[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On 19/07/16 21:05, M. J. Everitt wrote: > On 19/07/16 20:48, mr meowski wrote: > How old is old? I'm running a moderately vintage gentoo install here > now, but I've found all too often that if you get major > package/dependency breakage, you are better off starting back from a > stage3 and either copying your make.conf and/or world files over, and > starting afresh. Gentoo really is a rolling distro, and not to keep > up-to-date does cause its problems. > > Whilst not a 'guru' or expert at all, I do circulate with the gentoo > devs (a strange bunch, I'll admit) .. but I think they too, whilst > arguing that it -should- be possible, might also agree that starting > afresh would be a better option. > > However, I'm game for a shot .. what is portage complaining about, and > are you a systemd fan or ardently against? (marmite issue, I know!). > pastebin/bpaste works :) Ah ha! I was secretly hoping that you might reply - I was sure that I'd seen you mentioning using Gentoo in production before. Bit nefarious of me - I've posted RPi questions before effectively hoping that Gordon would pipe up with a canonical answer :] The instance is my single oldest surviving Linux install from way back in London and I'm (stupidly) slightly emotionally attached to the old fellow as I leant many, many things from running it. Once upon a time it was a physical machine and since then survived migration to two different computers, conversion to a VM (twice), went from x86 to x64 and from normal to testing. It's like an old knackered pet that I can't bear to see put down. All the data I want is safe and as a VM it's snapshotted of course so I can attempt any amount of otherwise dangerous rescue attempts on it bit I'd really like to keep the "same" instance alive, at least in spirit even if it's a veritable ship of Theseus at this point. About a year (?) ago systemd choked an "emerge world" and I was never able to untangle the mess, despite finding numerous threads from others who evidently had fixed it. It is already a systemd* box but something related to early systemd upgrades back when it was a bit unstable went badly wrong. Let me fire it up a minute! Thanks Michael :] Cheers * as for systemd, I'm pragmatically ambivalent about it - I've learnt to use it no problem as it's not going anywhere on Linux it seems. In this case I'm perfectly happy to keep it systemd as trying to back it out to OpenRC in a previous rescue attempt on a clone went very poorly indeed... -- The Mailing List for the Devon & Cornwall LUG https://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/listfaq