Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 20 Aug 2020 18:23:35 +0100
From:      RW <rwmaillists@googlemail.com>
To:        freebsd-questions@freebsd.org
Subject:   Re: Correct way to disconnect/eject an USB hard disk ?
Message-ID:  <20200820182335.4f2048c6@gumby.homeunix.com>
In-Reply-To: <20200820044352.678a588e.freebsd@edvax.de>
References:  <C6B11926-65BE-4A05-B421-9F3E9C9AACF6@kukulies.org> <20200819142952.3686baa8.freebsd@edvax.de> <DB8PR06MB64427319E175193D1B828135F65D0@DB8PR06MB6442.eurprd06.prod.outlook.com> <b9b46ecb-b1ef-8f67-7b41-ae3cf0d3864b@dreamchaser.org> <20200820044352.678a588e.freebsd@edvax.de>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
On Thu, 20 Aug 2020 04:43:52 +0200
Polytropon wrote:

> The USB controller accepts and confirms the last
> chunk of data, reports back "Got it, everything is fine."
> Now the umount command does what it says on the can, it
> unmounts the medium, and returns control to the user.
> Now _internally_ the USB stick is still busy _actually_
> writing the last chunk of data to its storage, ...
 
> The idea of "just give it a few seconds after you physically
> remove the USB stick" relies on the assumption that such
> kind of activity takes place outside the control of the OS,
> and the OS has no real way of discovering what's really
> happening at that point.

I'm wondering how "now safe to remove" notifications work in windows
and other operating systems. And is shutting down or suspending to
disk just guesswork?

I don't know, but I'd be surprised if there aren't mechanisms  in sata,
usb etc, to ensure data has been committed to non-volatile storage.



Want to link to this message? Use this URL: <http://docs.FreeBSD.org/cgi/mid.cgi?20200820182335.4f2048c6>