Date: Sun, 18 May 1997 11:38:46 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= <ache@nagual.pp.ru> To: Peter Wemm <peter@spinner.dialix.com> Cc: Bruce Evans <bde@zeta.org.au>, peter@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-lib@FreeBSD.ORG Subject: Re: cvs commit: src/lib/libc/gen sleep.c Message-ID: <Pine.BSF.3.96.970518113230.815A-100000@nagual.pp.ru> In-Reply-To: <199705180652.OAA04790@spinner.DIALix.COM>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 18 May 1997, Peter Wemm wrote: > Apache doesn't fail with non-BSD systems that sleep doesn't restart after > non-ALRM signals, eg: sunos, solaris, hpux, digital unix, etc.. This may be caused by another SYSVism, i.e. double features negate each other. > Apache doesn't use signals internally other than SIGALRM anyway. (well, > the user can send SIGHUP and SIGTERM, but that's an exception, not the > normal case during operation) It forks a lot, so got SIGCHLD at least. > eg: from Digital Unix: > If the sleep() function returns because the requested time has elapsed, it > returns 0 (zero). If the sleep() function returns because it caught a > signal, the function returns the number of seconds remaining in the suspen- > sion. Maybe we can keep SYSV behaviour too... Lets fix 'killed by SIGALRM' bug first, then I'll see, how it affects Apache. > I noticed that apache is having a lot of trouble with fin_wait_2 when using > keepalive connections to unix netscape3 and doesn't run very well with -X > when large files are transferred (more than socket buffer I suspect). > I found a mention of this in the docs where it explicitly listed Netscape > on FreeBSD as one known rogues that don't do client keepalive properly. It is not a case here, I test it with Lynx alone and Lynx can't use keepalive. -- Andrey A. Chernov <ache@null.net> http://www.nagual.pp.ru/~ache/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.970518113230.815A-100000>