Summary: | EBUILD: sys-apps/lvm2-user | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Mike Jones <gentoo> |
Component: | New packages | Assignee: | x86-kernel (DEPRECATED) <x86-kernel> |
Status: | RESOLVED FIXED | ||
Severity: | enhancement | CC: | diemumiee, max, releng, richard, spkane, tantive, zwelch |
Priority: | High | Keywords: | EBUILD |
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
URL: | http://www.sistina.com/products_lvm.htm | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
Ebuild for sys-apps/lvm2-user (LVM Userspace Utilities)
Ebuild: sys-apps/lvm2-user (Userspace utilities for LVM2) /etc/init.d/lvm2 init script for lvm2-user |
Description
Mike Jones
2003-04-21 22:19:57 UTC
Created attachment 10976 [details]
Ebuild for sys-apps/lvm2-user (LVM Userspace Utilities)
Okay, the dog can't enjoy himself this time :-(
Created attachment 10977 [details]
Ebuild: sys-apps/lvm2-user (Userspace utilities for LVM2)
This time it's the right version (was a few revs behind before!)
BUT! I've thrown in an lvm2 init script to make up for my deficiency! :-)
Created attachment 10978 [details]
/etc/init.d/lvm2 init script for lvm2-user
/etc/init.d/lvm2 init script for lvm2-user
to make up for my deficiencies :-(
quick question: why not make lvm2 userspace utils depend on device mapper and do the kernel check in the device-mapper ebuild? Besides that, why force lvm2 and dm to only install in 2.5 kernels? I have it happily running on 2.4.20 with the patch from sistina.com. I just posted a bug (bug 21908) with patches to checkfs and halt.sh that enable/disable LVM on boot/shutdown. I'd say that is a cleaner way of handling LVM (de)activation since there's already support for LVM1 in those files. Therefore there's probably no need for an init script (especially since there's a order problem with it -- see my bug please). Would it make sense also to make this bug depend on 19744 and 21908? Note: if bug 22768 gets added to the kernel the lvm2 and device-mapper ebuilds will need to allow people to build it on a 2.4 kernel, but could check for a version with the LVM2 backport patches applied. This ebuild will also need to be set with an rdepend on device-manager, which it isn't currently. When I tried to build lvm2 first. It failed, but it worked after I built device-mapper (note this is on a patched 2.4 kernel (see bug: 22768)) this is in portage now. *** Bug 25898 has been marked as a duplicate of this bug. *** I don't understand. Why was lvm2-2.00.05 added to Portage when version 2.00.06 is already released since 2003-08-20? Version 2.00.06 has less bugs. Please update! The same holds for device mapper. BTW, this was mentioned in bug 25898 which was marked as duplicate. > The same holds for device mapper.
My comment regarding device-mapper is wrong. Portage already contains the newest version.
:-)
shiny new version is now in portage. enjoy! |