Date: Wed, 29 Jun 2016 06:32:03 +0200 From: Edward Tomasz =?utf-8?Q?Napiera=C5=82a?= <trasz@FreeBSD.org> To: Mateusz Guzik <mjguzik@gmail.com> Cc: Hongjiang Zhang <honzhan@microsoft.com>, "freebsd-fs@freebsd.org" <freebsd-fs@freebsd.org> Subject: Re: ufs freeze does not work Message-ID: <20160629043203.GA82400@brick> In-Reply-To: <20160628195731.GA21323@dft-labs.eu> References: <CO2PR03MB2215FAA1AB86A669039B9540B5210@CO2PR03MB2215.namprd03.prod.outlook.com> <20160628065432.GA20716@brick> <SN2PR03MB2224903F61C7DD576EDDB05DB5220@SN2PR03MB2224.namprd03.prod.outlook.com> <20160628185523.GA82035@brick> <20160628195731.GA21323@dft-labs.eu>
next in thread | previous in thread | raw e-mail | index | archive | help
On 0628T2157, Mateusz Guzik wrote: > On Tue, Jun 28, 2016 at 08:55:23PM +0200, Edward Tomasz Napierała wrote: > > As I said, the suspension is released when the ufssuspend file descriptor > > gets closed - which is what happens when the calling process exits. It's > > a protection mechanism, to avoid the situation where the process malfunction > > (eg a crash) would leave the system in unrecoverable (suspended) state. > > > > You probably want your process to just execute another one, and wait until > > it exits. > > > > The example with freeze -f strongly hints this is supposed to work as a > drop in replacement for linux scripts. > > As such, maybe ufs should grow another operation which does not > automagically unfreeze. I'm not sure it's a good idea to provide an inferior mechanism just for backward compatibility with Linux. Especially given how easy it is to do it properly, modeling the utility after eg lockf(1).
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20160629043203.GA82400>