Date: Tue, 15 May 2007 13:36:54 +0200 From: Eike Bernhardt <eike@unorganized.net> To: freebsd-geom@freebsd.org Subject: Re: Growing a graid3? Message-ID: <46499B56.3020000@unorganized.net> In-Reply-To: <20070513032317.GA15556@garage.freebsd.pl> References: <46442175.60508@unorganized.net> <20070511084018.GA38860@garage.freebsd.pl> <E0243D68-8D97-4F1E-A055-D20AC6F7A5B5@unorganized.net> <20070511112335.698620bf@whirpool.lse.ac.uk> <D4487729-69D7-4909-AA5F-794F837169FB@unorganized.net> <464495F1.4080108@unorganized.net> <20070513032317.GA15556@garage.freebsd.pl>
next in thread | previous in thread | raw e-mail | index | archive | help
On 13.05.2007 5:23 Uhr, Pawel Jakub Dawidek wrote:
> On Fri, May 11, 2007 at 06:12:33PM +0200, Eike Bernhardt wrote:
>> On 11.05.2007 14:14 Uhr, Eike Bernhardt wrote:
>>> On 11.05.2007, at 12:23, vladimir konrad wrote:
>>>
>>>>> growfs: we are not growing (8191->8191)
>>>> is there a disk label on the raid?
>>> I haven't added one explicitly, just did a
>>>
>>> graid3 label test md{0,1,2}
>>> newfs /dev/raid3/test
>>>
>>> as stated in Pawel's example and the one in graid3(8)
>> I just testet this with md devices and gmirror, without adding a disk
>> label -- there growfs works as expected .. ie. resizes the filesystem
>> successfully.
>
> My guess is that growfs(8) doesn't handle sector different that 512
> properly... (graid3 gives (<number of components>-1)*512 sectors)
If I understand the growfs code correctly (and I'm not shure I do), then
it tries to adjust the device size in a superblock copy, and exits
since it believes that the filesystem is not growing (line 2141ff).
I'll try to get some more information, but I'm not really used to debug
filesystem code ...
Schoenen Gruss!
--
Eike Bernhardt http://unorganized.net/ ICQ:
11256658
"Do not meddle in the affairs of SysAdmins,
for you are crunchy and taste good with ketchup."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?46499B56.3020000>