headers
FreeBSD-gnats-submit | 5 Dec 2005 22:10
Picon
Favicon

Re: kern/89991: softupdates with mount -ur causes fs UNREFS

Thank you very much for your problem report.
It has the internal identification `kern/89991'.
The individual assigned to look at your
report is: freebsd-bugs. 

You can access the state of your problem report at any time
via this link:

http://www.freebsd.org/cgi/query-pr.cgi?pr=89991

>Category:       kern
>Responsible:    freebsd-bugs
>Synopsis:       softupdates with mount -ur causes fs UNREFS
>Arrival-Date:   Mon Dec 05 21:10:02 GMT 2005
_______________________________________________
freebsd-fs <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-unsubscribe <at> freebsd.org"
Permalink | Reply | 0 comments |
headers
Craig Rodrigues | 13 Dec 2005 16:19

XFS (read-only) support committed to CURRENT

Hi,

Read-only XFS support has been committed to FreeBSD-CURRENT.
Write access to XFS is not supported at this time.
The XFS for FreeBSD source code is based off of GPL'd sources
provided by SGI.

You can compile it into your kernel by adding "XFS" to your
kernel config file, or you can "kldload xfs".

If you have an XFS partition on your system, you can try mounting it
with:
mount -t xfs [device] [mntpoint]

Additional utilities such as mkfs.xfs are available in the sysutils/xfsprogs
port. 

Many thanks to Alexander Kabaev and Russell Cattelan for
starting and doing most of the work on the XFS for FreeBSD port,
and also to SGI for making the XFS source code freely available.

--

-- 
Craig Rodrigues        
rodrigc <at> crodrigues.org
_______________________________________________
freebsd-current <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe <at> freebsd.org"
(Continue reading)

Permalink | Reply | 0 comments |
headers
t c | 15 Dec 2005 11:12
Picon

filesystem full - freebsd 5.3

Hello,

I've got the following error messages in dmesg.today, but there are lots
inodes (and free space) on that partition (/home):

pid 50371 (rateup), uid 0 inumber 1130885 on /home: filesystem full
pid 42486 (httpd), uid 80 inumber 1059960 on /home: filesystem full
pid 50614 (virtual), uid 1004 inumber 966735 on /home: filesystem full

(many times each row...)

root <at> ns:~# df -hi
Filesystem     Size    Used   Avail Capacity iused   ifree       %iused
Mounted on
/dev/ad0s1a    4.8G    1.2G    3.2G    27%  159048  500406     24%   /
devfs              1.0K    1.0K      0B   100%       0       0
100%   /dev
/dev/ad0s1e    22G     16G    6.0G    72%  114104 2829894        4%   /home
/dev/ad0s1d    9.7G    3.2G    5.7G    36%   14119 1304791        1%   /var
devfs              1.0K    1.0K      0B   100%       0       0
100%   /var/named/dev

root <at> ns:/etc# mount
/dev/ad0s1a on / (ufs, local)
devfs on /dev (devfs, local)
/dev/ad0s1e on /home (ufs, local, soft-updates)
/dev/ad0s1d on /var (ufs, local, soft-updates)
devfs on /var/named/dev (devfs, local)

I've found a similar problem in the archives - without solution....
(Continue reading)

Permalink | Reply | 1 comment |
headers
Oliver Fromme | 15 Dec 2005 14:40
Picon

Re: filesystem full - freebsd 5.3

t c <namondo <at> gmail.com> wrote:
 > I've got the following error messages in dmesg.today, but there are lots
 > inodes (and free space) on that partition (/home):
 > 
 > pid 50371 (rateup), uid 0 inumber 1130885 on /home: filesystem full
 > pid 42486 (httpd), uid 80 inumber 1059960 on /home: filesystem full
 > pid 50614 (virtual), uid 1004 inumber 966735 on /home: filesystem full
 > (many times each row...)

Maybe the file system was really full at the time those
problems were logged.  Then some cronjob (e.g. logrotate
or whatever) cleaned up, and now you don't see any traces
of the problem anymore.

If you want to find out, you could monitor your free space
continously.  The easiest way to do that would probably be
a small shell script which appends `df -k /home` to a log
file in /var.  You can call the script every 5 minutes via
cron, for example.

Usually, when there are messages reporting that the file
system is full, it really _is_ full at that time.
In theory there could be some inconsistencies or other
damage of the filesystem, but in that case you should also
get other error messages.  If you want to be sure, umount
the file system and fsck it.  I bet there will be no
errors.

Best regards
   Oliver
(Continue reading)

Permalink | Reply | 0 comments |
headers
dlm-fb | 15 Dec 2005 16:24

Re: filesystem full - freebsd 5.3


Oliver Fromme <olli <at> lurza.secnetix.de> wrote:
:  t c <namondo <at> gmail.com> wrote:
:   > I've got the following error messages in dmesg.today, but there are lots
:   > inodes (and free space) on that partition (/home):
:   > 
:   > pid 50371 (rateup), uid 0 inumber 1130885 on /home: filesystem full
:   > pid 42486 (httpd), uid 80 inumber 1059960 on /home: filesystem full
:   > pid 50614 (virtual), uid 1004 inumber 966735 on /home: filesystem full
:   > (many times each row...)
:  [...]
:  Usually, when there are messages reporting that the file
:  system is full, it really _is_ full at that time.
:  In theory there could be some inconsistencies or other
:  damage of the filesystem, but in that case you should also
:  get other error messages.  If you want to be sure, umount
:  the file system and fsck it.  I bet there will be no
:  errors.

Perhaps more likely, he was trying to allocate full-size blocks, and
the only things available were fragments.  The output from df doesn't
distinguish between the two types of available space.  You can use
dumpfs(8) to do that.

	Dworkin
_______________________________________________
freebsd-fs <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-unsubscribe <at> freebsd.org"
(Continue reading)

Permalink | Reply | 3 comments |
headers
Oliver Fromme | 15 Dec 2005 17:14
Picon

Re: filesystem full - freebsd 5.3

dlm-fb <at> weaselfish.com wrote:
 > Oliver Fromme <olli <at> lurza.secnetix.de> wrote:
 > :  t c <namondo <at> gmail.com> wrote:
 > :   > I've got the following error messages in dmesg.today, but there are lots
 > :   > inodes (and free space) on that partition (/home):
 > :   > 
 > :   > pid 50371 (rateup), uid 0 inumber 1130885 on /home: filesystem full
 > :   > pid 42486 (httpd), uid 80 inumber 1059960 on /home: filesystem full
 > :   > pid 50614 (virtual), uid 1004 inumber 966735 on /home: filesystem full
 > :   > (many times each row...)
 > :  [...]
 > :  Usually, when there are messages reporting that the file
 > :  system is full, it really _is_ full at that time.
 > :  In theory there could be some inconsistencies or other
 > :  damage of the filesystem, but in that case you should also
 > :  get other error messages.  If you want to be sure, umount
 > :  the file system and fsck it.  I bet there will be no
 > :  errors.
 > 
 > Perhaps more likely, he was trying to allocate full-size blocks, and
 > the only things available were fragments.

I considered mentioning the fragments issue, too, but I
don't think it applies in this case.  His file system is
22 Gbyte with 6 Gbyte free.  I think it is very unlikely
that those 6 Gbyte are all fragments.

Best regards
   Oliver
(Continue reading)

Permalink | Reply | 2 comments |
headers
Steven Hartland | 15 Dec 2005 17:28
Picon
Favicon

Re: filesystem full - freebsd 5.3

Out of inodes?

    Steve

----- Original Message ----- 
From: "Oliver Fromme" <olli <at> lurza.secnetix.de>
 ...
> I considered mentioning the fragments issue, too, but I
> don't think it applies in this case.  His file system is
> 22 Gbyte with 6 Gbyte free.  I think it is very unlikely
> that those 6 Gbyte are all fragments.

================================================
This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is
addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or
otherwise disseminating it or any information contained in it. 

In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137
or return the E.mail to postmaster <at> multiplay.co.uk.

_______________________________________________
freebsd-fs <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-unsubscribe <at> freebsd.org"
Permalink | Reply | 1 comment |
headers
Oliver Fromme | 15 Dec 2005 17:41
Picon

Re: filesystem full - freebsd 5.3

Steven Hartland <killing <at> multiplay.co.uk> wrote:
 > Out of inodes?

No.  See the first message in this thread.

Best regards
   Oliver

--

-- 
Oliver Fromme,  secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing
Dienstleistungen mit Schwerpunkt FreeBSD: http://www.secnetix.de/bsd
Any opinions expressed in this message may be personal to the author
and may not necessarily reflect the opinions of secnetix in any way.

"The scanf() function is a large and complex beast that often does
something almost but not quite entirely unlike what you desired."
        -- Chris Torek
_______________________________________________
freebsd-fs <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-unsubscribe <at> freebsd.org"
Permalink | Reply | 0 comments |
headers
dlm-fb | 15 Dec 2005 19:52

Re: filesystem full - freebsd 5.3


Oliver Fromme <olli <at> lurza.secnetix.de> wrote:
:  I considered mentioning the fragments issue, too, but I
:  don't think it applies in this case.  His file system is
:  22 Gbyte with 6 Gbyte free.  I think it is very unlikely
:  that those 6 Gbyte are all fragments.

I've had 100GB+ filesystems that were 50% full, and everything free
was an isolated fragment (or short run but under the size of a block).
Web caches, news, and similar small-file applications can be pretty
pessimal for ufs, so mentioning the possibility seemed worthwhile.

	Dworkin
_______________________________________________
freebsd-fs <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-fs
To unsubscribe, send any mail to "freebsd-fs-unsubscribe <at> freebsd.org"
Permalink | Reply | 0 comments |
headers
Matthias Andree | 16 Dec 2005 12:15
Picon
Picon

Re: XFS (read-only) support committed to CURRENT

Craig Rodrigues <rodrigc <at> crodrigues.org> writes:

> Read-only XFS support has been committed to FreeBSD-CURRENT.
> Write access to XFS is not supported at this time.
> The XFS for FreeBSD source code is based off of GPL'd sources
> provided by SGI.

Hm. Does this mean that FreeBSD's XFS implementation is GPL'd like
ext2fs is? If so, allow me a question why XFS was chosen in preference
to ext3fs?

Ext3fs appears to have some advantages, easy migration from and to
ext2fs, shrinkable, data journalling, data ordering (write data blocks
before the file metadata is written) and so on.

I don't mean this should become an advocacy discussion, as XFS surely
has advantages, too, real-time capability and so on - but ext2fs is
already there and has write support.

Just curious.

--

-- 
Matthias Andree
_______________________________________________
freebsd-current <at> freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe <at> freebsd.org"
Permalink | Reply | 0 comments |

Gmane