From: "Petr Vandrovec" <VANDROVE@vc.cvut.cz>
To: "Richard B. Johnson" <root@chaos.analogic.com>
Cc: linux-kernel@vger.kernel.org, ertzog@bk.ru
Subject: Re: hot IDE change
Date: Wed, 23 Jan 2002 22:14:49 +0100 [thread overview]
Message-ID: <FBFD7B7521F@vcnet.vc.cvut.cz> (raw)
On 23 Jan 02 at 15:56, Richard B. Johnson wrote:
> > This question is more about hardware, but is also related to Linux.
> > If I have a harddisk, plugged into the motherboard (IDE cable and power),
> > can I turn it off, plugging out first power cable, then IDE cable.
> > Can it harm harddisk or motherboard?
> > If I can do it, then will Linux detect it back, if I make this
> > operation back: i.e. plug IDE cable, then power cable.
>
> Linux doesn't care as long as it has been dismounted.
Do not forget to run 'sync' after you dismount it.
> So, you decide to pull the IDE cable first. Well, if you made
> sure that all active bits were disconnected at the same time,
> (grin), maybe you can get away with it.
My non-hotpluggable bay first disconnect power, and it works fine,
every friday evening for more than two years... But I have this
device alone on its cable, so there is no traffic when I disconnect it.
And I probably lost warranty by doing that.
> Now, do you want to plug in another drive? If the five-volts
> isn't present before the +12V, you may back-feed the 5-volt
> logic through its protection diodes. If this generates enough
Unless something changed behind my back, do not do it. As kernel
have no idea that you removed drive, it will not invalidate its
caches. Maybe you could workaround this (at least you have to ask
it to rescan partition table), but I decided to not plug IDE HDD into
machine when kernel runs. So I reboot my system on each Monday morning,
and I plug hdd in when BIOS checks memory. It resets your uptime to zero,
but as I upgrade kernel much more often, it is not a big problem.
Only problem I had were ATAPI devices. There is stupid paragraph about
creating `virtual' slave when none is present, in the ATAPI standard,
and so even if you plug HDD into the box as a slave to ATAPI device,
you'll not see it until you powercycle your ATAPI device... At least
my Chineese CDROM behaves that way, so I put it on its own cable.
Best regards,
Petr Vandrovec
vandrove@vc.cvut.cz
next reply other threads:[~2002-01-23 21:15 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-01-23 21:14 Petr Vandrovec [this message]
-- strict thread matches above, loose matches on Subject: below --
2002-01-23 23:05 hot IDE change ertzog
2002-01-23 20:12 ` Ed Sweetman
2002-01-23 20:56 ` Richard B. Johnson
2002-01-23 21:18 ` Matus Horvath
2002-01-23 21:25 ` John Heil
2002-01-23 22:07 ` Wakko Warner
2002-01-24 9:19 ` Allan Sandfeld
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=FBFD7B7521F@vcnet.vc.cvut.cz \
--to=vandrove@vc.cvut.cz \
--cc=ertzog@bk.ru \
--cc=linux-kernel@vger.kernel.org \
--cc=root@chaos.analogic.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox