Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*separate\s+log\s+and\s+structure\s+from\s+user\s+data\s+device\?\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*separate\s+log\s+and\s+structure\s+from\s+user\s+data\s+device\?\s*$/: 28 ]

Total 28 documents matching your query.

1. separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Mon, 5 Jun 2006 09:49:10 +0300 (EEST); I saw with the logdev parameter it is possible to specify an external log device on a separate disk and partition. What would be even more interesting for my special purposes is whether even the fil; /archives/xfs/2006-06/msg00012.html (8,888 bytes)
2. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Mon, 5 Jun 2006 20:55:29 +1000; The realtime subvolume will indeed give you this split. See xfs(5) and mkfs/xfs(8) where most doco resides on this. Not 100% sure what unreliable means here from a software POV... would we be seeing; /archives/xfs/2006-06/msg00013.html (10,120 bytes)
3. Re: separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Mon, 5 Jun 2006 14:22:35 +0300 (EEST); Hi and thanks for your quick reply! Thanks for clarifying. That had been a bit unclear to me from the docs. Also realized only now that xfsctl has to be used to set an empty file's realtime bit to re; /archives/xfs/2006-06/msg00014.html (11,108 bytes)
4. Re: separate log and structure from user data device? (score: 1): Author: Keith Owens <kaos@xxxxxxx>; Date: Tue, 06 Jun 2006 00:19:08 +1000; The ATA Streaming Feature Set defines the Handle Stream Error (HSE) bit to mark data which is critical, and therefore needs full error recovery. That leaves all other data to be handled as best case,; /archives/xfs/2006-06/msg00015.html (12,018 bytes)
5. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Tue, 6 Jun 2006 09:13:48 +1000; Hi Jan, No worries. You can set the rtinherit bit on a directory, and all new files created there will be automatically written to the realtime device. Theres also a (probably undocumented, I really; /archives/xfs/2006-06/msg00016.html (11,464 bytes)
6. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Tue, 6 Jun 2006 10:12:58 +1000; Hmm, OK, interesting. Its a trivial thing from the filesystem POV - in XFS, it would require a change in fs/xfs/linux-2.6/xfs_buf.c - the call into submit_bio() there is the point all metadata and lo; /archives/xfs/2006-06/msg00017.html (10,990 bytes)
7. Re: separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Wed, 7 Jun 2006 14:05:57 +0300 (EEST); (Short-lived discussion/probing at http://lkml.org/lkml/2006/5/18/261) Don't know what a bio flag is but it sounds good :), like some ioctl() command to tag the block device into critical / noncritic; /archives/xfs/2006-06/msg00023.html (11,608 bytes)
8. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Thu, 8 Jun 2006 10:42:42 +1000; Its one way. Also try: wrote 104857600/104857600 bytes at offset 0 100 MiB, 100 ops; 0:00:01.00 (95.473 MiB/sec and 95.4726 ops/sec) wrote 104857600/104857600 bytes at offset 0 100 MiB, 100 ops; 0:00; /archives/xfs/2006-06/msg00033.html (11,690 bytes)
9. RE: separate log and structure from user data device? (score: 1): Author: "Patrick Freeman" <patrickf@xxxxxxxxxxxxxx>; Date: Wed, 7 Jun 2006 22:34:34 -0400; Of [ ... ] with XFS I just wanted to comment that this is about 320 MB/s which is very reasonable for a 6 disk raid0 volume (between 50 and 60 MB/s per drive with 6 disks). It would be kind of at th; /archives/xfs/2006-06/msg00036.html (10,094 bytes)
10. Re: separate log and structure from user data device? (score: 1): Author: pg_xfs@xxxxxxxxxxxxxxxxxx (Peter Grandi); Date: Thu, 8 Jun 2006 15:15:39 +0100; [ ... curious complaints about write performance ... ] Thats too kind :-). However, I am puzzled as you comment on 'O_DIRECT' but not on 'O_SYNC'. IIRC XFS performance, and in particular the strategy; /archives/xfs/2006-06/msg00037.html (10,017 bytes)
11. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Fri, 9 Jun 2006 08:50:26 +1000; 12PM -0600, Rishi Malik wrote: Thanks for the reply Nathan. I was unclear. I know it wasn't included until later, but I'm looking for the patch set t; /archives/xfs/2006-06/msg00040.html (10,810 bytes)
12. Re: separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Fri, 16 Jun 2006 14:38:19 +0300 (EEST); of logprint binaries can print a log from an XFS file system on a 64bit or 32bit os. The following file(s) were checked into: longdrop.melbourne.sgi.; /archives/xfs/2006-06/msg00073.html (13,223 bytes)
13. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Mon, 19 Jun 2006 14:44:49 +1000; workload, file sizes, write patterns, etc. There is no magic number that will give you the best performance in all cases. ... stripe unit = MD chunk; /archives/xfs/2006-06/msg00082.html (12,378 bytes)
14. Re: separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Tue, 20 Jun 2006 14:34:10 +0300 (EEST); disk? On Mon, 19 Jun 2006, Avuton Olrich wrote: On 6/19/06, Nathan Scott <nathans@xxxxxxx> wrote: How reproducible is it? Is it reproducible even af; /archives/xfs/2006-06/msg00096.html (11,975 bytes)
15. separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Mon, 5 Jun 2006 09:49:10 +0300 (EEST); supported by the underlying device XFS mounting filesystem dm-0 Filesystem "dm-0": X; /archives/xfs/2006-06/msg00176.html (8,888 bytes)
16. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Mon, 5 Jun 2006 20:55:29 +1000; specify an external log device on a separate disk and partition. What would be even more interesting for my special purposes is whether even the fil; /archives/xfs/2006-06/msg00177.html (10,120 bytes)
17. Re: separate log and structure from user data device? (score: 1): Author: Jan Wagner <jwagner@xxxxxxxxxxx>; Date: Mon, 5 Jun 2006 14:22:35 +0300 (EEST); split. See xfs(5) and mkfs/xfs(8) where most doco resides on this. Not 100% sure what unreliable means here from a software POV... would we be seeing; /archives/xfs/2006-06/msg00178.html (11,108 bytes)
18. Re: separate log and structure from user data device? (score: 1): Author: Keith Owens <kaos@xxxxxxx>; Date: Tue, 06 Jun 2006 00:19:08 +1000; rifying. That had been a bit unclear to me from the docs. Also realized only now that xfsctl has to be used to set an empty file's realtime bit to re; /archives/xfs/2006-06/msg00179.html (12,018 bytes)
19. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Tue, 6 Jun 2006 09:13:48 +1000; tream Error (HSE) bit to mark data which is critical, and therefore needs full error recovery. That leaves all other data to be handled as best case,; /archives/xfs/2006-06/msg00180.html (11,464 bytes)
20. Re: separate log and structure from user data device? (score: 1): Author: Nathan Scott <nathans@xxxxxxx>; Date: Tue, 6 Jun 2006 10:12:58 +1000; on a directory, and all new files created there will be automatically written to the realtime device. Theres also a (probably undocumented, I really; /archives/xfs/2006-06/msg00181.html (10,990 bytes)

Current List: 1 - 20
Page: [1] [2]