I seem to have hit an odd limit, that I didn't think existed. I have a 250G
WD IDE hard drive that I have just installed. Since I couldn't put a Ext3
filesystem on it (mount wouldn't recognize it) I decided to put a ReiserFS
filesystem on it. Since I have done that I have added 128G of data to the
drive. Now when I attempt to copy more data to it I get an error that there
is no more space on the drive.
I can touch a 0 byte file and delete it, but as soon as I attempt to move
anything over there with any size it errors out. df shows me as having 112G
free on that drive so I am a little confused as to what is giving me this
error. Is it the kernel that is not letting me write to the rest of the
drive or reiserfs or something completely different? Any help would be
welcome. Thanks.
Brent Norris
Assistant DTC, Edmonson County Schools
Ps. Please CC me on replies.
Norris, Brent wrote:
> I seem to have hit an odd limit, that I didn't think existed. I have a 250G
> WD IDE hard drive that I have just installed. Since I couldn't put a Ext3
> filesystem on it (mount wouldn't recognize it) I decided to put a ReiserFS
> filesystem on it. Since I have done that I have added 128G of data to the
> drive. Now when I attempt to copy more data to it I get an error that there
> is no more space on the drive.
Reiser can definitly do larger file systems than that (I have a Reiser
file system with over 300GB on).
Its worth trying a df -i to make sure you haven't run out of inodes -
but then you say you can create empty files.
Dave
I have about 2 filesystems currently at 600Gig.
Reiser has no inodes so I guarantee your out but it's irrelevant.
What does df -k actually show? You sure you don't have a process with
an open log file that's been removed?
Thus spake Dave Gilbert (Home) ([email protected]):
> Norris, Brent wrote:
> >I seem to have hit an odd limit, that I didn't think existed. I have a
> >250G
> >WD IDE hard drive that I have just installed. Since I couldn't put a Ext3
> >filesystem on it (mount wouldn't recognize it) I decided to put a ReiserFS
> >filesystem on it. Since I have done that I have added 128G of data to the
> >drive. Now when I attempt to copy more data to it I get an error that
> >there
> >is no more space on the drive.
>
> Reiser can definitly do larger file systems than that (I have a Reiser
> file system with over 300GB on).
>
> Its worth trying a df -i to make sure you haven't run out of inodes -
> but then you say you can create empty files.
>
> Dave
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
:wq!
---------------------------------------------------------------------------
Robert L. Harris | GPG Key ID: E344DA3B
@ x-hkp://pgp.mit.edu
DISCLAIMER:
These are MY OPINIONS ALONE. I speak for no-one else.
Life is not a destination, it's a journey.
Microsoft produces 15 car pileups on the highway.
Don't stop traffic to stand and gawk at the tragedy.
> Reiser has no inodes so I guarantee your out but it's
> irrelevant. What does df -k actually show? You sure you
> don't have a process with an open log file that's been removed?
No I am pretty sure that I don't have an open log files. This drive holds
mainly movies and songs. I still got the error after a reboot, so I don't
think it is possible that there is anything open on it, especially anything
taking up 112G. It is just a desktop machine so, that would be a real
rarity on it. Is it possible that there is some error in the file
structure? The filesystem is less than 12 hours old, but in that 12 hours
there have been a lot of writes to it (namely the whole 128G) so is it
possible something has been corrupted?
Like I said it was a real odd error to get, but this is the first time I
have worked with drives this big, so I didn't know if this was something
everyone knew about.
Brent
Its an IDE issue. You need 48-bit addressing to access above 128GB on
IDE drives. I'm not sure exactly what the IDE guys need from you...
Post the output of dmesg, lspci, .config.
Testing the latest kernel would be an idea, if you are not running it.
Robert L. Harris wrote:
>
>I have about 2 filesystems currently at 600Gig.
>
>Reiser has no inodes so I guarantee your out but it's irrelevant.
>What does df -k actually show? You sure you don't have a process with
>an open log file that's been removed?
>
>Thus spake Dave Gilbert (Home) ([email protected]):
>
>
>>Norris, Brent wrote:
>>
>>>I seem to have hit an odd limit, that I didn't think existed. I have a
>>>250G
>>>WD IDE hard drive that I have just installed. Since I couldn't put a Ext3
>>>filesystem on it (mount wouldn't recognize it) I decided to put a ReiserFS
>>>filesystem on it. Since I have done that I have added 128G of data to the
>>>drive. Now when I attempt to copy more data to it I get an error that
>>>there
>>>is no more space on the drive.
>>>
>>Reiser can definitly do larger file systems than that (I have a Reiser
>>file system with over 300GB on).
>>
>>Its worth trying a df -i to make sure you haven't run out of inodes -
>>but then you say you can create empty files.
>>
>>Dave
>>
>>-
>>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>the body of a message to [email protected]
>>More majordomo info at http://vger.kernel.org/majordomo-info.html
>>Please read the FAQ at http://www.tux.org/lkml/
>>
>
>:wq!
>---------------------------------------------------------------------------
>Robert L. Harris | GPG Key ID: E344DA3B
> @ x-hkp://pgp.mit.edu
>DISCLAIMER:
> These are MY OPINIONS ALONE. I speak for no-one else.
>
>Life is not a destination, it's a journey.
> Microsoft produces 15 car pileups on the highway.
> Don't stop traffic to stand and gawk at the tragedy.
>
>
>
> What version of the linux kernel are you using?
> Ie. provide some more info like what's 'uname -a'
> say on your system. There's been a few patches go
> by this list to fix this disk size problem. They
> should also be in the latest kernel releases too.
Ok, well that is a start then. I am running Redhat's latest kernel
2.4.20-20.9 though I also have 2.4.0test4 on the machine. I haven't tried
it under 2.4.0test4 though. I just didn't have enough time this morning
when the error surfaced. Are they in 2.4.22? Because looking through the
changelog, I see a bunch of AC's and other fixes for IDE items, but I don't
see anything about this boundrey? Did it go in before then?
Thanks for the help
Brent
Norris, Brent wrote:
>I seem to have hit an odd limit, that I didn't think existed. I have a 250G
>WD IDE hard drive that I have just installed. Since I couldn't put a Ext3
>filesystem on it (mount wouldn't recognize it) I decided to put a ReiserFS
>filesystem on it. Since I have done that I have added 128G of data to the
>drive. Now when I attempt to copy more data to it I get an error that there
>is no more space on the drive.
>
>I can touch a 0 byte file and delete it, but as soon as I attempt to move
>anything over there with any size it errors out. df shows me as having 112G
>free on that drive so I am a little confused as to what is giving me this
>error. Is it the kernel that is not letting me write to the rest of the
>drive or reiserfs or something completely different? Any help would be
>welcome. Thanks.
>
>
This probably has to do with the 128GB IDE disk boundary problems that
have been recently coming about and nothing to do with Reiserfs (except
that Reiserfs gets stuck with it). What version of the linux kernel are
you using? Ie. provide some more info like what's 'uname -a' say on your
system. There's been a few patches go by this list to fix this disk size
problem. They should also be in the latest kernel releases too.
IDE limit is 137GB, not 128GB.
2.6.0-test4 should be okay, unless you have Promise PDC20265,
then you should use 2.6.0-test5 (allows LBA48 DMA on PDC20265).
Latest 2.4-ac may be even better if you have i.e. ALi IDE controller.
It will use DMA transfers for < 137GB area and PIO for > 137GB,
whereas 2.6 will always use PIO.
--bartlomiej
On Thursday 25 of September 2003 16:03, Norris, Brent wrote:
> > What version of the linux kernel are you using?
> > Ie. provide some more info like what's 'uname -a'
> > say on your system. There's been a few patches go
> > by this list to fix this disk size problem. They
> > should also be in the latest kernel releases too.
>
> Ok, well that is a start then. I am running Redhat's latest kernel
> 2.4.20-20.9 though I also have 2.4.0test4 on the machine. I haven't tried
> it under 2.4.0test4 though. I just didn't have enough time this morning
> when the error surfaced. Are they in 2.4.22? Because looking through the
> changelog, I see a bunch of AC's and other fixes for IDE items, but I don't
> see anything about this boundrey? Did it go in before then?
>
> Thanks for the help
>
> Brent
On Thu, Sep 25, 2003 at 04:44:22PM +0200, Bartlomiej Zolnierkiewicz wrote:
> IDE limit is 137GB, not 128GB.
It's the same limit:
erik@zurix:~ >bc -l
bc 1.06
Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type warranty'.
128 * 1024 * 1024 * 1024 / 1000000000
137.43895347200000000000
Regards,
Erik
--
+-- Erik Mouw -- http://www.harddisk-recovery.com -- +31 70 370 12 90 --
| Lab address: Delftechpark 26, 2628 XH, Delft, The Netherlands
| Data lost?!
| Stay calm and contact Harddisk-recovery.com!
On Thursday 25 of September 2003 17:28, Erik Mouw wrote:
> On Thu, Sep 25, 2003 at 04:44:22PM +0200, Bartlomiej Zolnierkiewicz wrote:
> > IDE limit is 137GB, not 128GB.
>
> It's the same limit:
>
> erik@zurix:~ >bc -l
> bc 1.06
> Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc.
> This is free software with ABSOLUTELY NO WARRANTY.
> For details type warranty'.
> 128 * 1024 * 1024 * 1024 / 1000000000
> 137.43895347200000000000
No, it's not the same... 137GB == 128GiB ;-).
> Regards,
> Erik
On Sep 25, 2003 08:08 -0500, Norris, Brent wrote:
> I seem to have hit an odd limit, that I didn't think existed. I have a 250G
> WD IDE hard drive that I have just installed. Since I couldn't put a Ext3
> filesystem on it (mount wouldn't recognize it) I decided to put a ReiserFS
> filesystem on it.
Just FYI, we have lots of ext3 filesystems that are 2TB in size, so I don't
think it is an ext3 problem. What could be happening though is that when
you mke2fs the filesystem with your IDE problem it wraps writes over 128GB
back to zero and overwrites the superblock so mount doesn't see the ext3
superblock anymore.
Cheers, Andreas
--
Andreas Dilger
http://sourceforge.net/projects/ext2resize/
http://www-mddsp.enel.ucalgary.ca/People/adilger/
> Just FYI, we have lots of ext3 filesystems that are 2TB in
> size, so I don't think it is an ext3 problem. What could be
> happening though is that when you mke2fs the filesystem with
> your IDE problem it wraps writes over 128GB back to zero and
> overwrites the superblock so mount doesn't see the ext3
> superblock anymore.
Great!!! Nice info. That goes a long way toward explaining why it wouldn't
mount. It was so weird since mkfs.ext3 didn't error or anything, it just
finished normally, but then nothing would mount the drive. Thanks again.
Brent
> 2.6.0-test4 should be okay, unless you have Promise PDC20265,
> then you should use 2.6.0-test5 (allows LBA48 DMA on PDC20265).
Well I do have a PDC20265, so I installed the 2.6.0test5 rpm from
http://people.redhat.com/arjanv/2.5/RPMS.kernel/
That allowed me to get access to the whole drive. I made an ext3 filesystem
on it and everything mounted real nice. Unfortunately, now after about a
minute of hard copying the IDE channels timeout and reset. I get the
timeout on hdg (which is the 250G drive), but the filesystem on hdf (which
is a 120G drive that I am copying from) get corrupted. I didn't have this
problem with any of the other kernels, so I think it might have something to
do with what went in for test5. I am still looking at that though. I am
also going to try moving the drive that is currently hdf onto the standard
motherboard channels so that it isn't on that controller. I hope to get the
source rpm for 2.6.0test5 and the .config for it tonight and try recompiling
with some different options. Does any one have any suggestions on what I
should look for?
Thanks for all you help you guys have been really great.
Brent