lvm over raid5 - sizes
I have alot of ideas going through my head. Here is my situation:
I've had 5 IDEs with 4 partitions each. Each partition is in a RAID5 md:stripped. md0 = hda5, hdb5, hdc5, hdd5, hde5
md1 = hda6, hdb6, hdc6, hdd6, hde6
md2 = hda7, hdb7, hdc7, hdd7
md3 = hda8, hdb8, hdc8, hdd8
md0-3 are in an LVM2 volume group. This makes it really nice when expanding or adding another disk because I just pull say md3 out of the volume group... rebuild md3 bigger and then add it back.
I just upgraded, replacing hde with a bigger disk and while the pvmoves are taking FOREVER... I thought what is the right size for the RAID5 chunk-size?
I have chunk-size = 128k right now.
The pv extents size is (always?) 4MB.
Am I just shooting myself in the foot by not making my RAID chunk size atleast 1MB? Can someone give me a reason to increase it?
BTW, I'm using reiserfs with default sizes and my needs are mixed. I have many couple gig files, but also some "rsyncs" of client machines as backups/archives.
Thanks for any ideas,
Scott
Quoted 16 lines · Scott Serr
I have alot of ideas going through my head. Here is my situation:
I've had 5 IDEs with 4 partitions each. Each partition is in a RAID5 md:stripped.
md0 = hda5, hdb5, hdc5, hdd5, hde5
md1 = hda6, hdb6, hdc6, hdd6, hde6
md2 = hda7, hdb7, hdc7, hdd7
md3 = hda8, hdb8, hdc8, hdd8md0-3 are in an LVM2 volume group. This makes it really nice when expanding or adding another disk because I just pull say md3 out of the volume group... rebuild md3 bigger and then add it back.
I just upgraded, replacing hde with a bigger disk and while the pvmoves are taking FOREVER... I thought what is the right size for the RAID5 chunk-size?
Quoted 6 lines
I have chunk-size = 128k right now.
The pv extents size is (always?) 4MB.
Am I just shooting myself in the foot by not making my RAID chunk size atleast 1MB? Can someone give me a reason to increase it?
You're shooting yourself in the foot with RAID 5 and a crazy setup like this in the first place. I sure hope this isn't for a production box :)
Just get more disks and use RAID 1. It's faster, simpler, and more robust. You can still do all of the tricks that you're looking for.
Hey, maybe one of the disks has failed. Standard RAID 5 behaviour as designed in this case is to slow to a crawl. Alternatively maybe you're just experiencing what is known in the 'biz as "The Butterfly Effect".
With most striping, keep the chunk size as large as possible. It's a real waste to spend 10ms waiting for the disk platter to rotate, and then only read 128kB of data. Set it as high as possible.
Signature · 2 lines
Sam Vilain, sam /\T vilain |><>T net, PGP key ID: 0x05B52F13 (include my PGP key ID in personal replies to avoid spam filtering)
Well, I have a similar problem/desire.
I want to have RAID5 for redundancy/reliability. But I want to be able to resize the file system stored on the RAID5 array, as well as being able to add and remove hard disks to either increase/decrease available storage.
What would be your recommendation for a configuration?
At the moment, I'm considering using EVMS with its RAID5 regions, and put LVM2 on top of it.
Thanks in advance,
Erik.
Quoted 64 lines
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Sam Vilain
Sent: Monday, February 28, 2005 12:11 AM
To: LVM general discussion and development
Subject: Re: [linux-lvm] lvm over raid5 - sizesQuoted 3 lines · Scott Serr
I have alot of ideas going through my head. Here is my situation:
I've had 5 IDEs with 4 partitions each. Each partition is
in a RAID5
Quoted 8 lines
md:stripped.
md0 = hda5, hdb5, hdc5, hdd5, hde5
md1 = hda6, hdb6, hdc6, hdd6, hde6
md2 = hda7, hdb7, hdc7, hdd7
md3 = hda8, hdb8, hdc8, hdd8md0-3 are in an LVM2 volume group. This makes it really nice when expanding or adding another disk because I just pull say
md3 out of the
Quoted 2 lines
volume group... rebuild md3 bigger and then add it back. I just upgraded, replacing hde with a bigger disk and while
the pvmoves
Quoted 1 line
are taking FOREVER... I thought what is the right size for
the RAID5
Quoted 7 lines
chunk-size?
I have chunk-size = 128k right now.
The pv extents size is (always?) 4MB.
Am I just shooting myself in the foot by not making my RAID
chunk size
Quoted 1 line
atleast 1MB? Can someone give me a reason to increase it?
You're shooting yourself in the foot with RAID 5 and a crazy setup like this in the first place. I sure hope this isn't for a production box :)
Just get more disks and use RAID 1. It's faster, simpler, and more robust. You can still do all of the tricks that you're looking for.
Hey, maybe one of the disks has failed. Standard RAID 5 behaviour as designed in this case is to slow to a crawl. Alternatively maybe you're just experiencing what is known in the 'biz as "The Butterfly Effect".
With most striping, keep the chunk size as large as possible.
It's a real waste to spend 10ms waiting for the disk platter
to rotate, and then only read 128kB of data. Set it as high as possible.Signature · 7 lines
Sam Vilain, sam /\T vilain |><>T net, PGP key ID: 0x05B52F13 (include my PGP key ID in personal replies to avoid spam filtering) _______________________________________________ linux-lvm mailing list [email protected] https://www.redhat.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
Oh wow... I thought EMVS was a LVM2 "equivalent." Can't really give you any ideas about EMVS without doing some reading, but the "RAID5 region" sounds neat.
My Experience:
I've been using LVM2 ontop of software RAID5 for over a year. I've been happy with it. It seems a little slow... but I don't care too much. I know atleast some other people are doing it, and it does fit my needs. I've gone through the whole "grow" process twice now. It is time consuming to grow the disk, but you don't have to sit and watch... and you can have the system live if you wish. A "pvmove /dev/md0 /dev/md3" takes about 8 hours with their sizes about 174GB.
If you can afford 1 hot spare (like your other message said) you might want to go with RAID1, it would be faster.
Quoted 94 lines · Erik Ohrnberger
Well, I have a similar problem/desire.
I want to have RAID5 for redundancy/reliability. But I want to be able to resize the file system stored on the RAID5 array, as well as being able to add and remove hard disks to either increase/decrease available storage.
What would be your recommendation for a configuration?
At the moment, I'm considering using EVMS with its RAID5 regions, and put LVM2 on top of it.
Thanks in advance,
Erik.Quoted 53 lines
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Sam Vilain
Sent: Monday, February 28, 2005 12:11 AM
To: LVM general discussion and development
Subject: Re: [linux-lvm] lvm over raid5 - sizesQuoted 5 lines · Scott Serr
I have alot of ideas going through my head. Here is my situation:
I've had 5 IDEs with 4 partitions each. Each partition is
in a RAID5
Quoted 5 lines
md:stripped. md0 = hda5, hdb5, hdc5, hdd5, hde5 md1 = hda6, hdb6, hdc6, hdd6, hde6 md2 = hda7, hdb7, hdc7, hdd7 md3 = hda8, hdb8, hdc8, hdd8
md0-3 are in an LVM2 volume group. This makes it really nice when expanding or adding another disk because I just pull say
md3 out of the
Quoted 3 lines
volume group... rebuild md3 bigger and then add it back. I just upgraded, replacing hde with a bigger disk and while
the pvmoves
Quoted 3 lines
are taking FOREVER... I thought what is the right size for
the RAID5
Quoted 9 lines
chunk-size?
I have chunk-size = 128k right now.
The pv extents size is (always?) 4MB.
Am I just shooting myself in the foot by not making my RAID
chunk size
Quoted 3 lines
atleast 1MB? Can someone give me a reason to increase it?
Quoted 17 lines
You're shooting yourself in the foot with RAID 5 and a crazy setup like this in the first place. I sure hope this isn't for a production box :)
Just get more disks and use RAID 1. It's faster, simpler, and more robust. You can still do all of the tricks that you're looking for.
Hey, maybe one of the disks has failed. Standard RAID 5 behaviour as designed in this case is to slow to a crawl. Alternatively maybe you're just experiencing what is known in the 'biz as "The Butterfly Effect".
With most striping, keep the chunk size as large as possible. It's a real waste to spend 10ms waiting for the disk platter to rotate, and then only read 128kB of data. Set it as high as possible.
Signature · 6 lines
Sam Vilain, sam /\T vilain |><>T net, PGP key ID: 0x05B52F13 (include my PGP key ID in personal replies to avoid spam filtering) _______________________________________________ linux-lvm mailing list [email protected] https://www.redhat.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/_______________________________________________
linux-lvm mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/linux-lvm
read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
Quoted 31 lines · Sam Vilain
Quoted 16 lines · Scott Serr
I have alot of ideas going through my head. Here is my situation:
I've had 5 IDEs with 4 partitions each. Each partition is in a RAID5 md:stripped.
md0 = hda5, hdb5, hdc5, hdd5, hde5
md1 = hda6, hdb6, hdc6, hdd6, hde6
md2 = hda7, hdb7, hdc7, hdd7
md3 = hda8, hdb8, hdc8, hdd8md0-3 are in an LVM2 volume group. This makes it really nice when expanding or adding another disk because I just pull say md3 out of the volume group... rebuild md3 bigger and then add it back.
I just upgraded, replacing hde with a bigger disk and while the pvmoves are taking FOREVER... I thought what is the right size for the RAID5 chunk-size?Quoted 6 lines
I have chunk-size = 128k right now.
The pv extents size is (always?) 4MB.
Am I just shooting myself in the foot by not making my RAID chunk size atleast 1MB? Can someone give me a reason to increase it?
You're shooting yourself in the foot with RAID 5 and a crazy setup like this in the first place. I sure hope this isn't for a production box :)
Just get more disks and use RAID 1. It's faster, simpler, and more robust. You can still do all of the tricks that you're looking for.
Quoted 7 lines
Hey, maybe one of the disks has failed. Standard RAID 5 behaviour as designed in this case is to slow to a crawl. Alternatively maybe you're just experiencing what is known in the 'biz as "The Butterfly Effect".
With most striping, keep the chunk size as large as possible. It's a real waste to spend 10ms waiting for the disk platter to rotate, and then only read 128kB of data. Set it as high as possible.
It's just a home server. I don't need alot of speed, but I rather have it sorta optimized. This is as many disks that will fit in my PowerEdge 400SC, and it's my own money... so these are the reasons for not having RAID1 with more disks. Thanks for the tips though, I will try to up the size. I couldn't find anything on google about Butterfly Effect - do you have any links?
The disks (during the pvmove) sound like they sketch around ALOT. I suppose if I had 1MB chunks this would cut down the seeks by 1/8th. ;)
My requirements for this disk system:
- not loose data if one disk goes south.
- expand filesystems as I get bigger disks. - maximize available disk space, limit cost.
- room for 5 disks
I'm open to suggestions that are less complicated.
Quoted 3 lines · Scott Serr
Quoted 2 lines
designed in this case is to slow to a crawl. Alternatively maybe you're just experiencing what is known in the 'biz as "The Butterfly Effect".
size. I couldn't find anything on google about Butterfly Effect - do you have any links?
How strange! The term isn't in the Jargon file.
For now you'll just have to go to dictionary.com and search for "walking drives"
Signature · 2 lines
Sam Vilain, sam /\T vilain |><>T net, PGP key ID: 0x05B52F13 (include my PGP key ID in personal replies to avoid spam filtering)