I've had an odd issue trying to set up kickstart with Centos 7.
First of all, the relevant portion of the file itself. The following works, although it's not optimal.
zerombr clearpart --all --initlabel --drives=sda,sdb part raid.11 --size=500 --ondrive=sda part raid.12 --size=12000 --grow --ondrive=sda part raid.21 --size=500 --ondrive=sdb part raid.22 --size=12000 --grow --ondrive=sdb raid /boot --fstype xfs --device md1 --level=RAID1 raid.11 raid.21 raid pv.01 --device md0 --level=RAID1 raid.12 raid.22 volgroup sysvg pv.01 logvol / --vgname=sysvg --size=8000 --name=lv_root --fstype=xfs logvol swap --vgname=sysvg --name=lv_swap --grow --size=1024 --maxsize=2048 logvol /mnt/edata --vgname=sysvg --percent=90 --grow --name=mntedata
This is one of the few configurations so far that it works for me, and it doesn't if I do it like this:
logvol /mnt/edata --vgname=sysvg --percent=100 --grow --name=mntedata
I.e. if instead of
--percent=90 I use
--percent=100 it comes back with a
"ValueError: not enough free space in volume group".
The same thing (and same error) occurs I don't use
--percent at all, but use instead something like
In fact the only other way I've managed to get this to work is if I define
--maxsize=<X>, but it's not a good solution for me since I want the kickstart script to work with different volume sizes based on the VM or physical machine.
So my questions:
1) Why doesn't
--percent=100 work but
--percent=90 does? It's supposed to already take into consideration any other statically defined volumes.
I don't want to waste space from my volumes so this workaround is far from optimal.
2) Is there another way to do what I'm trying above? Essentially to define one volume which grows to use all of the available space in the group.
I don't see why it doesn't, and pretty much this identical setup works perfectly in Centos 6.