Skip to contentSkip to conversation

LV's above 2TB, PV's above 1TB, LBD?

linux-lvm Tad Kollar 4 participants 4 messages Nov 14, 2002 – Nov 19, 2002

lore
Share & tools
lore

Tad Kollar

author
#0 lore raw

Using LVM2 tools from CVS, and kernel 2.5.47 w/dm patches...

Seems to be a sign error someplace... 2+ TB requests are treated as 4+ TB:

lvm> lvextend -l 10980 /dev/vgtest2/lvtest2
Extending logical volume lvtest2 to 2.68 TB
logical volumes cannot be larger than 4.00 TB

Also, pvcreate treats 1+ TB devices as 2+ TB and rejects them.

Aside from this, now that the 2.5 kernel supports LBD's, can the limit on LV's be raised to 16 TB?

Thanks,
Tad

#1 lore raw
Quoted 3 lines · Tad Kollar

lvm> lvextend -l 10980 /dev/vgtest2/lvtest2
Extending logical volume lvtest2 to 2.68 TB
logical volumes cannot be larger than 4.00 TB

A missing division by 2 when producing the error message:-)

Quoted 1 line

Also, pvcreate treats 1+ TB devices as 2+ TB and rejects them.

Maybe another missing factor of 2 somewhere?
Can you provide some *specific* output demonstrating this problem (like you did for the first problem)?

Quoted 2 lines

Aside from this, now that the 2.5 kernel supports LBD's, can the limit on LV's be raised to 16 TB?

For the *new* format of metadata (in beta 4 released today), yes:
for now, I've knocked out the LV size restriction in CVS. Further tweaks to the tools may also be required. Please try it and let us know whether it works, or what else needs changing.

Alasdair

Signature · 1 line
[email protected]
#2 lore raw
Quoted 6 lines · Alasdair G Kergon
Quoted 2 lines

Aside from this, now that the 2.5 kernel supports LBD's, can the limit on LV's be raised to 16 TB?

For the *new* format of metadata (in beta 4 released today), yes:
for now, I've knocked out the LV size restriction in CVS. Further tweaks to the tools may also be required. Please try it and let us know whether it works, or what else needs changing.

Is the new metdata format really required?
Isn't the unit used in the metadata PEs? So should it not work with the old format if only the PEs are large enough?

Signature · 2 lines
Ragnar Kjørstad
Big Storage
#3 lore raw
Quoted 11 lines · Ragnar Kjørstad
Quoted 6 lines · Alasdair G Kergon
Quoted 2 lines

Aside from this, now that the 2.5 kernel supports LBD's, can the limit on LV's be raised to 16 TB?

For the *new* format of metadata (in beta 4 released today), yes:
for now, I've knocked out the LV size restriction in CVS. Further tweaks to the tools may also be required. Please try it and let us know whether it works, or what else needs changing.

Is the new metdata format really required?
Isn't the unit used in the metadata PEs? So should it not work with the old format if only the PEs are large enough?

The constraint in the LVM1 format is ~2^16 physical extents maximum. If we support more than this amount, we'ld be LVM1 format incompatible. Going beyond that constraint is possible with the new format Alasdair was refering to but you loose LVM1 format compatibility.
IOW: no way back to running such configuration with LVM1.

But yes, with larger physical extent sizes _and_ less than ~2^16 extents maximum you can stay with LVM1 format in order to be backwards compatible.

Quoted 12 lines
Signature · 9 lines
-- 
Ragnar Kjørstad
Big Storage

_______________________________________________
linux-lvm mailing list
[email protected]
http://lists.sistina.com/mailman/listinfo/linux-lvm
read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
Signature · 15 lines
Regards,
Heinz    -- The LVM Guy --

*** Software bugs are stupid.
    Nevertheless it needs not so stupid people to solve them ***

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Sistina Software Inc.
Senior Consultant/Developer                       Am Sonnenhang 11
                                                  56242 Marienrachdorf
                                                  Germany
[email protected]                           +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-