xfs
[Top] [All Lists]

RE: Problems with xfs_grow on large LVM + XFS filesystem 20TB size check

To: Eric Sandeen <sandeen@xxxxxxxxxxx>
Subject: RE: Problems with xfs_grow on large LVM + XFS filesystem 20TB size check 2 failed
From: Lance Reed <lreed@xxxxxxxxxxxxxx>
Date: Wed, 30 Apr 2008 17:53:31 -0400
Accept-language: en-US
Acceptlanguage: en-US
Cc: "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
In-reply-to: <481752F9.8040600@xxxxxxxxxxx>
References: <6A32BC807C106440B7E23208F280DDAF01D21F36FD@xxxxxxxxxxxxxxxxxxxxx> <481650F5.40205@xxxxxxxxxxx> <6A32BC807C106440B7E23208F280DDAF01D21F3718@xxxxxxxxxxxxxxxxxxxxx> <481656F6.5030300@xxxxxxxxxxx> <48166E18.10008@xxxxxxx> <48166F42.50104@xxxxxxxxxxx> <6A32BC807C106440B7E23208F280DDAF01D21F384B@xxxxxxxxxxxxxxxxxxxxx> <48175046.5050405@xxxxxxxxxxx> <6A32BC807C106440B7E23208F280DDAF01D21F3870@xxxxxxxxxxxxxxxxxxxxx> <481752F9.8040600@xxxxxxxxxxx>
Sender: xfs-bounce@xxxxxxxxxxx
Thread-index: AciqGbqXQyhDIKWIQNisOqyX9VoPsAA8hsLg
Thread-topic: Problems with xfs_grow on large LVM + XFS filesystem 20TB size check 2 failed
Great!
That is exactly what I needed to know.

One follow up question:

Can I assume that the bug:
TAKE 959978 - growing an XFS filesystem by more than 2TB is broken
is a problem only with the the xfs_growfs code?  The reason I asked is that 
when I first made the original filesystem, I created it using mkfs.xfs and it 
succeeded fine for 10.5 TB.

# mkfs.xfs /dev/VolGroupNAS200/LogVolNAS200
meta-data=/dev/VolGroupNAS200/LogVolNAS200 isize=256    agcount=32, 
agsize=83886080 blks
         =                       sectsz=512   attr=0
data     =                       bsize=4096   blocks=2684354560, imaxpct=25
         =                       sunit=0      swidth=0 blks, unwritten=1
naming   =version 2              bsize=4096
log      =internal log           bsize=4096   blocks=32768, version=1
         =                       sectsz=512   sunit=0 blks, lazy-count=0
realtime =none                   extsz=4096   blocks=0, rtextents=

so I am assuming the rest of the XFS setup can handle large Filesystems fine.  
I am just trying to confirm that the probem is TAKE 959978 and that doing it in 
less than 2 TB increments should be fine.

Thanks again to the List for all the assitance.

Lance

-----Original Message-----
From: Eric Sandeen [mailto:sandeen@xxxxxxxxxxx]
Sent: Tuesday, April 29, 2008 12:55 PM
To: Lance Reed
Cc: markgw@xxxxxxx; xfs@xxxxxxxxxxx
Subject: Re: Problems with xfs_grow on large LVM + XFS filesystem 20TB size 
check 2 failed

Lance Reed wrote:
> Thanks,
>
> Sorry, I am a bit confused on the "data section" vs. the "real-time section"?
>
> Is it enough to just run "xfs_growfs -D XXX  /mntpoint" and the rest should 
> fall into place.
>
> Again, sorry for being dense.
> I really appraciate the rapid feedback.
>

Unless you specifically made a filesystem with a realtime subvol, just
ignore it, it's not created by default.

So yes, -D <size> (in blocks) is what you want.  A little cumbersome but
not too bad :)

-Eric


<Prev in Thread] Current Thread [Next in Thread>