* [linux-lvm] Few doubts related to LVM 2 striping
@ 2009-10-16 5:43 Gopala Krishna
0 siblings, 0 replies; only message in thread
From: Gopala Krishna @ 2009-10-16 5:43 UTC (permalink / raw)
To: linux-lvm
[-- Attachment #1: Type: text/plain, Size: 4299 bytes --]
Hi,
I would like to know little more details on how LV stripe is spreadout on
available devices in a VG.
I have created a logical volume and the lvdisplay -v -m output is shown
below. From this out put no information about the stripes spread out. it is
giving column information but it is not giving where each stripe lies.
Look at the lvdisplay output below. I have created VG with default PE size
of 4 MB and then created striped LV of stripe width 4 KB.
root:~ # lvdisplay -v -m /dev/vg1/lv1
Using logical volume(s) on command line
--- Logical volume ---
LV Name /dev/vg1/lv1
VG Name vg1
LV UUID wIwIAU-Lq8J-Cva9-0eyb-fECN-h5TS-nrMUMu
LV Write Access read/write
LV Status available
# open 0
LV Size 1008.00 MB
Current LE 252
Segments 1
Allocation inherit
Read ahead sectors 0
Block device 253:7
--- Segments ---
Logical extent 0 to 251:
Type striped
Stripes 3
Stripe size 4 KB
Stripe 0:
Physical volume /dev/sdm
Physical extents 0 to 83
Stripe 1:
Physical volume /dev/sdl
Physical extents 0 to 83
Stripe 2:
Physical volume /dev/sdk
Physical extents 0 to 83
Here the stripe stripe width is 4 KB and there are 3 stripe columns. I would
like to know how each stripe width is distributed. does it distributed round
robin fashion among the devices or anything else? Above output gives only
column information. Is there any command or an option I can use to get to
know more on stripes spreadout?
For example, in above case, LE/PE size is 4 MB(look at the vgdisplay output
shown below)
So my question is:
does the stripe spreads out as follow?
first 4 KB (stripe width) sits on /dev/sdm, PE 0 and next 4 KB on /dev/sdl
andPE 0 and 3 rd stripe gets in to 0 th PE of /dev/sdk and then again this
process repeates and start occypying 0 th PE of /dev/sdm and so on. When it
finishes with 0 th PE (4 MB) of each device, then corresponding stripe
occupies .
Following is a representation of above statement:
stripe-1 PE: 0 /dev/sdm
stripe-2 PE:0 /dev/sdl
stripe-3 PE:0 /dev/sdk
stripe-4 PE:0 /dev/sdm
stripe-5 PE:0 /dev/sdl
;
;
stripe-xx PE:1 /dev/sdm
stripe-xx+1 PE:1 /dev/sdl
stripe-XX+2 PE:1 /dev/sdk
stripe-xx+3 PE:1 /dev/sdm
OR
First few sets of 4 KB stripes (i.e PE size/ stripe width) occupies
/dev/sdm and next set of stripes /dev/sdl and then /dev/sdk and comes back
to /dev/sdm. i.e it continuously writes so many stripes to align to PE size.
Following is a representation of above statement:
stripe-1 PE:0 /dev/sdm
stripe-2 PE:0 /dev/sdm
;
stripe-x PE:0 /dev/sdm
stripe-x+1 PE:0 /dev/sdl
stripe-X+2 PE:0 /dev/sdl
;
;
stripe-y PE:0 /dev/sdl
stripe-y+1 PE:0 /dev/sdk
;
stripe-z PE:0 /dev/sdk
stripe-z+1 PE:1 /dev/sdm (repeates with physical extent of each
device)
stripe-Z+2 PE:1 /dev/sdm
;
;
i.e stripes are going to occupy on the device until at least 1 PE is fulll
before mnoving to next device.
OR
Does it something different from both of these. If so how to identify where
exactly particular stripe of data lies on which disk of the VG. Is there any
command or metadata dump or debug info where I can get those information?
Please let me know. Thanks for your time.
Thanks,
Gopal.
====================vgdisplay output====
spea64:~ # vgdisplay -v /dev/vg1
Using volume group(s) on command line
Finding volume group "vg1"
--- Volume group ---
VG Name vg1
System ID
Format lvm2
Metadata Areas 4
Metadata Sequence No 8
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 1
Open LV 0
Max PV 0
Cur PV 4
Act PV 4
VG Size 1.94 GB
PE Size 4.00 MB
Total PE 496
Alloc PE / Size 252 / 1008.00 MB
Free PE / Size 244 / 976.00 MB
VG UUID dp2pO1-qqNQ-BRrM-jrZ8-zkbo-y7OF-CgS5eo
[-- Attachment #2: Type: text/html, Size: 5428 bytes --]
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2009-10-16 5:43 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-10-16 5:43 [linux-lvm] Few doubts related to LVM 2 striping Gopala Krishna
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.