). A real run-away program can kill off the entire system -- and then you don't really know *why* the system went haywire. Having a single user run into a limit is better than having the entire system come to a standstill. And now that I know about /etc/login.conf, it looks pretty interesting. Note that in my case, the machine is basically a one-person machine, but still I preferred running into the limit on one session than having a system-wide problem. --- Garance Alistair Drosehn = gad@eclipse.its.rpi.edu Senior Systems Programmer or drosih@rpi.edu Rensselaer Polytechnic Institute To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?v04011706b1a5d8932b5c>