Re: [CentOS] RAID 5 setup?

2010-03-27 Thread Chan Chung Hang Christopher
Benjamin Franz wrote:
 Robert Heller wrote:
 I suspect that this is a simular case to what I did: I have a server
 with 4 drives.  I have several (small) RAID1 partitions (/boot, /,
 /usr, /var, etc.) with 4 mirrors and one large RAID5 with three
 partitions and a hot spare (a LVM volumn group, containing /home and
 some other partitions). I would guess that the admin with the 8-way
 RAID1 for the OS probably also has a 6 or 8 disk RAID5 or RAID6 for
 the bulk of the disks
 Yup. 8 way RAID1 for the OS, 8 way RAID6 for the data. I was hoping when 
 I setup the 8-way RAID1 for the OS that I would get really good read 
 speeds since md is supposed to stripe reads from RAID1, but in practice 
 the RAID6 completely kills it for read performance (~61 MB/sec from the 
 RAID1 partition vs ~200 MB/sec from the RAID6 partition).

What are you running? I think there was a patch that evened out the 
reads across all members as it would at times solely read from one and 
then another...
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-27 Thread John R Pierce
Robert Spangler wrote:
 On Thursday 25 March 2010 18:10, Robert Heller wrote:

   
  The prefered way to go would be RAID10 (RAID1 (mirror) + RAID0 (stripe)).
  Form pairs as RAID1, then strip the pairs.  With 8 disks, this would 4
  pairs, 1.5TB/pair = 1.5*4 = 6TB total.
 

 I am just starting to look into this RAID and I was wondering wouldn't RAID01 
 be better then RAID10?  In a 4 disc system having the first two using 
 stripping and then backing them up the second set with mirrors?

 My though is having D1 and D2 as the primary drives stripping and then having 
 D3 backup D1 and D4 backup D2.

 And if enough room place a couple more drives in the system as hot standby's.

 Or am I looking at this all wrong?
   


for all practical purposes its the same thing. if it was really 
stripe then mirror, a naive mirror handler would think it would have to 
remirror both drives when one half of one of the stripesets failed and 
was replaced. but in fact, the mirror handlres tend to be well aware 
of whats going on.  mirror 0+1  aand stripe that with mirrored 2+3, 
and its really all the samethe native raid10 in newer mdraid is 
cleaner because you don't end up with extra partial volume metadevices...


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-27 Thread Ross Walker
On Mar 27, 2010, at 5:07 AM, John R Pierce pie...@hogranch.com wrote:

 Robert Spangler wrote:
 On Thursday 25 March 2010 18:10, Robert Heller wrote:


 The prefered way to go would be RAID10 (RAID1 (mirror) + RAID0  
 (stripe)).
 Form pairs as RAID1, then strip the pairs.  With 8 disks, this  
 would 4
 pairs, 1.5TB/pair = 1.5*4 = 6TB total.


 I am just starting to look into this RAID and I was wondering  
 wouldn't RAID01
 be better then RAID10?  In a 4 disc system having the first two using
 stripping and then backing them up the second set with mirrors?

 My though is having D1 and D2 as the primary drives stripping and  
 then having
 D3 backup D1 and D4 backup D2.

 And if enough room place a couple more drives in the system as hot  
 standby's.

 Or am I looking at this all wrong?



 for all practical purposes its the same thing. if it was really
 stripe then mirror, a naive mirror handler would think it would have  
 to
 remirror both drives when one half of one of the stripesets failed and
 was replaced. but in fact, the mirror handlres tend to be well  
 aware
 of whats going on.  mirror 0+1  aand stripe that with mirrored  
 2+3,
 and its really all the samethe native raid10 in newer mdraid is
 cleaner because you don't end up with extra partial volume  
 metadevices...

RAID0+1 is never a good configuration because a single drive failure  
in a RAID0 stripe fails out the whole stripe, and with say an 4x2  
RAID0+1, you are more likely to hit a disk failure with 4 drives in a  
RAID0 then 2 in a RAID1.

That's why RAID1+0 came about.

-Ross
  
   
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-27 Thread Benjamin Franz
Chan Chung Hang Christopher wrote:
 Benjamin Franz wrote:
   
 Yup. 8 way RAID1 for the OS, 8 way RAID6 for the data. I was hoping when 
 I setup the 8-way RAID1 for the OS that I would get really good read 
 speeds since md is supposed to stripe reads from RAID1, but in practice 
 the RAID6 completely kills it for read performance (~61 MB/sec from the 
 RAID1 partition vs ~200 MB/sec from the RAID6 partition).
 

 What are you running? I think there was a patch that evened out the 
 reads across all members as it would at times solely read from one and 
 then another...

   

I'm running fully up-to-date CentOS 5.4, kernel 2.6.18-164.15.1. The 
test was done with a default run of bonnie++. Watching the disk I/O 
while it ran suggested that it was only using some of the disks at a 
time to read (but changing which ones every few seconds).

8 x RAID1

Version  1.96   --Sequential Output-- --Sequential Input- 
--Random-
Concurrency   1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- 
--Seeks--
MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  
/sec %CP
pbox16.freerun. 16G   586  95 17263   5 13016   1  2263  97 60793   3 
644.2   4
Latency   203ms  188s1521ms   14714us 557ms
1265ms
Version  1.96   --Sequential Create-- Random 
Create
pbox16  -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
  files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  
/sec %CP
 16 12418  16 + +++ 10999  12 27730  34 + +++ 
29916  32
Latency   109us 638us 681us 200us  25us  
39us

8 x RAID6
Version  1.96   --Sequential Output-- --Sequential Input- 
--Random-
Concurrency   1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- 
--Seeks--
MachineSize K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  
/sec %CP
pbox16.freerun. 16G   611  96 90060  15 55711  14  2407  96 200901  27 
522.3  15
Latency 13157us 604ms1727ms   32420us 142ms   
73131us
Version  1.96   --Sequential Create-- Random 
Create
pbox16  -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
  files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  
/sec %CP
 16 10298  16 + +++ 24208  32 30395  45 + +++ 
32511  43
Latency 18504us 598us 613us  87us  25us 
121us
1.96,1.96,pbox16,1,1269604576,16G,,611,96,90060,15,55711,14,2407,96,200901,27,522.3,15,16,10298,16,+,+++,24208,32,30395,45,+,+++,32511,43,13157us,604ms,1727ms,32420us,142ms,73131us,18504us,598us,613us,87us,25us,121us

-- 
Benjamin Franz
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-27 Thread Robert Spangler
On Saturday 27 March 2010 05:07, John R Pierce wrote:

  for all practical purposes its the same thing. if it was really
  stripe then mirror, a naive mirror handler would think it would have to
  remirror both drives when one half of one of the stripesets failed and
  was replaced. but in fact, the mirror handlres tend to be well aware
  of whats going on.  mirror 0+1  aand stripe that with mirrored 2+3,
  and its really all the samethe native raid10 in newer mdraid is
  cleaner because you don't end up with extra partial volume metadevices...

Thank you kindly for your reply.



-- 

Regards
Robert

Linux User #296285
http://counter.li.org
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-27 Thread Robert Spangler
On Saturday 27 March 2010 09:22, Ross Walker wrote:

   for all practical purposes its the same thing. if it was really
   stripe then mirror, a naive mirror handler would think it would have
   to
   remirror both drives when one half of one of the stripesets failed and
   was replaced. but in fact, the mirror handlres tend to be well
   aware
   of whats going on.  mirror 0+1  aand stripe that with mirrored
   2+3,
   and its really all the samethe native raid10 in newer mdraid is
   cleaner because you don't end up with extra partial volume
   metadevices...

  RAID0+1 is never a good configuration because a single drive failure
  in a RAID0 stripe fails out the whole stripe, and with say an 4x2
  RAID0+1, you are more likely to hit a disk failure with 4 drives in a
  RAID0 then 2 in a RAID1.

  That's why RAID1+0 came about.

Thnx for clearing this up.


-- 

Regards
Robert

Linux User #296285
http://counter.li.org
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Ross Walker
On Mar 25, 2010, at 9:36 PM, Christopher Chan christopher.c...@bradbury.edu.hk 
  wrote:

 On Friday, March 26, 2010 09:12 AM, John R Pierce wrote:
 Christopher Chan wrote:
 but with RAID 10, data is safe after many types of failures.


 Except for the case when a mirror dies after which the whole thing  
 is
 toast but in theory you can survive up to four disks going down.


 if you have a 8 drive raid1+0, and a random drive fails, you can  
 survive
 any other drive failing *except* the mirror of the failed one.   so  
 if a

 That's what I said right? I did not say when a mirror is broken...


 second drive fails, there's only a 1 in 7 chance that its the 'fatal'
 one.   on a 4 drive raid10, its a 1 in 3 chance.   meanwhile, a  
 raid10
 can rebuild from a hotspare in like an hour, if the system isn't  
 busy,
 and a few hours in the background if its busy and active.

 yeah, I thought the raid10 module would be able to rebuild  
 automatically
 from a hotspare and would therefore be better than using nested  
 raid1+0.
 I better stop the nested raid1+0 thing...does ananconda support the
 raid10 module during install yet? I mean rather, is the raid10 module
 included in the installation initrd yet?

No, not yet, but I always recommend setting up your data arrays  
manually so your intimately familiar with how they are constructed and  
the mdadm command usage is fresh in your head.

Did you know with Neil's raid10 implementation you can store 3 copies  
of the data so ANY two drives can fail before you start playing  
Russian roulette!

-Ross

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Benjamin Franz
Ross Walker wrote:

 No, not yet, but I always recommend setting up your data arrays  
 manually so your intimately familiar with how they are constructed and  
 the mdadm command usage is fresh in your head.

 Did you know with Neil's raid10 implementation you can store 3 copies  
 of the data so ANY two drives can fail before you start playing  
 Russian roulette!
   

You can do that with RAID1+0, too. You can setup RAID1 with more than 2 
drives. I have one system with an 8-way RAID1 for the OS.

-- 
Benjamin Franz
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Chan Chung Hang Christopher
Benjamin Franz wrote:
 Ross Walker wrote:
 No, not yet, but I always recommend setting up your data arrays  
 manually so your intimately familiar with how they are constructed and  
 the mdadm command usage is fresh in your head.

 Did you know with Neil's raid10 implementation you can store 3 copies  
 of the data so ANY two drives can fail before you start playing  
 Russian roulette!
   
 
 You can do that with RAID1+0, too. You can setup RAID1 with more than 2 
 drives. I have one system with an 8-way RAID1 for the OS.
 

That's some serious redundancy dude!

Good for reads too...
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread James Bensley

  I have one system with an 8-way RAID1 for the OS.


For real or is that a typo? Is that incase you go on holiday for a week and
a drive-dies-a-day?

-- 
Regards,
James.

http://www.jamesbensley.co.cc/
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Robert Heller
At Fri, 26 Mar 2010 17:16:04 + CentOS mailing list centos@centos.org 
wrote:

 
 
 
 
   I have one system with an 8-way RAID1 for the OS.
 
 
 For real or is that a typo? Is that incase you go on holiday for a week and
 a drive-dies-a-day?
 

I suspect that this is a simular case to what I did: I have a server
with 4 drives.  I have several (small) RAID1 partitions (/boot, /,
/usr, /var, etc.) with 4 mirrors and one large RAID5 with three
partitions and a hot spare (a LVM volumn group, containing /home and
some other partitions). I would guess that the admin with the 8-way
RAID1 for the OS probably also has a 6 or 8 disk RAID5 or RAID6 for
the bulk of the disks.


-- 
Robert Heller -- 978-544-6933
Deepwoods Software-- Download the Model Railroad System
http://www.deepsoft.com/  -- Binaries for Linux and MS-Windows
hel...@deepsoft.com   -- http://www.deepsoft.com/ModelRailroadSystem/

   
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Benjamin Franz
Robert Heller wrote:

 I suspect that this is a simular case to what I did: I have a server
 with 4 drives.  I have several (small) RAID1 partitions (/boot, /,
 /usr, /var, etc.) with 4 mirrors and one large RAID5 with three
 partitions and a hot spare (a LVM volumn group, containing /home and
 some other partitions). I would guess that the admin with the 8-way
 RAID1 for the OS probably also has a 6 or 8 disk RAID5 or RAID6 for
 the bulk of the disks
Yup. 8 way RAID1 for the OS, 8 way RAID6 for the data. I was hoping when 
I setup the 8-way RAID1 for the OS that I would get really good read 
speeds since md is supposed to stripe reads from RAID1, but in practice 
the RAID6 completely kills it for read performance (~61 MB/sec from the 
RAID1 partition vs ~200 MB/sec from the RAID6 partition).

In a deeply ironic turn of events, one of the hard drives in that 
machine died in a way that freaked the hardware controller driver out 
and caused a kernel panic last week. The machine wouldn't finishing 
booting until I physically removed the bad drive. The RAID1 was fine 
afterwards, but the RAID6 had to be manually re-assembled (no 
corruption, it just wouldn't automatically start until I intervened).

-- 
Benjamin Franz

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Les Mikesell
On 3/26/2010 12:45 PM, Robert Heller wrote:
 At Fri, 26 Mar 2010 17:16:04 + CentOS mailing listcentos@centos.org  
 wrote:





 I have one system with an 8-way RAID1 for the OS.


 For real or is that a typo? Is that incase you go on holiday for a week and
 a drive-dies-a-day?


 I suspect that this is a simular case to what I did: I have a server
 with 4 drives.  I have several (small) RAID1 partitions (/boot, /,
 /usr, /var, etc.) with 4 mirrors and one large RAID5 with three
 partitions and a hot spare (a LVM volumn group, containing /home and
 some other partitions). I would guess that the admin with the 8-way
 RAID1 for the OS probably also has a 6 or 8 disk RAID5 or RAID6 for
 the bulk of the disks.

If you are really paranoid you can split things up into raid1 mirrors of 
2 drives each mounted into logical places.  At the expense of having to 
mange the space in smaller chunks (and losing half to redundancy) you 
get the ability to control head contention among jobs and to recover 
data from any single disk after a problem.

-- 
   Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Ross Walker
On Mar 26, 2010, at 10:25 AM, Benjamin Franz jfr...@freerun.com wrote:

 Ross Walker wrote:

 No, not yet, but I always recommend setting up your data arrays
 manually so your intimately familiar with how they are constructed  
 and
 the mdadm command usage is fresh in your head.

 Did you know with Neil's raid10 implementation you can store 3 copies
 of the data so ANY two drives can fail before you start playing
 Russian roulette!


 You can do that with RAID1+0, too. You can setup RAID1 with more  
 than 2
 drives. I have one system with an 8-way RAID1 for the OS.

Yes, but you can use a lot less drives to get the same IOPS with  
Neil's module (which technically isn't RAID10 at all but a whole new  
RAID level).

-Ross

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-26 Thread Robert Spangler
On Thursday 25 March 2010 18:10, Robert Heller wrote:

  The prefered way to go would be RAID10 (RAID1 (mirror) + RAID0 (stripe)).
  Form pairs as RAID1, then strip the pairs.  With 8 disks, this would 4
  pairs, 1.5TB/pair = 1.5*4 = 6TB total.

I am just starting to look into this RAID and I was wondering wouldn't RAID01 
be better then RAID10?  In a 4 disc system having the first two using 
stripping and then backing them up the second set with mirrors?

My though is having D1 and D2 as the primary drives stripping and then having 
D3 backup D1 and D4 backup D2.

And if enough room place a couple more drives in the system as hot standby's.

Or am I looking at this all wrong?


-- 

Regards
Robert

Linux User #296285
http://counter.li.org
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Les Mikesell
On 3/25/2010 2:24 PM, Slack-Moehrle wrote:

 Can anyone provide a tutorial or advice on how to configure a software RAID 5 
 from the command-line (since I did not install Gnome)?

 I have 8 x 1.5tb Drives.


Make matching partitions on each disk with fdisk, setting the type to FD 
(raid autodetect), then 'mdadm create ...' to specify the options and 
start it. See the create section in 'man mdadm'. You'll need at least 
--raid-level=  --raid-devices=  --auto=yes.

Then you'll probably want to add an entry in /etc/fstab to mount the new 
md device somewhere.

-- 
   Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Robert Heller
At Thu, 25 Mar 2010 12:24:57 -0700 (PDT) CentOS mailing list 
centos@centos.org wrote:

 
 
 Can anyone provide a tutorial or advice on how to configure a software RAID 5 
 from the command-line (since I did not install Gnome)?
 
 I have 8 x 1.5tb Drives.

mdadm --create /dev/md0 --level=5 --raid-devices=7 /dev/sd[abcdefg]1

The above will create a level 5 RAID named /dev/md0 of /dev/sda1
/dev/sdb1 /dev/sda1 /dev/sdc1 /dev/sdd1 /dev/sde1 /dev/sdf1, with
hot-spare /dev/sdg1

Note: RAID5 is not really recomended for such large disks.  You run the
risk of a complete data loss if one disk fails and the another disk
fails during the rebuild.


 
 -Jason
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos
 
  

-- 
Robert Heller -- 978-544-6933
Deepwoods Software-- Download the Model Railroad System
http://www.deepsoft.com/  -- Binaries for Linux and MS-Windows
hel...@deepsoft.com   -- http://www.deepsoft.com/ModelRailroadSystem/
   
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread James Bensley
I used this guide for my first RAID on an Ubuntu box, its very straight
forward. Its all command line based so everything here I have used in CentOS
(apart from the writer sets the RAID flag on his drives via the GParted GUI
but this can be done via terminal);

http://bfish.xaedalus.net/2006/11/software-raid-5-in-ubuntu-with-mdadm/

-- 
Regards,
James.

http://www.jamesbensley.co.cc/
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Boris Epstein
On Thu, Mar 25, 2010 at 3:36 PM, Robert Heller hel...@deepsoft.com wrote:
 At Thu, 25 Mar 2010 12:24:57 -0700 (PDT) CentOS mailing list 
 centos@centos.org wrote:



 Can anyone provide a tutorial or advice on how to configure a software RAID 
 5 from the command-line (since I did not install Gnome)?

 I have 8 x 1.5tb Drives.

 mdadm --create /dev/md0 --level=5 --raid-devices=7 /dev/sd[abcdefg]1

 The above will create a level 5 RAID named /dev/md0 of /dev/sda1
 /dev/sdb1 /dev/sda1 /dev/sdc1 /dev/sdd1 /dev/sde1 /dev/sdf1, with
 hot-spare /dev/sdg1

 Note: RAID5 is not really recomended for such large disks.  You run the
 risk of a complete data loss if one disk fails and the another disk
 fails during the rebuild.



 -Jason
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos



 --
 Robert Heller             -- 978-544-6933
 Deepwoods Software        -- Download the Model Railroad System
 http://www.deepsoft.com/  -- Binaries for Linux and MS-Windows
 hel...@deepsoft.com       -- http://www.deepsoft.com/ModelRailroadSystem/

 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos


Robert,

Why is the size a factor here? Why would this be OK with smaller
disks? How would you partition this instead?

Thanks.

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Hakan Koseoglu
On Thu, Mar 25, 2010 at 9:07 PM, Boris Epstein borepst...@gmail.com wrote:
 Note: RAID5 is not really recomended for such large disks.  You run the
 risk of a complete data loss if one disk fails and the another disk
 fails during the rebuild.
 Why is the size a factor here? Why would this be OK with smaller
 disks? How would you partition this instead?
As the disks get bigger, rebuild time also increases and the
performance of the disks don't increase linearly with their storage.
This means that when you are rebuilding a disk, the chances of one of
your other disks failing becomes significantly large. Most suggest
RAID6 these days as a minimum, mirroring and striping appears to be
the most popular.

-- 
Hakan (m1fcj) - http://www.hititgunesi.org
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Rainer Duffner


Am 25.03.2010 um 22:07 schrieb Boris Epstein:





Robert,

Why is the size a factor here? Why would this be OK with smaller
disks? How would you partition this instead?

Thanks.

Boris.




This has been discussed before.

The root of the problem lies in the fact that when a disk fails, you  
have to read-out the data from the other disks to re-build the RAID.

Reads from disks have a certain probability to contain an error.
The larger the disk and the larger the array, the more probable it is  
to encounter such an error while rebuilding the RAID (and if that  
happens, you're RAID is just a piece of scrap-metal)


http://www.google.com/search?q=the+end+of+raid

RAID5 works OK-ish for a couple of 146GB SAS-disks.



Rainer___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Boris Epstein
On Thu, Mar 25, 2010 at 5:27 PM, Hakan Koseoglu ha...@koseoglu.org wrote:
 On Thu, Mar 25, 2010 at 9:07 PM, Boris Epstein borepst...@gmail.com wrote:
 Note: RAID5 is not really recomended for such large disks.  You run the
 risk of a complete data loss if one disk fails and the another disk
 fails during the rebuild.
 Why is the size a factor here? Why would this be OK with smaller
 disks? How would you partition this instead?
 As the disks get bigger, rebuild time also increases and the
 performance of the disks don't increase linearly with their storage.
 This means that when you are rebuilding a disk, the chances of one of
 your other disks failing becomes significantly large. Most suggest
 RAID6 these days as a minimum, mirroring and striping appears to be
 the most popular.

 --
 Hakan (m1fcj) - http://www.hititgunesi.org
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos


Hakan,

You surely do have a point there. However, it is still not all that
likely that a disk will fail during the rebuild time in question (what
are we talking? some hours max?)

Boris.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Slack-Moehrle

As the disks get bigger, rebuild time also increases and the
performance of the disks don't increase linearly with their storage.
This means that when you are rebuilding a disk, the chances of one of
your other disks failing becomes significantly large. Most suggest
RAID6 these days as a minimum, mirroring and striping appears to be
the most popular.

I looked up RAID6 and see the addition or a parity bit.

http://en.wikipedia.org/wiki/Standard_RAID_levels#RAID_6

RAID 10 is also something I looked at. Striped, then Mirrored

So:

8 x 1.5tb = 12tb

RAID 5 = 12tb - 1.5tb for parity data = 10.5tb space available

RAID 10 = 4 x 1.5 = 6tb - 1.5tb for parity data = 4.5tb per stripe then mirror 
it.

but with RAID 10, data is safe after many types of failures.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Les Mikesell
On 3/25/2010 4:43 PM, Boris Epstein wrote:
 On Thu, Mar 25, 2010 at 5:27 PM, Hakan Koseogluha...@koseoglu.org  wrote:
 On Thu, Mar 25, 2010 at 9:07 PM, Boris Epsteinborepst...@gmail.com  wrote:
 Note: RAID5 is not really recomended for such large disks.  You run the
 risk of a complete data loss if one disk fails and the another disk
 fails during the rebuild.
 Why is the size a factor here? Why would this be OK with smaller
 disks? How would you partition this instead?
 As the disks get bigger, rebuild time also increases and the
 performance of the disks don't increase linearly with their storage.
 This means that when you are rebuilding a disk, the chances of one of
 your other disks failing becomes significantly large. Most suggest
 RAID6 these days as a minimum, mirroring and striping appears to be
 the most popular.



 You surely do have a point there. However, it is still not all that
 likely that a disk will fail during the rebuild time in question (what
 are we talking? some hours max?)

The common problem is that there are unused portions of the drives that 
go bad but are unnoticed for a long time.  Then one fails badly enough 
to get kicked out of the raid. Then when you rebuild, you have to 
reconstruct parity for even the unused parts of the drive and you hit 
previously unnoticed bad spots in the process.  I think the last Centos 
update added some sort of raid scan as a cron job that might detect bad 
spots earler, but I'm not sure what it actually does.

-- 
   Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Robert Heller
At Thu, 25 Mar 2010 22:27:56 +0100 CentOS mailing list centos@centos.org 
wrote:

 
 
 
 
 Am 25.03.2010 um 22:07 schrieb Boris Epstein:
 
 
 
  Robert,
 
  Why is the size a factor here? Why would this be OK with smaller
  disks? How would you partition this instead?
 
  Thanks.
 
  Boris.
 
 
 
 This has been discussed before.
 
 The root of the problem lies in the fact that when a disk fails, you  
 have to read-out the data from the other disks to re-build the RAID.
 Reads from disks have a certain probability to contain an error.
 The larger the disk and the larger the array, the more probable it is  
 to encounter such an error while rebuilding the RAID (and if that  
 happens, you're RAID is just a piece of scrap-metal)

Or as was done recently at the Wendell Free Library, your disks become
raw materials for an after school art project... :-)

 
 http://www.google.com/search?q=the+end+of+raid
 
 RAID5 works OK-ish for a couple of 146GB SAS-disks.

More than a couple of disks for RAID5 -- at least 3 are needed for RAID5.

 
 
 
 Rainer
 MIME-Version: 1.0
 
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos
 
   

-- 
Robert Heller -- 978-544-6933
Deepwoods Software-- Download the Model Railroad System
http://www.deepsoft.com/  -- Binaries for Linux and MS-Windows
hel...@deepsoft.com   -- http://www.deepsoft.com/ModelRailroadSystem/

 
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Robert Heller
At Thu, 25 Mar 2010 17:07:47 -0400 CentOS mailing list centos@centos.org 
wrote:

 
 On Thu, Mar 25, 2010 at 3:36 PM, Robert Heller hel...@deepsoft.com wrote:
  At Thu, 25 Mar 2010 12:24:57 -0700 (PDT) CentOS mailing list 
  centos@centos.org wrote:
 
 
 
  Can anyone provide a tutorial or advice on how to configure a software 
  RAID 5 from the command-line (since I did not install Gnome)?
 
  I have 8 x 1.5tb Drives.
 
  mdadm --create /dev/md0 --level=5 --raid-devices=7 /dev/sd[abcdefg]1
 
  The above will create a level 5 RAID named /dev/md0 of /dev/sda1
  /dev/sdb1 /dev/sda1 /dev/sdc1 /dev/sdd1 /dev/sde1 /dev/sdf1, with
  hot-spare /dev/sdg1
 
  Note: RAID5 is not really recomended for such large disks.  You run the
  risk of a complete data loss if one disk fails and the another disk
  fails during the rebuild.
 
 
 
  -Jason
  ___
  CentOS mailing list
  CentOS@centos.org
  http://lists.centos.org/mailman/listinfo/centos
 
 
 
  --
  Robert Heller             -- 978-544-6933
  Deepwoods Software        -- Download the Model Railroad System
  http://www.deepsoft.com/  -- Binaries for Linux and MS-Windows
  hel...@deepsoft.com       -- http://www.deepsoft.com/ModelRailroadSystem/
 
  ___
  CentOS mailing list
  CentOS@centos.org
  http://lists.centos.org/mailman/listinfo/centos
 
 
 Robert,
 
 Why is the size a factor here? Why would this be OK with smaller
 disks? How would you partition this instead?

There was a thread some time back (a few weeks? Couple of months?) about
how as disk size got so much larger, the error rate hasn't really gotten
much better.  With such large disks, the number of I/O operations needed
to do a rebuild of a RAID 5 array is so large that one will be
increasingly likely to hit an error, at which point all bets are off. 
(There are some papers talking about this -- I don't have the links, but
I think they are in the list archives.)

The prefered way to go would be RAID10 (RAID1 (mirror) + RAID0 (stripe)).
Form pairs as RAID1, then strip the pairs.  With 8 disks, this would 4
pairs, 1.5TB/pair = 1.5*4 = 6TB total.

 
 Thanks.
 
 Boris.
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos
 
   
 

-- 
Robert Heller -- 978-544-6933
Deepwoods Software-- Download the Model Railroad System
http://www.deepsoft.com/  -- Binaries for Linux and MS-Windows
hel...@deepsoft.com   -- http://www.deepsoft.com/ModelRailroadSystem/

   
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread John R Pierce
Boris Epstein wrote:
 You surely do have a point there. However, it is still not all that
 likely that a disk will fail during the rebuild time in question (what
 are we talking? some hours max?)
   

8 disks is about the upper limit I'd suggest for a single raid group on 
any sort of system.

rebuilding a 8x1.5TB raid5 could easily take a full day or more will 
you have an online hotspare?  if not, then the rebuild time includes how 
long it takes you to realize there's a bad drive, procure and install 
the replacement, /AND/ the umpteen hours for the rebuild.

personally, I prefer using raid10 or 1+0 (more or less the same thing), 
and for anything above a 2 disk mirror, I prefer to use a proper 
hardware raid controller... for 8+ disks, I'd likely be looking at 
external storage arrays such as the IBM DS3000 family.


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread John R Pierce
Slack-Moehrle wrote:
 RAID 10 = 4 x 1.5 = 6tb - 1.5tb for parity data = 4.5tb per stripe then 
 mirror it.
   

no -1.5 on that.   you don't have parity when you are mirroring.  
8x1.5TB raid10 is simply 4*1.5 = 6TB I'd still want hot spare.  

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Mike McCarty
Robert Heller wrote:
 At Thu, 25 Mar 2010 22:27:56 +0100 CentOS mailing list centos@centos.org 
 wrote:
[...]

 
 The root of the problem lies in the fact that when a disk fails, you  
 have to read-out the data from the other disks to re-build the RAID.
 Reads from disks have a certain probability to contain an error.
 The larger the disk and the larger the array, the more probable it is  
 to encounter such an error while rebuilding the RAID (and if that  
 happens, you're RAID is just a piece of scrap-metal)
 
 Or as was done recently at the Wendell Free Library, your disks become
 raw materials for an after school art project... :-)

It depends on how redundant the array is. With enough
redundancy, one can rebuild even if more than one disc
fails. RAID is essentially indistinguishable from ECC.
If the number of errors (failed reads from discs) does
not exceed the correction ability of the code used
(usualy a Reed-Solomon BCH style code) then the reconstruction
can proceed.

Mike
-- 
p=p=%c%s%c;main(){printf(p,34,p,34);};main(){printf(p,34,p,34);}
Oppose globalization and One World Governments like the UN.
This message made from 100% recycled bits.
You have found the bank of Larn.
I speak only for myself, and I am unanimous in that!
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Christopher Chan
On Friday, March 26, 2010 05:52 AM, Slack-Moehrle wrote:

 As the disks get bigger, rebuild time also increases and the
 performance of the disks don't increase linearly with their storage.
 This means that when you are rebuilding a disk, the chances of one of
 your other disks failing becomes significantly large. Most suggest
 RAID6 these days as a minimum, mirroring and striping appears to be
 the most popular.

 I looked up RAID6 and see the addition or a parity bit.

 http://en.wikipedia.org/wiki/Standard_RAID_levels#RAID_6

RAID6 allows you to survive failure of any two disks in the array at the 
cost of two disks of space.



 RAID 10 is also something I looked at. Striped, then Mirrored

They recommend that you mirror and then stripe the mirrors. But that is 
probably old school now with Neil Brown's raid10 personality. Does 
anyone do nested raid1+0 setups anymore?


 but with RAID 10, data is safe after many types of failures.

Except for the case when a mirror dies after which the whole thing is 
toast but in theory you can survive up to four disks going down.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread John R Pierce
Christopher Chan wrote:
 but with RAID 10, data is safe after many types of failures.
 

 Except for the case when a mirror dies after which the whole thing is 
 toast but in theory you can survive up to four disks going down.
   

if you have a 8 drive raid1+0, and a random drive fails, you can survive 
any other drive failing *except* the mirror of the failed one.   so if a 
second drive fails, there's only a 1 in 7 chance that its the 'fatal' 
one.   on a 4 drive raid10, its a 1 in 3 chance.   meanwhile, a raid10 
can rebuild from a hotspare in like an hour, if the system isn't busy, 
and a few hours in the background if its busy and active.




___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RAID 5 setup?

2010-03-25 Thread Christopher Chan
On Friday, March 26, 2010 09:12 AM, John R Pierce wrote:
 Christopher Chan wrote:
 but with RAID 10, data is safe after many types of failures.


 Except for the case when a mirror dies after which the whole thing is
 toast but in theory you can survive up to four disks going down.


 if you have a 8 drive raid1+0, and a random drive fails, you can survive
 any other drive failing *except* the mirror of the failed one.   so if a

That's what I said right? I did not say when a mirror is broken...


 second drive fails, there's only a 1 in 7 chance that its the 'fatal'
 one.   on a 4 drive raid10, its a 1 in 3 chance.   meanwhile, a raid10
 can rebuild from a hotspare in like an hour, if the system isn't busy,
 and a few hours in the background if its busy and active.

yeah, I thought the raid10 module would be able to rebuild automatically 
from a hotspare and would therefore be better than using nested raid1+0. 
I better stop the nested raid1+0 thing...does ananconda support the 
raid10 module during install yet? I mean rather, is the raid10 module 
included in the installation initrd yet?
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos