Date: Sun, 21 Aug 2005 12:19:46 +0000 From: Randy Bush <randy@psg.com> To: Robert Watson <rwatson@FreeBSD.org> Cc: arch@FreeBSD.org Subject: Re: [CFR] reflect resolv.conf update to running application Message-ID: <17160.29026.1724.73259@roam.psg.com> References: <ygefyt4yiaz.wl%ume@mahoroba.org> <20050821003536.P14178@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> (1) Has anyone characterized the significant of the cost of doing a stat() > for every DNS lookup for a significant workload? Does it matter? > Most performance-critical network paths don't do name lookups in order > to prevent indefinite stalls in lookup, but doing, say, 1,000 > additional stats a second is not a small issue. are you gonna be doing all that many of them, given that the dns queries over the net will not be blindingly fast to say the least? > (2) By reading the configuration file more frequently and more quickly > after a change, we increase the chances of a race condition in which > the resolve reads a partially written resolv.conf file during an > update. Does this happen in practice? I've always been very leery of > re-reading configuration files automatically based on a time-stamp, as > updates to files are not atomic at all. hmmmmm dunno about others' use patterns, but in my world, resolv.conf only changes when my mobile platform moves layer three binding (i.e. not an an 802.11 access point move). so the frequency is low, and it is usually not issuing dns queries as i move. but when i get to the new binding, i am annoyed if i have to whack the resolver. randy
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?17160.29026.1724.73259>