Date: Thu, 05 Dec 1996 11:07:37 -0800 From: Paul Traina <pst@shockwave.com> To: =?KOI8-R?Q?Andrey_Chernov=2C_=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= <ache@nagual.ru> Cc: Peter Wemm <peter@spinner.dialix.com>, "Jordan K. Hubbard" <jkh@freefall.freebsd.org>, CVS-committers@freefall.freebsd.org, cvs-all@freefall.freebsd.org, cvs-include@freefall.freebsd.org Subject: Re: cvs commit: src/include utmp.h Message-ID: <199612051907.LAA04245@precipice.shockwave.com> In-Reply-To: Your message of "Thu, 05 Dec 1996 21:11:26 %2B0300." <Pine.BSF.3.95.961205210444.222A-100000@nagual.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
We already change UT_NAMESIZE, lets change UT_HOSTSIZE in the same round Or change it back. :-) Usually bigger UT_HOSTSIZE gives much better statistics, because too many of hosts now exceedes 16 char limit. I agree, 64 isn't bad, as you suggest. One other "cautionary note" (I'm full of 'em today). A lot of stuff assumes that wtmp records are the same as utmp records. Make sure we check for programs that access wtmp as well as utmp. We're not totally screwed if we have a shared libutil() because many updated programs use logwtmp(), but nonetheless, it's something that need be checked for lossage. Just don't break my xterms please. :-(
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612051907.LAA04245>