1 |
On 6/24/19 2:40 AM, Peter Humphrey wrote: |
2 |
> Yes, I've done the same on two boxes that have no need of lvm. It does |
3 |
> seem wasteful though. |
4 |
|
5 |
Probably. |
6 |
|
7 |
I dislike the fact that other things that need device mapper have to |
8 |
drag LVM along, or apply (what I call) a device-mapper-only /hack/. |
9 |
|
10 |
I feel like device-mapper should be it's own package that other things |
11 |
depend on; LVM, RAID (mdadm, et al.), multi-path, LUKS (cryptsetup). |
12 |
|
13 |
> I forget the detail now, but a recent-ish version of sys-fs/cryptsetup |
14 |
> found it needed a hard dependency on some of the code in lvm2. |
15 |
|
16 |
Did you apply (what I call) the device-mapper-only /hack/. Or was LVM |
17 |
pulled in for device-mapper? |
18 |
|
19 |
> It seems to me that we have here an opportunity for redesign of certain |
20 |
> packages. ("We" the community, that is.) |
21 |
|
22 |
Agreed. |
23 |
|
24 |
> On this box, which does need lvm for RAID-1 on two SSDs: |
25 |
|
26 |
Do you /need/ LVM? Or is it extra that comes with device-mapper? |
27 |
|
28 |
|
29 |
|
30 |
-- |
31 |
Grant. . . . |
32 |
unix || die |