Read Only

[Michael Scully]

Cameron:

	I wasn't aware that Outlook would embed the in-reply to stuff in
there.  Thanks for the tip.  Often I will reply to some old message merely
to get the email address quickly, but change the subject line. I can stop
that practice for the list.

	Your logic makes sense in terms of what may be happening.  But I'm
not sure what is different about this system of any others.  I have external
USB drives on probably a dozen RHEL systems: versions 3, 4, and 5, both
32bit and x86_64.  I set up identical hardware to this one 4 weeks ago, and
I have no problem there.  I've been using small Maxtor Mini's drives for
removable backups, instead of tape, as well as larger 500/750/1000 GB
external drives for longer-term archiving.  None of the other servers have
this read-only dilemma.  

Scully

-----Original Message-----
From: redhat-list-bounces at redhat.com [mailto:redhat-list-bounces at redhat.com]
On Behalf Of Cameron Simpson
Sent: Thursday, April 17, 2008 4:51 PM
To: General Red Hat Linux discussion list
Subject: Re: Read Only

On 17Apr2008 15:21, Michael Scully <agentscully at flexiblestrategies.com>
wrote:
| Greetings:

Greeting.

Please DO NOT start a new thread by replying to an old thread.
Your mailer has kept the In-Reply-To: header, and made your message
appear to be part of the "hijacked email" thread. This is both Very
Annoying and may also prevent your message being seen by people ignoring
the hijack thread. Indeed, what you just did is _called_ thread
hijacking, amusingly.

| 	I've been using external USB hard drives on RHEL for more than a
| year.  But this condition only happens on one system.  I make ext3
| filesystems on the devices (I have more than one here).  I can mount them
| fine and write to them fine.  But after a time period of inactivity, the
| drives spin down for power savings.  When they wake back up, the file
| systems become READ ONLY.  A simple umount and mount is all that's needed
to
| remedy this, but I can't always do that in some automated processes.

I haven't seen it, but I speculate that the OS tries to do an update to
the drive at some point when it is spun down, and the drive does not spin
out and accept things fast enough for the OS to consider it successful
i.e. the OS gives up before the drive is ready. This causes the OS to
consider it a write error, and common practice for various filesystems
is to make the filesystem read-only to prevent further damage.

If you look at the output of the "dmesg" command after this happens you
_should_ see some evidence of this happening.

Cheers,
-- 
Cameron Simpson <cs at zip.com.au> DoD#743
http://www.cskk.ezoshosting.com/cs/

And while its true that the meek shall inheret the earth, we will just push
them down and take it back again.       - A Whitney Brown