Summary: | sys-fs/udev-145-r1 on openvz system has unwanted depend on gentoo-sources in VC's | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Joakim <moonwalker> |
Component: | New packages | Assignee: | udev maintainers <udev-bugs> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | anton.bugs, dschridde+gentoobugs, gentoo, kernel-misc, tom |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Joakim
2009-10-30 11:02:10 UTC
Some more info on what's happening.
# nano /etc/portage/package.mask
GNU nano 2.1.10 File: /etc/portage/package.mask
>dev-lang/python-2.7
#>=sys-fs/udev-145-r1
sys-kernel/gentoo-sources
# emerge -uDalN system
These are the packages that would be merged, in order:
Calculating dependencies ... done!
!!! The following update has been skipped due to unsatisfied dependencies:
sys-fs/udev:0
!!! All ebuilds that could satisfy "virtual/linux-sources" have been masked.
!!! One of the following masked packages is required to complete your request:
- sys-kernel/gentoo-sources-2.6.31-r4 (masked by: package.mask)
/etc/portage/package.mask:
#>=sys-fs/udev-146-r1
- sys-kernel/gentoo-sources-2.6.31-r3 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.31-r2 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.31-r1 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.31 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30-r8 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30-r7 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30-r6 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30-r5 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30-r4 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30-r3 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.30 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.29-r6 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.29-r5 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.28-r6 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.28-r5 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.27-r10 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.27-r8 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.26-r4 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.25-r9 (masked by: package.mask)
- sys-kernel/gentoo-sources-2.6.16-r13 (masked by: package.mask)
For more information, see the MASKED PACKAGES section in the emerge
man page or refer to the Gentoo Handbook.
(dependency required by "sys-fs/udev-146-r1" [ebuild])
Nothing to merge; would you like to auto-clean packages? [Yes/No]
We know about this. It is because linux-info.eclass depends on virtual/linux-sources, even if this is only a non-fatal check to be done. As tempopary workaround you can set I_KNOW_WHAT_I_AM_DOING to a value and then the eclass will not depend on linux-sources. Actually, I already had I_KNOW_WHAT_I_AM_DOING set but it didn't seem to matter, wanted to pull in gentoo-sources anyway. What I did now instead, as a workaround, was # emerge -O udev to build it without dependencies, then it appear to works as expected with I_KNOW_WHAT_I_AM_DOING and # emerge udev, telling me that I am on my own as it cannot check the kernl config options - but before that when upgrading from udev-145 or earlier I_KNOW_WHAT_I_AM_DOING doesn't prevent to pull in gentoo-sources. So using the -O flag seem tobe the way to go for workaround + I_KNOW_WHAT_I_AM_DOING *** Bug 293147 has been marked as a duplicate of this bug. *** (In reply to comment #2) > > As tempopary workaround you can set I_KNOW_WHAT_I_AM_DOING to a value and then > the eclass will not depend on linux-sources. > I just wanted to stress again that the I_KNOW_WHAT_I_AM_DOING flag really doesn't work. With every release of new udev same sh*t, you have o do emerge -aO1 udev to pull in udev without deps before anything else, then you can run things like emerge -uDaN system|world I thought of it, if the problem is with linux-info.eclass deps, then maybe it's a linux-info.eclass bug or it's a bug to pull in that eclass for udev do with it, just checking kernel version and some flags? Just speculating... The logic per I_KNOW_WHAT_I_AM_DOING is now default, and the linux-info no longer brings in linux-sources at all. I am slightly confused, I see no bump for udev package and last entry in Changelog is of 30 Dec 2009 so just wonder if this has been fixed "outside" of udev or it simply hasn't made it to the tree yet? The fix is in the eclass yes, not the ebuild. zzam noted that in comment #2 that was the source of the problem. |