Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*Might\s+have\s+found\s+a\s+bug\.\.\.\s*$/: 22 ]

Total 22 documents matching your query.

1. Re: Might have found a bug... (score: 1)
Author: xxxxxxxxxxx>
Date: Fri, 11 May 2001 12:41:10 -0400
Yes don't do that. XFS uses delayed allocation / delayed write by default any data written immediacy before a crash probably won't be on disk, this is known behavior and isn't considered a bug, it's
/archives/xfs/2001-05/msg00598.html (8,911 bytes)

2. Re: Might have found a bug... (score: 1)
Author: xxxxxxxxxxx>
Date: Fri, 11 May 2001 13:18:40 -0400 (EDT)
I understand, but the file should have been kept to its original state, no? I had garbage all over the file after the reset... -- marc. 3 out of 4 Americans make up 75% of the population.
/archives/xfs/2001-05/msg00600.html (9,115 bytes)

3. Re: Might have found a bug... (score: 1)
Author: xxxxxxxxxxx>
Date: 12 May 2001 11:51:14 -0400
Is this true even if I call fsync() or fdatasync() on the file? I am wondering about this because some applications (e.g., qmail) create a temp file, call fsync() to flush it to stable storage, use r
/archives/xfs/2001-05/msg00617.html (9,395 bytes)

4. Re: Might have found a bug... (score: 1)
Author: J. LoPresti)
Date: Sat, 12 May 2001 11:16:07 -0500
Yes that is exactly what an application should do to ensure the data has been written to disk. -- Russell Cattelan cattelan@xxxxxxxxxxx
/archives/xfs/2001-05/msg00618.html (9,457 bytes)

5. Re: Might have found a bug... (score: 1)
Author: xxxxxxxxxxx>
Date: Sat, 12 May 2001 19:06:10 -0500 (CDT)
-- Austin Gonyou Systems Architect, CCNA Coremetrics, Inc. Phone: 512-796-9023 email: austin@xxxxxxxxxxxxxxx
/archives/xfs/2001-05/msg00620.html (10,152 bytes)

6. Re: Might have found a bug... (score: 1)
Author: xxxxxxxxxxx>
Date: Mon, 14 May 2001 07:16:11 -0500
I am going to attempt to explain a little of how XFS writes data to disk, and what this means when the power is turned off. For most transactions, the modifications to the filesystem are committed to
/archives/xfs/2001-05/msg00661.html (12,439 bytes)

7. Re: Might have found a bug... (score: 1)
Author: xxxxxxxxxxx>
Date: Tue, 15 May 2001 20:15:14 -0000
At the risk of pushing the point here... I understand all that has been said about delayed writes, caches etc. However, with regard to the original question in this thread, is it still "expected beha
/archives/xfs/2001-05/msg00754.html (13,568 bytes)

8. Re: Might have found a bug... (score: 1)
Author: t@xxxxxxxxx>
Date: Tue, 15 May 2001 22:23:41 +0200
As I understand it the problem is that XFS/Linux doesn't support unwritten extents yet (unlike Irix XFS). An unwritten extent would tell XFS to zero the data on crash recovery, if it wasn't turned in
/archives/xfs/2001-05/msg00756.html (9,680 bytes)

9. Re: Might have found a bug... (score: 1)
Author: <ak@xxxxxxx>
Date: Tue, 15 May 2001 15:29:37 -0500
Sounds like a reasonable explaination to me - running xfs_bmap on the file with crud in it will tell you if it has extents or not - and if you are seeing anything but zeros this has to be the case.
/archives/xfs/2001-05/msg00757.html (9,969 bytes)

10. Re: Might have found a bug... (score: 1)
Author: 17@xxxxxxxx>
Date: Tue, 15 May 2001 17:05:47 -0400
If you followed the thread and my last explanation you will note that the file was not full of garbage but NULL's. the file had no extents which means it consisted entirely of a whole aka NULL's As t
/archives/xfs/2001-05/msg00762.html (14,844 bytes)

11. Re: Might have found a bug... (score: 1)
Author: @xxxxxxxxxx>
Date: Tue, 15 May 2001 15:27:03 -0700
Indeed, this is the case: the original file contains NULL's but the .swp file contains the correct contents, and in my case, it included the modifications done < 1 sec before hitting the power off bu
/archives/xfs/2001-05/msg00779.html (10,017 bytes)

12. Re: Might have found a bug... (score: 1)
Author: <mark@xxxxxxxxxxxxxxxxx>
Date: Fri, 11 May 2001 12:41:10 -0400
Yes don't do that. XFS uses delayed allocation / delayed write by default any data written immediacy before a crash probably won't be on disk, this is known behavior and isn't considered a bug, it's
/archives/xfs/2001-05/msg01962.html (8,911 bytes)

13. Re: Might have found a bug... (score: 1)
Author: an <cattelan@xxxxxxxxxxx>
Date: Fri, 11 May 2001 13:18:40 -0400 (EDT)
I understand, but the file should have been kept to its original state, no? I had garbage all over the file after the reset... -- marc. 3 out of 4 Americans make up 75% of the population.
/archives/xfs/2001-05/msg01964.html (9,115 bytes)

14. Re: Might have found a bug... (score: 1)
Author: ichaelv@xxxxxxxxxxxxxxxx>
Date: 12 May 2001 11:51:14 -0400
Is this true even if I call fsync() or fdatasync() on the file? I am wondering about this because some applications (e.g., qmail) create a temp file, call fsync() to flush it to stable storage, use r
/archives/xfs/2001-05/msg01981.html (9,395 bytes)

15. Re: Might have found a bug... (score: 1)
Author: xxx (Patrick J. LoPresti)
Date: Sat, 12 May 2001 11:16:07 -0500
Yes that is exactly what an application should do to ensure the data has been written to disk. -- Russell Cattelan cattelan@xxxxxxxxxxx
/archives/xfs/2001-05/msg01982.html (9,457 bytes)

16. Re: Might have found a bug... (score: 1)
Author: <austin@xxxxxxxxxxxxxxx>
Date: Sat, 12 May 2001 19:06:10 -0500 (CDT)
-- Austin Gonyou Systems Architect, CCNA Coremetrics, Inc. Phone: 512-796-9023 email: austin@xxxxxxxxxxxxxxx
/archives/xfs/2001-05/msg01984.html (10,152 bytes)

17. Re: Might have found a bug... (score: 1)
Author: II <jijo@xxxxxxxxxxxxxxx>
Date: Mon, 14 May 2001 07:16:11 -0500
I am going to attempt to explain a little of how XFS writes data to disk, and what this means when the power is turned off. For most transactions, the modifications to the filesystem are committed to
/archives/xfs/2001-05/msg02025.html (12,439 bytes)

18. Re: Might have found a bug... (score: 1)
Author: rsen" <mkp@xxxxxxxxxxxxx>
Date: Tue, 15 May 2001 20:15:14 -0000
At the risk of pushing the point here... I understand all that has been said about delayed writes, caches etc. However, with regard to the original question in this thread, is it still "expected beha
/archives/xfs/2001-05/msg02118.html (13,568 bytes)

19. Re: Might have found a bug... (score: 1)
Author: ris.buggenhout@xxxxxxxxx>
Date: Tue, 15 May 2001 22:23:41 +0200
As I understand it the problem is that XFS/Linux doesn't support unwritten extents yet (unlike Irix XFS). An unwritten extent would tell XFS to zero the data on crash recovery, if it wasn't turned in
/archives/xfs/2001-05/msg02120.html (9,680 bytes)

20. Re: Might have found a bug... (score: 1)
Author: >
Date: Tue, 15 May 2001 15:29:37 -0500
Sounds like a reasonable explaination to me - running xfs_bmap on the file with crud in it will tell you if it has extents or not - and if you are seeing anything but zeros this has to be the case.
/archives/xfs/2001-05/msg02121.html (9,969 bytes)


This search system is powered by Namazu