headers
Bas | 24 Jun 2002 08:50
Picon

Fix for typo in jfsutils/mkfs/mkfs.c

Hi,

This patch fixes the 'Emergnecy help' typo fix in mkfs.c.

Bye,
Bas.

----- Original Message ----- 
From: "Steve Best" <sbest <at> us.ibm.com>
To: <jfs-discussion <at> oss.software.ibm.com>
Sent: Friday, June 21, 2002 8:22 PM
Subject: [Jfs-discussion] [ANNOUNCE] JFS release 1.0.20

> Release 1.0.20 of JFS was made available today.
> 
> Drop 58 on June 21, 2002 (jfs-2.4-1.0.20.tar.gz
> and jfsutils-1.0.20.tar.gz) includes fixes to the file
> system and utilities.
> 
> Utilities changes
> 
> - don't display heartbeat during log format if output is redirected
>   (eliminates strange characters in redirected output from mkfs.jfs
>   with external log, fsck.jfs Phase 9)
> - fix mkfs.jfs to set version in JFS superblock properly if external
>    log is used (enables JFS external log compatibility with EVMS 1.1-pre2
>    or greater)
> - enhance jfsutils to support enormous disks like 8TB+ (Peter C.)
> - remove unused variables (Christoph Hellwig)
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
SyvertsonJ | 24 Jun 2002 23:43

file system resize

Greetings,
  On a previous post in May I saw you were hoping to release the resize
utility by the end of the month of may.

I have not seen any information in the changelog in the last 3 releases.
Also, the 1.1.0-pre1,2,3 release of evms are supposed to have added the File
system interface Modules now.  A note was put int there that the filesystem
interface module for JFS was in there but I did not see the utility come out
for JFS yet.

Has EVMS been interfaced at this point now with JFS so expansins of volumes
will expand the filesystem for JFS?

Has the utility been snuck in without any changelog notice?

By the way, I have had no problems to date with evms 1.0.0 with JFS 1.0.17
on top of it to date (no lost data or kernel trouble) 2.4.18 kernel.  I'm
looking forward to the fs and evms integration.

Thank you for some excellent software . . . Jon Syvertson - HON INDUSTRIES
Permalink | Reply | 0 comments |
headers
Barry Arndt | 25 Jun 2002 17:09
Picon
Favicon

Re: file system resize


Jon,

The extend function is not quite complete.

After several discussions, the EVMS team decided that there were more
benefits to including the file system FSIMs in the EVMS package rather
than with the file systems' individual packages.  The JFS FSIM is included
in all of the 1.1.0-pre releases of EVMS (as are FSIMs for ext2/3,
reiserfs,
and swap).  The JFS FSIM does not yet support extend, but will when
the extend work is completed in the JFS file system.

You bring up a good point...  I will include JFS FSIM notices in the
ChangeLog and/or NEWS to keep JFS users informed of JFS
FSIM updates.

Thanks for your note, and for keeping us updated on your JFS usage.

Barry Arndt
barndt <at> us.ibm.com
IBM Linux Technology Center
JFS for Linux http://oss.software.ibm.com/jfs

SyvertsonJ <SyvertsonJ <at> honi.com> <at> www-124.southbury.usf.ibm.com on
06-24-2002 04:43:32 PM

Sent by:    jfs-discussion-admin <at> www-124.southbury.usf.ibm.com

To:    "'jfs-discussion <at> www-124.ibm.com'"
(Continue reading)

Permalink | Reply | 0 comments |
headers
Roy Sigurd Karlsbakk | 26 Jun 2002 11:55
Favicon
Gravatar

Fwd: Large Block Device testing -- preliminary results.

hi

thought you'd probably be interested in this

roy

----------  Forwarded Message  ----------

Subject: Large Block Device testing -- preliminary results.
Date: Wed, 26 Jun 2002 14:36:03 +1000
From: Peter Chubb <peter <at> chubb.wattle.id.au>
To: linux-kernel <at> vger.kernel.org

So far I've been able to test JFS on two megaraid controllers
software-raided together to give a 4-terabyte filesystem on a 32-bit
system; and a single 2.4TB filesystem on a 64-bit machine.

On the 32-bit system:

$ df -k
Filesystem           1k-blocks      Used Available Use% Mounted on
/dev/hda2              2403452   1991036    290324  88% /
/dev/hda1                46667     10566     33692  24% /boot
/dev/md0             4016755544  33794668 3982960876   1% /mnt

$ cat /proc/partitions
major minor  #blocks  name

   9     0 4016911104 md0
   8     0 1990639616 sda
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bernhard Mogens Ege | 27 Jun 2002 11:52
Picon
Picon

mpstat: free larger than alloc?

I was checking the /proc/fs/jfs files and noticed that mpstat had this
to say:

JFS Metapage statistics
=======================
metapages in use = 0
page allocations = 945055
page frees = 945056
lock waits = 29
allocation waits = 0

Is it safe for page frees to be larger than page allocations?

I know the server might have a hardware problem (is yet to be located
and confirmed). I also see these message in the logs, but am uncertain 
what they actually mean?

Jun 27 01:04:40 boston kernel: find_entry called with index = 0
Jun 27 04:02:10 boston kernel: find_entry called with index = 0

Dual P3 Xeon 700MHz, 512Mb RAM, Mylex DAC960 (5 disks, raid 5), jfs
1.0.20, kernel 2.4.18, loglevel=1.

Additional question: How do I change the loglevel?

regards,

Bernhard Ege
Permalink | Reply | 5 comments |
headers
Dave Kleikamp | 27 Jun 2002 14:26
Picon
Favicon

Re: mpstat: free larger than alloc?

> page allocations = 945055
> page frees = 945056
> 
> Is it safe for page frees to be larger than page allocations?

I don't think the keeping of these statistics is SMP-safe, so on an SMP
machine, these numbers aren't completely reliable.

> 
> I know the server might have a hardware problem (is yet to be located
> and confirmed). I also see these message in the logs, but am uncertain 
> what they actually mean?
> 
> Jun 27 01:04:40 boston kernel: find_entry called with index = 0
> Jun 27 04:02:10 boston kernel: find_entry called with index = 0

The directory contains a Btree containing the contents of the directory and
a table needed to map each directory entry with a persistent "cookie"
necessary to make readdirs work correctly as changes are made to the
directory.  That message indicates that some directory entries do not
have a valid cookie.

If you have any entries in lost+found, these are created without the cookie
table.  (We need to fix this.)  If that's not the case, I'm not aware of
any known problems resulting in this condition.
> 
> Dual P3 Xeon 700MHz, 512Mb RAM, Mylex DAC960 (5 disks, raid 5), jfs
> 1.0.20, kernel 2.4.18, loglevel=1.
> 
> Additional question: How do I change the loglevel?
(Continue reading)

Permalink | Reply | 4 comments |
headers
Timon Heinze | 1 Jul 2002 14:58
Picon

Cannot mount jfs-filesystem anymore

Hi,

After a clean reboot I cannot mount my jfs-filesystem anymore. 
Everything was working fine before the reboot and i changed nothing 
before that on the system.

The output of "mount /dev/hda3 /mnt/date -t jfs" is:

attempt to access beyond end of device
03:03: rw=0, want=118546560, limit=118546344
attempt to access beyond end of device
03:03: rw=1, want=118546560, limit=118546344
assert(bp->l_wqnext == NULL)
invalid operand: 0000
chani kernel: CPU:    0
EIP:    0010:[<c017f7be>]    Not tainted
EFLAGS: 00010086
eax: 0000001d   ebx: c1546a80   ecx: 00000001   edx: 00000001
esi: d3b10ac0   edi: d3b00000   ebp: 02035b70   esp: d3babdfc
ds: 0018   es: 0018   ss: 0018
Process mount (pid: 37, stackpage=d3bab000)
Stack: c021ebb9 c021ebef 00000206 c1546a80 c017f78e d3b10ac0 00000005 
c017f1b2
        d3b10ac0 00000000 c1546a80 c1546b34 d3ee0600 d3b10ac0 d3b10bc0 
00000000
        00000000 00004000 00000246 00000000 d3eb79c0 d3ee0600 d3ee0600 
c017edea
Call Trace: [<c017f78e>] [<c017f1b2>] [<c017edea>] [<c017ee81>] 
[<c0168e86>]
    [<c0166173>] [<c013222e>] [<c013263b>] [<c0141b81>] [<c0141e32>]
(Continue reading)

Permalink | Reply | 0 comments |

Gmane