Date: Tue, 13 Jul 1999 14:27:54 -0700 (PDT) From: Matthew Dillon <dillon@apollo.backplane.com> To: Noriyuki Soda <soda@sra.co.jp> Cc: Jason Thorpe <thorpej@nas.nasa.gov>, "Brian F. Feldman" <green@FreeBSD.ORG>, bright@rush.net, dcs@newsguy.com, freebsd-hackers@FreeBSD.ORG, jon@oaktree.co.uk, tech-userlevel@netbsd.org Subject: Re: Replacement for grep(1) (part 2) Message-ID: <199907132127.OAA80947@apollo.backplane.com>
next in thread | raw e-mail | index | archive | help
:> Secondly, for such a server to fail to run is just as bad as if
:> the system were to run out of swap.
:
:> IRIX has a swap reservation flag too, a left-over from the SysV days.
:> It is a totally useless flag.
:
:That's wrong.
:On such systems, critical server has a chance to save it's data to
:filesystem.
:On 4.4BSD derived systems, it cannot be guaranteed.
:--
:soda
You are assuming that the situation actually occurs. In real life,
it will not occur unless the critical server is running away with
memory.
I have never, ever run one of BEST's servers out of swap. It has never
been an issue.
And, I can only repeat, again, that long before a reasonably configured
FreeBSD system runs out of swap it would become unusable from the I/O
overload.
And you also haven't bothered to address my other point: In order to
configure a system that guarentees backing store, you need to configure
that system with 8x or more swap then you would a normal system. If
you think there was a chance of a normal system running out of swap,
what do you think would happen if you configured a normal system with
8x swap?
-Matt
Matthew Dillon
<dillon@backplane.com>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199907132127.OAA80947>