Re: XFS file system

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Sunday 01 January 2006 06:47, Zoltan Boszormenyi wrote:
>Florin Andrei írta:
>>On Mon, 2005-12-26 at 18:46 +0100, Zoltan Boszormenyi wrote:
>>>Well, I have installed FC3 that way (XFS on most partitions)
>>>and now I regret it. All the time I have a power failure (very
>>> rare) or a kernel crash, all the files that were opened O_RDWR
>>> contain only zeroes after reboot.
>>
>>It's best if there's a match between the tools and the purpose.
>>
>>XFS was designed first and foremost for performance. Clean behaviour
>> in the case of sudden loss of power to the processing unit is not
>> exactly a high-priority design constraint for systems such as
>> these...
>>
>>http://www.sgi.com/products/
>>
>>...since the processing unit is always behind at least one layer of
>>backup/redundant/uninterruptible power supplies.
>>
>>However, the highest performance possible is one of (if not THE)
>> most important design constraints.
>>Another assumption behind XFS is that storage is the best quality
>> and therefore it does not lie when it reports back that the data
>> was flushed all the way to the magnetic layer (which is something
>> that cheap IDE drives/cards lie about, sometimes).
>>
>>Bottom line: use Ext3 for the general-purpose partitions. Use XFS
>> for your MythTV partition, or the one used to do video capture, DVD
>> backups, ISO images, etc.
>>At least that's what I do.
>
>I wanted XFS for that purpose, too.
>
>Anyway, I intend to retire this 2 years old Maxtor,
>please suggest a SATA drive that would be reliable.
>Is IBM/Hitachi Deskstar known to not lie about the flush?
>
Is that not a typo above, and should be IBM/Hitachi DeathStar?

>>>And most of the time the Oops points to the xfs driver.
>>
>>It's been many years since I've last seen an oops on healthy
>> hardware while using a "vanilla" Fedora / Red Hat install, with or
>> without XFS.
>
>I have reported a "Bad page state" for vanilla 2.6.14 on LKML, XFS
> was definitely involved:
>
>Bad page state at prep_new_page (in process 'cc1', page
> ffff8100011e1278) flags:0x4000000000000004 mapping:0000000000000000
> mapcount:0 count:-2228224 Backtrace:
>
>Call Trace:<ffffffff8015def1>{bad_page+113}
><ffffffff8015ea01>{buffered_rmqueue+609}
>       <ffffffff8015ec93>{__alloc_pages+243}
><ffffffff8016add7>{do_no_page+279}
>       <ffffffff8016b339>{__handle_mm_fault+425}
><ffffffff8017c913>{do_sync_read+211}
>       <ffffffff80352646>{do_page_fault+998}
><ffffffff8822505e>{:xfs:xfs_inactive_free_eofblocks+174}
>       <ffffffff88225240>{:xfs:xfs_release+160}
> <ffffffff80195f13>{dput+35} <ffffffff8017dce7>{__fput+375}
> <ffffffff8010f369>{error_exit+0}
>
>Trying to fix it up, but a reboot is needed
>
>Best regards,
>Zoltán Böszörményi

-- 
Cheers, Gene
People having trouble with vz bouncing email to me should add the word
'online' between the 'verizon', and the dot which bypasses vz's
stupid bounce rules.  I do use spamassassin too. :-)
Yahoo.com and AOL/TW attorneys please note, additions to the above
message by Gene Heskett are:
Copyright 2005 by Maurice Eugene Heskett, all rights reserved.


[Index of Archives]     [Current Fedora Users]     [Fedora Desktop]     [Fedora SELinux]     [Yosemite News]     [Yosemite Photos]     [KDE Users]     [Fedora Tools]     [Fedora Docs]

  Powered by Linux