Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 18 Dec 1999 13:47:21 +1030
From:      Greg Lehey <grog@lemis.com>
To:        Bruce DeVault <internal@intechsoft.net>
Cc:        Andy Dills <andy@xecu.net>, Andrew Heybey <ath@niksun.com>, freebsd-questions@FreeBSD.ORG
Subject:   Re: Maxtor 40GB HD
Message-ID:  <19991218134721.E1108@freebie.lemis.com>
In-Reply-To: <006401bf489c$66862a70$0c01a8c0@wkbruce>
References:  <Pine.GSO.4.10.9912162305320.25025-100000@shell.xecu.net> <006401bf489c$66862a70$0c01a8c0@wkbruce>

next in thread | previous in thread | raw e-mail | index | archive | help
On Friday, 17 December 1999 at  8:38:35 -0600, Bruce DeVault wrote:
>> On 16 Dec 1999, Andrew Heybey wrote:
>>
>>>> This is backed up by somebody who emailed me stating that with
>>>> their 27.5 GB drive, they had to leave off the last couple of
>>>> sectors to get it to work.
>>>>
>>>> So, is this a bug, a kernel config, or anything you recognize?
>>>> And, anything I can do to fix it?
>>>
>>> First, it is not a fundamental bug.  Proof by counter-example:
>>>
>>> Filesystem  1K-blocks     Used    Avail Capacity  Mounted on
>>> /dev/ccd0c   66984252  4803840 56821672     8%    /netvcr
>>
>> Yes, I'm not questioniong the limit of fs sizes, but that's a concat'ed
>> fs you have there. I'm talking about a single fs on a single 40GB disk.
>>
>>> and there are others with much larger file systems than this.
>>
>> In GB or sectors?
>>
>>> Second, you did not provide any useful information to help anyone
>>> figure this out such as a) what version of FreeBSD you are using, b)
>>> exactly what arguments you are giving to newfs, c) what error messages
>>> newfs prints when it fails, or d) what your disk label is.
>>
>> Version is 3.3, standard newfs args from sysinstall (-b 2192 -something
>> 1024), the error message is a simple write error. What info from the
>> disklabel would you like?
>
> This is almost verbatim the problem I had with IBM Deskstar 34GB drives. I
> could newfs to almost the size of the drive with no errors, but using the
> whole drive yielded a newfs error. I posted about this problem several
> times, perhaps three or four weeks ago. Although people were generally
> helpful, and several were quite helpful, there's been no acknowledgement
> that there's any kind of problem with the way things work.
>
> In the end, I had to use less than the full size of the drives, and I ended
> up striping two about 30GB filesystems with vinum.
>
> I'm looking forward to seeing an actual resolution to this problem, because
> I'd like to be able to reinstall the thing correctly when a fix is
> available, but I can't afford to take the thing down to play with it anymore
> just to try things at semi-random.

I've seen a number of messages about this problem.  As far as I can
see, it's a bug in the IDE driver.  It doesn't happen with Vinum or
ccd, and I'm pretty sure it doesn't happen with SCSI drives, though I
can't point to a proof right now.

We're in the process of changing to a completely new IDE driver,
called ata.  The old driver is wd.  4.0-RELEASE will include the ata
driver as standard, so it's a good time to test it now.  If anybody's
adventurous, please let me know.

Greg
--
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991218134721.E1108>