On Mon, May 4, 2009 at 2:55 PM, Dan Johansson <dan.johans...@dmj.nu> wrote: > On Sunday 03 May 2009, Duncan wrote: >> Dan Johansson <dan.johans...@dmj.nu> posted >> 200905031107.24458.dan.johans...@dmj.nu, excerpted below, on Sun, 03 May >> >> 2009 11:07:17 +0200: >> > I noticed that emerge wanted to downgrade my >> > app-emulation/vmware-modules-1.0.0.15-r2 >> > >> > # emerge --update --deep --verbose --reinstall changed-use world >> > --pretend >> > >> > These are the packages that would be merged, in order: >> > >> > Calculating dependencies... done! >> > [ebuild U ] dev-perl/XML-Parser-2.36 [2.34-r1] 0 kB [?=>0] [ebuild >> > UD] app-emulation/vmware-modules-1.0.0.15-r2 [1.0.0.23] 0 kB [0] >> > >> > Total: 2 packages (1 upgrade, 1 downgrade), Size of downloads: 0 kB >> > >> > I really don't get it >> >> It's likely that your current version has been masked, for some reason. >> Try this to find out why: >> >> emerge --pretend =app-emulation/vmware-modules-1.0.0.23 >> >> That should tell you why it can't remerge that version, spitting out the >> comment from package.mask or otherwise telling you what's wrong >> (keywords, blocker, whatever). >> >> FWIW, it lets me pretend-install that version here, but I don't have >> VMWare installed (it's proprietaryware which I don't do) so whatever >> blocker there might be I'm not seeing, and if the mask was just added in >> the last 24-36 hours or so, I'd not see it either as I've not synced in >> that time. > No it's not the vmware-modules ebuild - it's not masked or keyworded. > # emerge --pretend =app-emulation/vmware-modules-1.0.0.23 > > These are the packages that would be merged, in order: > > Calculating dependencies... done! > [ebuild R ] app-emulation/vmware-modules-1.0.0.23 > > I think it's a change in the vmware-server ebuild - without version change > (:-( > > # grep vmware-modules /usr/portage/app-emulation/vmware-server/*.ebuild > ~app-emulation/vmware-modules-1.0.0.15 > !<app-emulation/vmware-modules-1.0.0.15 > !>=app-emulation/vmware-modules-1.0.0.16 > > The !>=...1.0.0.16 must be a new addition in the last one to two weeks. Before > that this was not a problem. > > If the devs think that 1.0.0.23 is a bad idea then OK I'll downgrade....
I suggest using the vmware overlay... it seems to be much more up-to-date, especially when vmware-modules gets updated after a new kernel, etc. I'm using it and haven't run into this problem.
