xfs
[Top] [All Lists]

Re: Stress test 015 need to check partition size

To: "Marcelo E. Magallon" <marcelo.magallon@xxxxxxxxxxx>
Subject: Re: Stress test 015 need to check partition size
From: Andrew Gildfind <ajag@xxxxxxxxxxxxxxxxxxxxxxx>
Date: Tue, 19 Jun 2001 09:23:41 +1000
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <200106182243.IAA03759@snort.melbourne.sgi.com>; from dxm@snort on Tue, Jun 19, 2001 at 08:43:13AM +1000
References: <20010618095701.A984@ysabell.wh.vaih> <200106182243.IAA03759@snort.melbourne.sgi.com>
Sender: owner-linux-xfs@xxxxxxxxxxx
On Tue, Jun 19, 2001 at 08:43:13AM +1000, Daniel Moore wrote:
> 
> "Marcelo E. Magallon" writes:
>  => Hi,
>  => 
>  =>  015 is trying to create a 50MB data section on the device, but don't
>  =>  check if the operation is sucessful:
>  => 
>  =>  mkfs -t xfs -f -d size=50m $SCRATCH_DEV >/dev/null
>  => 
>  =>  016 does check, but exits with an error instead of using _notrun.
> 
> That's because there's a non-specified arbitary minimum size for the
> stress test partitions. I'd suggest using something more like 100Mb...

I've run into this before. According to mkfs.xfs(8) min allocation group size
is 16 MB. xfs_repair needs three allocation groups to verify the superblock,
which means the smallest filesystem that I've been able to use for the tests
(see 041/042) is 3x16Mb.

> 
> PS. We're still running the QA against the daily CVS trees, but our
> QA suite is in maintenance mode rather than under active developemnt
> (for now at least).
> 
> -----------------------------------------------------
>  Daniel Moore                  dxm@xxxxxxx
>  R&D Software Engineer         Phone: +61-3-98348209
>  SGI Performance Tools Group   Fax:   +61-3-98132378
> -----------------------------------------------------
> 

-- 
Andrew Gildfind - R&D Software Engineer - SGI Melbourne Australia
email: ajag@xxxxxxx - work: +61.3.9834.8200 mobile: 0412.834.183

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