Intel X25-E running Linux using XFS

[Volker Kuhlmann]

As requested before, please remove mailing list admin addresses
(smartmontools-database-owner at lists.sourceforge.net) from cc. Thanks.


Cc: Pavel Machek <pavel at suse.cz>,
	"cc: smartmontools-support"@lists.sourceforge.net,
	smartmontools-database-owner at lists.sourceforge.net,
	linux-raid at vger.kernel.org, linux-kernel at vger.kernel.org,
	linux-ide at vger.kernel.org, xfs at oss.sgi.com,
	Alan Piszcz <ap at solarrain.com>,
	"X-Mailing-List: linux-raid"@vger.kernel.org
Bcc: 
Subject: Re: Intel X25-E running Linux using XFS
Reply-To: 
In-Reply-To: <alpine.DEB.1.10.0812121516190.17857 at p34.internal.lan>

On Sat 13 Dec 2008 09:23:19 NZDT +1300, Justin Piszcz wrote:

>
>
>
> On Fri, 12 Dec 2008, Pavel Machek wrote:
>
>>
>> Hi!
>>
>>> Due to the 100k write cycle lifetime, I will not be performing any more
>>> benchmarks on the SSD itself, such benchmarks are available on
>>> techreport.com but just wanted to send out the smart data for the
>>> smartmontools list and point out it is working with linux+ich7+xfs.
>>>
>>> Overall, whenever I do anything now its just like the entire machine is
>>> sitting in a ramdisk (similar when you boot knoppix and load to ramdisk).
>>>
>>> The techreport benchmarks are here:
>>> http://techreport.com/articles.x/15931
>>
>> Actually I'd like to know... how does it behave during powerfails
>> while it is being written to?
>>
>> Flashes I seen (USB / SD) developed bad sectors in that case, and
>> ext2/ext3 can't handle that. (And I don't think xfs can handle that
>> either).
>
> I have a UPS for all of my machines, so that will not be a problem in my 
> case.  I do not wish to trash a $720 SSD or my current install, if someone 
> else wants to try feel free though :)
>
> With barriers, presumably it would be safe-- but as always, a UPS/and 
> barriers and/or RAID, there is no excuse not to have regular backups incase 
> of a failure at any time..
>
> I do recall however one instance where there was an apt-get dist-upgrade 
> going and the power was cut to a machine on purpose (before/not using 
> barriers) and suffice to say-- it got mucked/suffered many corruptions
> all over the place.
>
> Justin.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Volker Kuhlmann			is list0570 with the domain in header
http://volker.dnsalias.net/	Please do not CC list postings to me.