Using Linux Raid Systems

David Brown | 1 Jul 2011 09:23

Re: misunderstanding of spare and raid devices? - and one question more

On 30/06/2011 23:28, NeilBrown wrote:
> On Thu, 30 Jun 2011 16:21:57 +0200 Karsten Römke<k.roemke <at> gmx.de>  wrote:
>
>> Hi Phil
>>>
>>> If your CPU has free cycles, I suggest you run raid6 instead of raid5+spare.
>>>
>>> Phil
>>>
>> I started the raid 6 array and get:
>>
>> Personalities : [raid0] [raid1] [raid10] [raid6] [raid5] [raid4]
>> md0 : active raid6 sde5[4] sdd5[3] sdc5[2] sdb2[1] sda3[0]
>>         13759296 blocks level 6, 64k chunk, algorithm 2 [5/5] [UUUUU]
>>         [=================>...]  resync = 87.4% (4013184/4586432) finish=0.4min speed=20180K/sec
>                                    ^^^^^^
> Note: resync
>
>>
>> when I started the raid 5 array I get
>>
>> md0 : active raid5 sdd5[4] sde5[5](S) sdc5[2] sdb2[1] sda3[0]
>>         13759296 blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
>>         [=>...................]  recovery =  6.2% (286656/4586432) finish=0.9min speed=71664K/sec
>                                    ^^^^^^^^
> Note: recovery.
>
>>
>> so I have to expect a three times less write speed - or is this calculation
>> to simple ?

(Continue reading)

Mathias Burén | 1 Jul 2011 10:26

Picon

mdadm -git compile error

Morning,

Using the git of today I see this:

 $ make
gcc -Wall -Werror -Wstrict-prototypes -Wextra -Wno-unused-parameter
-ggdb -DSendmail=\""/usr/sbin/sendmail -t"\"
-DCONFFILE=\"/etc/mdadm.conf\" -DCONFFILE2=\"/etc/mdadm/mdadm.conf\"
-DMAP_DIR=\"/dev/.mdadm\" -DMAP_FILE=\"map\"
-DMDMON_DIR=\"/dev/.mdadm\"
-DFAILED_SLOTS_DIR=\"/dev/.mdadm/failed-slots\" -DUSE_PTHREADS   -c -o
Assemble.o Assemble.c
gcc -Wall -Werror -Wstrict-prototypes -Wextra -Wno-unused-parameter
-ggdb -DSendmail=\""/usr/sbin/sendmail -t"\"
-DCONFFILE=\"/etc/mdadm.conf\" -DCONFFILE2=\"/etc/mdadm/mdadm.conf\"
-DMAP_DIR=\"/dev/.mdadm\" -DMAP_FILE=\"map\"
-DMDMON_DIR=\"/dev/.mdadm\"
-DFAILED_SLOTS_DIR=\"/dev/.mdadm/failed-slots\" -DUSE_PTHREADS   -c -o
Build.o Build.c
gcc -Wall -Werror -Wstrict-prototypes -Wextra -Wno-unused-parameter
-ggdb -DSendmail=\""/usr/sbin/sendmail -t"\"
-DCONFFILE=\"/etc/mdadm.conf\" -DCONFFILE2=\"/etc/mdadm/mdadm.conf\"
-DMAP_DIR=\"/dev/.mdadm\" -DMAP_FILE=\"map\"
-DMDMON_DIR=\"/dev/.mdadm\"
-DFAILED_SLOTS_DIR=\"/dev/.mdadm/failed-slots\" -DUSE_PTHREADS   -c -o
Create.o Create.c
gcc -Wall -Werror -Wstrict-prototypes -Wextra -Wno-unused-parameter
-ggdb -DSendmail=\""/usr/sbin/sendmail -t"\"
-DCONFFILE=\"/etc/mdadm.conf\" -DCONFFILE2=\"/etc/mdadm/mdadm.conf\"
-DMAP_DIR=\"/dev/.mdadm\" -DMAP_FILE=\"map\"

(Continue reading)

Robin Hill | 1 Jul 2011 10:50

Picon

Re: misunderstanding of spare and raid devices? - and one question more

On Fri Jul 01, 2011 at 09:23:43 +0200, David Brown wrote:

> On 30/06/2011 23:28, NeilBrown wrote:
> > On Thu, 30 Jun 2011 16:21:57 +0200 Karsten Römke<k.roemke <at> gmx.de>  wrote:
> >
> >> Hi Phil
> >>>
> >>> If your CPU has free cycles, I suggest you run raid6 instead of raid5+spare.
> >>>
> >>> Phil
> >>>
> >> I started the raid 6 array and get:
> >>
> >> Personalities : [raid0] [raid1] [raid10] [raid6] [raid5] [raid4]
> >> md0 : active raid6 sde5[4] sdd5[3] sdc5[2] sdb2[1] sda3[0]
> >>         13759296 blocks level 6, 64k chunk, algorithm 2 [5/5] [UUUUU]
> >>         [=================>...]  resync = 87.4% (4013184/4586432) finish=0.4min speed=20180K/sec
> >                                    ^^^^^^
> > Note: resync
> >
> >>
> >> when I started the raid 5 array I get
> >>
> >> md0 : active raid5 sdd5[4] sde5[5](S) sdc5[2] sdb2[1] sda3[0]
> >>         13759296 blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
> >>         [=>...................]  recovery =  6.2% (286656/4586432) finish=0.9min speed=71664K/sec
> >                                    ^^^^^^^^
> > Note: recovery.
> >
> >>

(Continue reading)

David Brown | 1 Jul 2011 12:18

Re: misunderstanding of spare and raid devices? - and one question more

On 01/07/2011 10:50, Robin Hill wrote:
> On Fri Jul 01, 2011 at 09:23:43 +0200, David Brown wrote:
>
>> On 30/06/2011 23:28, NeilBrown wrote:
>>> On Thu, 30 Jun 2011 16:21:57 +0200 Karsten Römke<k.roemke <at> gmx.de>   wrote:
>>>
>>>> Hi Phil
>>>>>
>>>>> If your CPU has free cycles, I suggest you run raid6 instead of raid5+spare.
>>>>>
>>>>> Phil
>>>>>
>>>> I started the raid 6 array and get:
>>>>
>>>> Personalities : [raid0] [raid1] [raid10] [raid6] [raid5] [raid4]
>>>> md0 : active raid6 sde5[4] sdd5[3] sdc5[2] sdb2[1] sda3[0]
>>>>          13759296 blocks level 6, 64k chunk, algorithm 2 [5/5] [UUUUU]
>>>>          [=================>...]  resync = 87.4% (4013184/4586432) finish=0.4min speed=20180K/sec
>>>                                     ^^^^^^
>>> Note: resync
>>>
>>>>
>>>> when I started the raid 5 array I get
>>>>
>>>> md0 : active raid5 sdd5[4] sde5[5](S) sdc5[2] sdb2[1] sda3[0]
>>>>          13759296 blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
>>>>          [=>...................]  recovery =  6.2% (286656/4586432) finish=0.9min speed=71664K/sec
>>>                                     ^^^^^^^^
>>> Note: recovery.
>>>

(Continue reading)

Robin Hill | 1 Jul 2011 13:29

Picon

Re: misunderstanding of spare and raid devices? - and one question more

On Fri Jul 01, 2011 at 12:18:22PM +0200, David Brown wrote:

> On 01/07/2011 10:50, Robin Hill wrote:
> > On Fri Jul 01, 2011 at 09:23:43 +0200, David Brown wrote:
> >
> >> What's the difference between a "resync" and a "recovery"?  Is it that a
> >> "resync" will read the whole stripe, check if it is valid, and if it is
> >> not it then generates the parity, while a "recovery" will always
> >> generate the parity?
> >>
> >  From the names, recovery would mean that it's reading from N-1 disks,
> > and recreating data/parity to rebuild the final disk (as when it
> > recovers from a drive failure), whereas resync will be reading from all
> > N disks and checking/recreating the parity (as when you're running a
> > repair on the array).
> >
> > The main reason I can see for doing a resync on RAID6 rather than a
> > recovery is if the data reconstruction from the Q parity is far slower
> > that the construction of the Q parity itself (I've no idea how the
> > mathematics works out for this).
> >
> 
> Well, data reconstruction from Q parity /is/ more demanding than 
> constructing the Q parity in the first place (the mathematics is the 
> part that I know about).  That's why a two-disk degraded raid6 array is 
> significantly slower (or, more accurately, significantly more 
> cpu-intensive) than a one-disk degraded raid6 array.
> 
> But that doesn't make a difference here - you are rebuilding one or two 
> disks, so you have to use the data you've got whether you are doing a

(Continue reading)

David Brown | 1 Jul 2011 14:45

Re: misunderstanding of spare and raid devices? - and one question more

On 01/07/2011 13:29, Robin Hill wrote:
> On Fri Jul 01, 2011 at 12:18:22PM +0200, David Brown wrote:
>
>> On 01/07/2011 10:50, Robin Hill wrote:
>>> On Fri Jul 01, 2011 at 09:23:43 +0200, David Brown wrote:
>>>
>>>> What's the difference between a "resync" and a "recovery"?  Is it that a
>>>> "resync" will read the whole stripe, check if it is valid, and if it is
>>>> not it then generates the parity, while a "recovery" will always
>>>> generate the parity?
>>>>
>>>    From the names, recovery would mean that it's reading from N-1 disks,
>>> and recreating data/parity to rebuild the final disk (as when it
>>> recovers from a drive failure), whereas resync will be reading from all
>>> N disks and checking/recreating the parity (as when you're running a
>>> repair on the array).
>>>
>>> The main reason I can see for doing a resync on RAID6 rather than a
>>> recovery is if the data reconstruction from the Q parity is far slower
>>> that the construction of the Q parity itself (I've no idea how the
>>> mathematics works out for this).
>>>
>>
>> Well, data reconstruction from Q parity /is/ more demanding than
>> constructing the Q parity in the first place (the mathematics is the
>> part that I know about).  That's why a two-disk degraded raid6 array is
>> significantly slower (or, more accurately, significantly more
>> cpu-intensive) than a one-disk degraded raid6 array.
>>
>> But that doesn't make a difference here - you are rebuilding one or two

(Continue reading)

NeilBrown | 1 Jul 2011 15:02

Picon

Re: misunderstanding of spare and raid devices? - and one question more

On Fri, 01 Jul 2011 14:45:00 +0200 David Brown <david <at> westcontrol.com> wrote:

> On 01/07/2011 13:29, Robin Hill wrote:
> > On Fri Jul 01, 2011 at 12:18:22PM +0200, David Brown wrote:
> >
> >> On 01/07/2011 10:50, Robin Hill wrote:
> >>> On Fri Jul 01, 2011 at 09:23:43 +0200, David Brown wrote:
> >>>
> >>>> What's the difference between a "resync" and a "recovery"?  Is it that a
> >>>> "resync" will read the whole stripe, check if it is valid, and if it is
> >>>> not it then generates the parity, while a "recovery" will always
> >>>> generate the parity?
> >>>>
> >>>    From the names, recovery would mean that it's reading from N-1 disks,
> >>> and recreating data/parity to rebuild the final disk (as when it
> >>> recovers from a drive failure), whereas resync will be reading from all
> >>> N disks and checking/recreating the parity (as when you're running a
> >>> repair on the array).
> >>>
> >>> The main reason I can see for doing a resync on RAID6 rather than a
> >>> recovery is if the data reconstruction from the Q parity is far slower
> >>> that the construction of the Q parity itself (I've no idea how the
> >>> mathematics works out for this).
> >>>
> >>
> >> Well, data reconstruction from Q parity /is/ more demanding than
> >> constructing the Q parity in the first place (the mathematics is the
> >> part that I know about).  That's why a two-disk degraded raid6 array is
> >> significantly slower (or, more accurately, significantly more
> >> cpu-intensive) than a one-disk degraded raid6 array.

(Continue reading)

Krzysztof Wojcik | 1 Jul 2011 18:34

Picon

[PATCH] imsm: FIX: getinfo_super_imsm_volume() doesn't fill all disk information

From: Adam Kwolek <adam.kwolek <at> intel.com>

Issue:
getinfo_super_imsm_volume() clears entire 'info' structure before filling with new
information. Disk number and raid_disk can be required later by caller
but it is also cleared.

Resolution:
Patch backs up all disk information before zeroing info structure and
restore it before function return.
The patch aslo reverses workaround for the same problem introduced
in commit:

80e4abc99c7f5a16c56c1c4084bfad63aec03c01
FIX: Cannot create volume

Signed-off-by: Adam Kwolek <adam.kwolek <at> intel.com>
Signed-off-by: Krzysztof Wojcik <krzysztof.wojcik <at> intel.com>
---
 Create.c      |    9 ---------
 super-intel.c |   13 ++++++++++++-
 2 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/Create.c b/Create.c
index 48115db..8d88aa1 100644
--- a/Create.c
+++ b/Create.c
 <at>  <at>  -856,15 +856,6  <at>  <at>  int Create(struct supertype *st, char *mddev,
 					/* getinfo_super might have lost these ... */
 					inf->disk.major = major(stb.st_rdev);

(Continue reading)

Simon Matthews | 2 Jul 2011 06:41

Picon

Re: Can't start array and Negative "Used Dev Size"

Neil,

On Tue, Jun 28, 2011 at 10:18 PM, NeilBrown <neilb <at> suse.de> wrote:
> On Tue, 28 Jun 2011 21:29:37 -0700 Simon Matthews
> <simon.d.matthews <at> gmail.com> wrote:
>
>> Problem 1: "Used Dev Size"
>> ====================
>> Note: the system is a Gentoo box, so perhaps I have missed a kernel
>> configuration option or use flag to deal with large hard drives.
>>
>> A week or two ago, I resized a raid1 array using 2x3TB drives. I went
>
> Oopps.  That array is using 0.90 metadata which can only handle up to 2TB
> devices.  The 'resize' code should catch that you are asking the impossible,
> but it doesn't it seems.
>
> You need to simply recreate the array as 1.0.
> i.e.
>  mdadm -S /dev/md5
>  mdadm  -C /dev/md5 --metadata 1.0 -l1 -n2 --assume-clean

Before I do this (tomorrow), do I need to add the partitions to the command:

mdadm  -C /dev/md5 --metadata 1.0 -l1 -n2 --assume-clean /dev/sdd2 /dev/sdc2

Simon
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo <at> vger.kernel.org

(Continue reading)

Simon Matthews | 2 Jul 2011 08:19

Picon

Re: Can't start array and Negative "Used Dev Size"

Neil,

On Fri, Jul 1, 2011 at 9:41 PM, Simon Matthews
<simon.d.matthews <at> gmail.com> wrote:
> Neil,
>
> On Tue, Jun 28, 2011 at 10:18 PM, NeilBrown <neilb <at> suse.de> wrote:
>> On Tue, 28 Jun 2011 21:29:37 -0700 Simon Matthews
>> <simon.d.matthews <at> gmail.com> wrote:
>>
>>> Problem 1: "Used Dev Size"
>>> ====================
>>> Note: the system is a Gentoo box, so perhaps I have missed a kernel
>>> configuration option or use flag to deal with large hard drives.
>>>
>>> A week or two ago, I resized a raid1 array using 2x3TB drives. I went
>>
>> Oopps.  That array is using 0.90 metadata which can only handle up to 2TB
>> devices.  The 'resize' code should catch that you are asking the impossible,
>> but it doesn't it seems.
>>
>> You need to simply recreate the array as 1.0.
>> i.e.
>>  mdadm -S /dev/md5
>>  mdadm  -C /dev/md5 --metadata 1.0 -l1 -n2 --assume-clean
>
> Before I do this (tomorrow), do I need to add the partitions to the command:
>
> mdadm  -C /dev/md5 --metadata 1.0 -l1 -n2 --assume-clean /dev/sdd2 /dev/sdc2

(Continue reading)

Group

gmane.linux.raid.

Advertisement

Search Archive

Page

1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 43

Articles in period: 423

July 2011

Language

Change language

Options

Current view: Threads only / Showing only 20 lines / Not hiding cited text.
Change to All messages, whole messages, or hide cited text.

Post a message
NNTP Newsgroup
Classic Gmane web interface
XML

XML

RSS Feed
List Information

About Gmane

Recent entries

Problem with Raid1 when all drives failed (20 Jun 2013)
Growing a 5-drive RAID6 - some initial questions (19 Jun 2013)
Proper way to delete an old RAID1? (18 Jun 2013)
Set AUTOCHECK=false (18 Jun 2013)
Optimal value for bitmap-chunk option (18 Jun 2013)
[PATCH 0/6] raid6check fixes (18 Jun 2013)
UDEV Rule for Timeouts (18 Jun 2013)
Quick Loans (17 Jun 2013)
Quick Loans (17 Jun 2013)
Quick Loans (17 Jun 2013)
Please assist with missing RAID 5 array. (16 Jun 2013)
RAID 5: low sequential write performance? (16 Jun 2013)
Mdadm server eating drives (14 Jun 2013)
Mdadm server eating drives (14 Jun 2013)
Fwd: Help with failed RAID-5 -> 6 migration (13 Jun 2013)
RAID-10 keeps aborting (13 Jun 2013)
RAID-10 keeps aborting (12 Jun 2013)
Fwd: Help with failed RAID-5 -> 6 migration (12 Jun 2013)
Mdadm server eating drives (12 Jun 2013)
Quick Loans(Apply Today) (12 Jun 2013)
Fwd: mismatches in raid1, clarification please (11 Jun 2013)

Archives

239	June 2013
263	May 2013
370	April 2013
314	March 2013
491	February 2013
350	January 2013
180	December 2012
305	November 2012
304	October 2012
370	September 2012
319	August 2012
436	July 2012
403	June 2012
470	May 2012
473	April 2012
350	March 2012
276	February 2012
317	January 2012
294	December 2011
342	November 2011
328	October 2011
439	September 2011
246	August 2011
423	July 2011
444	June 2011
473	May 2011
450	April 2011
439	March 2011
907	February 2011
844	January 2011
380	December 2010
445	November 2010
180	October 2010
229	September 2010
260	August 2010
267	July 2010
388	June 2010
243	May 2010
424	April 2010
499	March 2010
396	February 2010
402	January 2010
414	December 2009
532	November 2009
461	October 2009
540	September 2009
372	August 2009
215	July 2009
416	June 2009
547	May 2009
454	April 2009
269	March 2009
211	February 2009
373	January 2009
466	December 2008
324	November 2008
246	October 2008
105	September 2008
116	August 2008
347	July 2008
296	June 2008
373	May 2008
203	April 2008
369	March 2008
332	February 2008
348	January 2008
286	December 2007
244	November 2007
493	October 2007
185	September 2007
175	August 2007
331	July 2007
411	June 2007
370	May 2007
195	April 2007
290	March 2007
247	February 2007
239	January 2007
169	December 2006
210	November 2006
280	October 2006
319	September 2006
327	August 2006
314	July 2006
352	June 2006
286	May 2006
360	April 2006
307	March 2006
314	February 2006
482	January 2006
335	December 2005
265	November 2005
193	October 2005
251	September 2005
287	August 2005
255	July 2005
245	June 2005
192	May 2005
201	April 2005
368	March 2005
196	February 2005
524	January 2005
232	December 2004
282	November 2004
287	October 2004
194	September 2004
359	August 2004
208	July 2004
217	June 2004
266	May 2004
193	April 2004
272	March 2004
235	February 2004
299	January 2004
117	December 2003
189	November 2003
232	October 2003
140	September 2003
116	August 2003
105	July 2003
211	June 2003
203	May 2003
158	April 2003
157	March 2003
257	February 2003
117	January 2003
79	December 2002
133	November 2002
162	October 2002
128	September 2002
101	August 2002
139	July 2002
174	June 2002
116	May 2002
76	April 2002

Design

Zawodny | Right Menu | Left Menu

Your Own Design

Paste the URL of your CSS below. Download CSS template