headers
Tim Shimmin | 1 Jun 2007 05:32
Picon
Favicon

PARTIAL TAKE 965537 - xfsprogs AC_TYPE_U32 broken by defining HAVE___U32_T instead of HAVE___U32

s/HAVE___U32_T/HAVE___U32/g
Thanks to Eric Sandeen for finding this.

Date:  Fri Jun  1 13:31:17 AEST 2007
Workarea:  chook.melbourne.sgi.com:/build/tes/xfs-cmds
Inspected by:  sandeen <at> sandeen.net

The following file(s) were checked into:
  longdrop.melbourne.sgi.com:/isms/xfs-cmds/master-melb

Modid:  master-melb:xfs-cmds:28754a
xfsprogs/doc/CHANGES - 1.240 - changed
http://oss.sgi.com/cgi-bin/cvsweb.cgi/xfs-cmds/xfsprogs/doc/CHANGES.diff?r1=text&tr1=1.240&r2=text&tr2=1.239&f=h
xfsprogs/aclocal.m4 - 1.26 - changed
http://oss.sgi.com/cgi-bin/cvsweb.cgi/xfs-cmds/xfsprogs/aclocal.m4.diff?r1=text&tr1=1.26&r2=text&tr2=1.25&f=h
xfsprogs/m4/package_types.m4 - 1.4 - changed
http://oss.sgi.com/cgi-bin/cvsweb.cgi/xfs-cmds/xfsprogs/m4/package_types.m4.diff?r1=text&tr1=1.4&r2=text&tr2=1.3&f=h
	- s/HAVE___U32_T/HAVE___U32/g
	  Thanks to Eric Sandeen for finding this.
Permalink | Reply | 0 comments |
headers
Timothy Shimmin | 1 Jun 2007 06:39
Picon
Favicon

Re: [patch 1/3] Fix XFS_IOC_FSGEOMETRY_V1 in compat mode


--On 31 May 2007 11:26:15 PM +1000 David Chinner <dgc <at> sgi.com> wrote:
>> Who would want to use XFS_IOC_FSGEOMETRY_V1?
>>
>> Okay it turns out a whole bunch of our xfs-cmds :-)
>> (Such as xfsdump as Michal mentioned)
>> On Sep/2002, Nathan changed a bunch of them to use v1.
>>  xfsprogs-2.3.0 (03 September 2002)
>>        - Several changes to geometry ioctl callers which will make
>>          the tools useable on older kernel versions too.
>> So he did this so that new tools would work on the older kernels which
>> didn't support the new geom version.
>> So I guess we are stuck with v1 now.
>
> Not necessarily - we could change the tools to use v4, and if that
> didn't exist, then try v1. That way we don't need to support v1 in
> linux, and the tools still run on old kernels.....
>
The problem with that is the old tools won't run on new kernels.
If you get a new kernel and use an old xfsdump then you are out of luck.
Not sure if we want to require people to bump up to new userspace for this.

--Tim
Permalink | Reply | 1 comment |
headers
Ruben Porras | 1 Jun 2007 18:39
Picon

XFS shrink functionality

Hello, 

I'm investigating the possibility to write myself the necessary code to
shrink an xfs filesystem (I'd be able to dedicate a day/week). Trying to
know if something is already done I came across the mails of a previous
intent [0], [1] (I'm cc'ing the people involved). 

At a first glance the patch is a little outdated and will no more apply
(as of linux 2.16.18, which is the last customised kernel that I was
able to run under a XEN environment), because at least the function
xfs_fs_geometry is changed. 

I'm really curious about what happened to this patches and why they were
discontinued. The second part never was made public, and there was also
no answer. Was there any flaw in any of the posted code or anything in
XFS that makes it especially hard to shrink [3] that discouraged the
development?

After that, the first questions that arouse are,
would there be some assistance/groove in from the developers? 
How doable is it?
What are the programmers requirements from your point of view?

Thank you.

[1] http://oss.sgi.com/archives/xfs/2005-08/msg00142.html
[2] http://oss.sgi.com/archives/xfs/2005-09/msg00038.html
[3] the only limitation that I might think of is not being able to
shrink past the internal journal.
(Continue reading)

Permalink | Reply | 20 comments |
headers
Justin Piszcz | 2 Jun 2007 23:24

Kernel 2.6.22-rc3 safe to migrate to?

Wondering as their were a lot of XFS related issues early on in 
development..?  The 2.6.22-rc3 kernel has the core 2 duo coretemp patch by 
ruik which I want be running as long as 2.6.22-rc3 does not have any 
severe XFS issues?

Justin.
Permalink | Reply | 0 comments |
headers
Raz Ben-Jehuda(caro | 2 Jun 2007 23:07
Picon

corruption bug in 2.6.17

Mandy Hello

I have got into a serious trouble with xfs . I had some directories
returning "no such file or directory". This xfs file system is over a
raid5 of 4 disks, with 1MB stripe unit size. files are hierarchly
being "XFS_IOC_FSSETXATTR" extended to 1MB.

when I tried to create a new file I got the bellow oops.
could it be that your fix for 2.6.17.7 solves this problem ?
thank you.
raz

4499008.609000] Filesystem "md1": XFS internal error
xfs_dir2_block_addname at line 105 of file fs/xfs/xfs_dir2_block.c.
Caller 0xc10e7e1c
[4499008.634000]  <c10e94c3> xfs_dir2_block_addname+0x88b/0x8a9
<c10e7e1c> xfs_dir2_createname+0x11a/0x179
[4499008.654000]  <c10e7e1c> xfs_dir2_createname+0x11a/0x179
<c10d737e> xfs_bmap_last_offset+0xce/0x128
[4499008.673000]  <c10e889e> xfs_dir2_isblock+0x32/0x82  <c10e7e1c>
xfs_dir2_createname+0x11a/0x179
[4499008.690000]  <c112662c> kmem_zone_alloc+0x57/0xc5  <c111a348>
xfs_trans_ijoin+0x35/0x7f
[4499008.707000]  <c112117d> xfs_create+0x45b/0x7aa  <c112dd89>
xfs_vn_mknod+0x2f9/0x37e
[4499008.725000]  <c112dec4> xfs_vn_lookup+0x79/0x96  <c1077acc>
lookup_mnt+0x32/0x5c
[4499008.741000]  <c1068ee2> do_lookup+0x50/0xa8  <c107300c> dput+0x23/0x183
[4499008.755000]  <c1068f8a> __link_path_walk+0x50/0xe17  <c1077acc>
lookup_mnt+0x32/0x5c
(Continue reading)

Permalink | Reply | 2 comments |
headers
Justin Piszcz | 2 Jun 2007 23:39

Re: corruption bug in 2.6.17

Raz, 2.6.17-2.6.17.6 had the nasty corruption bug you're seeing, best to 
restore from backup and/or check the FAQ and get off that kernel range 
asap.

On Sun, 3 Jun 2007, Raz Ben-Jehuda(caro) wrote:

> Mandy Hello
>
> I have got into a serious trouble with xfs . I had some directories
> returning "no such file or directory". This xfs file system is over a
> raid5 of 4 disks, with 1MB stripe unit size. files are hierarchly
> being "XFS_IOC_FSSETXATTR" extended to 1MB.
>
> when I tried to create a new file I got the bellow oops.
> could it be that your fix for 2.6.17.7 solves this problem ?
> thank you.
> raz
>
> 4499008.609000] Filesystem "md1": XFS internal error
> xfs_dir2_block_addname at line 105 of file fs/xfs/xfs_dir2_block.c.
> Caller 0xc10e7e1c
> [4499008.634000]  <c10e94c3> xfs_dir2_block_addname+0x88b/0x8a9
> <c10e7e1c> xfs_dir2_createname+0x11a/0x179
> [4499008.654000]  <c10e7e1c> xfs_dir2_createname+0x11a/0x179
> <c10d737e> xfs_bmap_last_offset+0xce/0x128
> [4499008.673000]  <c10e889e> xfs_dir2_isblock+0x32/0x82  <c10e7e1c>
> xfs_dir2_createname+0x11a/0x179
> [4499008.690000]  <c112662c> kmem_zone_alloc+0x57/0xc5  <c111a348>
> xfs_trans_ijoin+0x35/0x7f
> [4499008.707000]  <c112117d> xfs_create+0x45b/0x7aa  <c112dd89>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Christian Kujau | 2 Jun 2007 23:59
Picon

Re: Kernel 2.6.22-rc3 safe to migrate to?

On Sat, 2 Jun 2007, Justin Piszcz wrote:
> Wondering as their were a lot of XFS related issues early on in 
> development..?  The 2.6.22-rc3 kernel has the core 2 duo coretemp patch by 
> ruik which I want be running as long as 2.6.22-rc3 does not have any severe 
> XFS issues?

Tracking -rc and running 2.6.22-rc3 for a couple of days now, no 
problems with xfs since this infamous 2.6.17 thingy[0]. But that's 
pretty useless information I guess, since your setup will be different 
from mine. Did you have anything "special" in mind when asking this?

C.

[0] http://oss.sgi.com/projects/xfs/faq.html#dir2
--

-- 
BOFH excuse #339:

manager in the cable duct
Permalink | Reply | 0 comments |
headers
Justin Piszcz | 3 Jun 2007 00:09

Re: Kernel 2.6.22-rc3 safe to migrate to?


On Sat, 2 Jun 2007, Christian Kujau wrote:

> On Sat, 2 Jun 2007, Justin Piszcz wrote:
>> Wondering as their were a lot of XFS related issues early on in 
>> development..?  The 2.6.22-rc3 kernel has the core 2 duo coretemp patch by 
>> ruik which I want be running as long as 2.6.22-rc3 does not have any severe 
>> XFS issues?
>
> Tracking -rc and running 2.6.22-rc3 for a couple of days now, no problems 
> with xfs since this infamous 2.6.17 thingy[0]. But that's pretty useless 
> information I guess, since your setup will be different from mine. Did you 
> have anything "special" in mind when asking this?
>
> C.
>
> [0] http://oss.sgi.com/projects/xfs/faq.html#dir2
> -- 
> BOFH excuse #339:
>
> manager in the cable duct
>

Thanks, I'll give it a go then--!
Permalink | Reply | 0 comments |
headers
Justin Piszcz | 3 Jun 2007 00:27

Re: Kernel 2.6.22-rc3 safe to migrate to?

Jun  2 18:23:23 p34 upsd[2225]: Can't connect to UPS [belkin] 
(newhidups-auto): No such file or directory
Jun  2 18:24:23 p34 upsmon[2228]: Poll UPS [belkin <at> localhost] failed - 
Driver not connected

Hmm, something changes with USB in 2.6.22-rc3 form 2.6.21..

On Sat, 2 Jun 2007, Christian Kujau wrote:

> On Sat, 2 Jun 2007, Justin Piszcz wrote:
>> Wondering as their were a lot of XFS related issues early on in 
>> development..?  The 2.6.22-rc3 kernel has the core 2 duo coretemp patch by 
>> ruik which I want be running as long as 2.6.22-rc3 does not have any severe 
>> XFS issues?
>
> Tracking -rc and running 2.6.22-rc3 for a couple of days now, no problems 
> with xfs since this infamous 2.6.17 thingy[0]. But that's pretty useless 
> information I guess, since your setup will be different from mine. Did you 
> have anything "special" in mind when asking this?
>
> C.
>
> [0] http://oss.sgi.com/projects/xfs/faq.html#dir2
> -- 
> BOFH excuse #339:
>
> manager in the cable duct
>
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Christian Kujau | 3 Jun 2007 00:04
Picon

Re: corruption bug in 2.6.17

On Sun, 3 Jun 2007, Raz Ben-Jehuda(caro) wrote:
> when I tried to create a new file I got the bellow oops.
> could it be that your fix for 2.6.17.7 solves this problem ?

If you're running a 2.6.17 kernel (< 2.6.17.7) when you're getting these 
errors, I guess it's most likely this particular 2.6.17-bug. Please read 
http://oss.sgi.com/projects/xfs/faq.html#dir2 very carefully and upgrade 
xfsprogs and your kernel.

C.
--

-- 
BOFH excuse #55:

Plumber mistook routing panel for decorative wall fixture
Permalink | Reply | 0 comments |

Gmane