- 1. Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 10:00:43 -0500
- I've just transferred my old ext2 FS into XFS FS and did the following test. What I got surprises me. First, I modify a file with vim, save it, and go back to bash shell. Then, I hit the power butto
- /archives/xfs/2001-12/msg00075.html (8,223 bytes)
- 2. Re: Files on XFS not safe?! (score: 1)
- Author: Joshua Baker-LePain <jlb17@xxxxxxxx>
- Date: Wed, 5 Dec 2001 10:07:14 -0500 (EST)
- http://oss.sgi.com/projects/xfs/faq.html#nulls -- Joshua Baker-LePain Department of Biomedical Engineering Duke University
- /archives/xfs/2001-12/msg00076.html (8,932 bytes)
- 3. Re: Files on XFS not safe?! (score: 1)
- Author: Andrew Klaassen <ak@xxxxxxx>
- Date: Wed, 5 Dec 2001 10:09:18 -0500
- http://oss.sgi.com/projects/xfs/faq.html#nulls Andrew Klaassen
- /archives/xfs/2001-12/msg00077.html (9,610 bytes)
- 4. Re: Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 10:47:41 -0500
- Thanks. Is this true for all the journaling FS, say ReiserFS, JFS? Xianglong Yuan
- /archives/xfs/2001-12/msg00078.html (10,517 bytes)
- 5. Re: Files on XFS not safe?! (score: 1)
- Author: Harri Haataja <harri.haataja@xxxxxxxxxxxxxx>
- Date: Wed, 5 Dec 2001 17:56:47 +0200
- Maybe I might bring up this http://kerneltrap.com/article.php?sid=389 -- Real Programmers don't write in FORTRAN. FORTRAN is for wimp engineers who wear white socks. They get excited over finite stat
- /archives/xfs/2001-12/msg00079.html (10,453 bytes)
- 6. Re: Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 11:14:21 -0500
- Is there any way to recover the old content? It is still at somewhere in the disk, isn't it? Xianglong Yuan -- Dept. Materials Science and Engineering MIT, Room 13-4050 Cambridge, MA 02139
- /archives/xfs/2001-12/msg00080.html (9,888 bytes)
- 7. Re: Files on XFS not safe?! (score: 1)
- Author: Walt H <waltabbyh@xxxxxxxxxxxxxx>
- Date: Wed, 05 Dec 2001 08:15:56 -0800
- The potential is certainly there. Most jfs use delayed writing. When you save something to disk, it isn't immediately flushed to disk. Had you ran 'sync' prior to pushing the power button, perhaps th
- /archives/xfs/2001-12/msg00081.html (12,022 bytes)
- 8. Re: Files on XFS not safe?! (score: 1)
- Author: Steve Lord <lord@xxxxxxx>
- Date: 05 Dec 2001 10:28:33 -0600
- No, it was in the memory on your machine when you turned off the power, now it is gone. The problem here is partially the editor, partially you, and partially XFS. o The editor because it does not us
- /archives/xfs/2001-12/msg00082.html (11,072 bytes)
- 9. Re: Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 12:03:25 -0500
- I presume there is a log that has the information when the data-flashing is scheduled to start, whether it has done or not, and what the original file state (meta-data?) is. Now, if the system crash
- /archives/xfs/2001-12/msg00086.html (12,685 bytes)
- 10. Re: Files on XFS not safe?! (score: 1)
- Author: Steve Lord <lord@xxxxxxx>
- Date: 05 Dec 2001 11:20:26 -0600
- Writing file data out to disk in XFS does not go anywhere near the log, only the allocation of extents goes in the log. Also, think about what the editor is probably doing: o Create a new file with a
- /archives/xfs/2001-12/msg00089.html (11,414 bytes)
- 11. Re: Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 12:49:13 -0500
- Thanks, Steve, it becomes much clear to me now as why it is. This problem is critical to us as we run structure simulation for up to several weeks and dump intermediate data every couple hours and a
- /archives/xfs/2001-12/msg00093.html (11,943 bytes)
- 12. Re: Files on XFS not safe?! (score: 1)
- Author: Andi Kleen <ak@xxxxxxx>
- Date: Wed, 5 Dec 2001 19:26:38 +0100
- The scenario Steve described obviously does not apply to appending to big files, because there is no rename() from a temporary file involved. If it has been flushed they will be on disk. You can for
- /archives/xfs/2001-12/msg00096.html (10,262 bytes)
- 13. Re: Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 14:19:27 -0500
- But only be on disk after flushing, correct? What if crashing during data dumping before it can issue a fsync(), or even during fsync()? I think the best bet is to remove the old file only after ens
- /archives/xfs/2001-12/msg00105.html (10,924 bytes)
- 14. Re: Files on XFS not safe?! (score: 1)
- Author: Andi Kleen <ak@xxxxxxx>
- Date: Wed, 5 Dec 2001 20:24:33 +0100
- Yes, but until that the old data is not gone because the file is not truncated . The problem with the editors is that the old data is usually lost in some temporary file; that won't be the problem he
- /archives/xfs/2001-12/msg00107.html (11,915 bytes)
- 15. Re: Files on XFS not safe?! (score: 1)
- Author: Steve Lord <lord@xxxxxxx>
- Date: 05 Dec 2001 13:23:48 -0600
- You can also use O_SYNC when you open the file, and with xfs the file data and associated metadata will be on disk before the write call returns. There is a mount option osyncisdsync which makes go f
- /archives/xfs/2001-12/msg00108.html (11,965 bytes)
- 16. Re: Files on XFS not safe?! (score: 1)
- Author: Xianglong Yuan <yuanx@xxxxxxx>
- Date: Wed, 5 Dec 2001 15:47:18 -0500
- I see what I am testing using vim is different from what my simulation actually recording data appendingly :-<< But, anyhow, ordered mode in ext3 seems to be my best bet. I'll do some reading to fin
- /archives/xfs/2001-12/msg00118.html (10,716 bytes)
- 17. Re: Files on XFS not safe?! (score: 1)
- Author: Dan Hollis <goemon@xxxxxxxxx>
- Date: Wed, 5 Dec 2001 13:49:50 -0800 (PST)
- FWIW I never had reiserfs nulling files... it is either more resistant to such situations or it operates in a totally different way than XFS. -Dan -- [-] Omae no subete no kichi wa ore no mono da. [-
- /archives/xfs/2001-12/msg00124.html (9,811 bytes)
- 18. Re: Files on XFS not safe?! (score: 1)
- Author: Peter Wächtler <pwaechtler@xxxxxxxxxxxxx
- Date: Thu, 06 Dec 2001 10:47:52 +0100
- Dan Hollis schrieb: But perhaps "scrambled tails"? AFAIK, ReiserFS moves the tails of files to put several into one disk block. I lost a lot of work due to this - ok it was just setiathome :))
- /archives/xfs/2001-12/msg00161.html (10,446 bytes)
- 19. Re: Files on XFS not safe?! (score: 1)
- Author: Dan Hollis <goemon@xxxxxxxxx>
- Date: Thu, 6 Dec 2001 01:55:36 -0800 (PST)
- Never had scrambled tails either.. You can turn off tailmerging with mount option 'notail' ... -Dan -- [-] Omae no subete no kichi wa ore no mono da. [-]
- /archives/xfs/2001-12/msg00162.html (9,975 bytes)
- 20. Re: Files on XFS not safe?! (score: 1)
- Author: Federico Sevilla III <jijo@xxxxxxxxxxxxxxxxxxxx>
- Date: Thu, 6 Dec 2001 17:58:43 +0800 (PHT)
- Both ext2 and ReiserFS I find are less prone, but not immune. And I've had messups with ReiserFS, too. They don't get nulled out, though. I'd just get some random character in my newly-edited config
- /archives/xfs/2001-12/msg00163.html (10,058 bytes)
This search system is powered by
Namazu