On 29/09/2013 13:58, Volker Armin Hemmann wrote: > Am 29.09.2013 13:03, schrieb Greg Woodbury: >> On 09/29/2013 06:55 AM, Volker Armin Hemmann wrote: >> >>> why do you bring up udev and systemd AT ALL? >>> >>> They are not the problem or the reason why seperate /usr is prone to >>> break. >>> >> Except that systemd *is* why a seperate /usr is broken now. >> Parts of the libraries that systemd depend on we *deliberately* placed >> in /usr despite the fact that they are needed to bbring the system to >> an operational state. For *years* things required to boot the system >> were defined to be in the root file system, and items not required >> until after mounting had been accomplished were to be placed in /usr. >> >> BTW: There is a standard (The File System Hierarch Standard - FSS) >> that existed and described this behaviour. It was killed off by >> deliberate vendor refusals to support or adhere to it. In >> frustration, the folks involved simply gave up. >> > > things were broken way before that. As much as I hate systemd, it is not > the root cause of the problem. > > The problems were caused by people saying that seperate /usr was a good > idea, so / would not fill up and similar idiocies. The problems were > caused by people saying that lvm is a good idea - for desktops. Those > people who are fighting against the kernel auto assembling raids are to > blame too. > > Systemd is just another point in a very long list.
Volker, we agree. The problem as I see it is that we have an artificial, arbitrary separation between "boot time" stuff and "something that happens later" stuff. There is no clear definition of what these things are and the only real technical criteria advanced thus far is quoted above: "after mounting had been accomplished" That worked in the 80s when SysV came out. But times move on, new methods and hardware were developed and computing is now a very different beast to what it was 30 years ago. Nowadays we have a boatload of actions that can/may be needed to happen before fstab can be read to mount the rest of the system. /usr has become, whether we like it or not, an indespensable part of the userland start up process, and the only way out of this is to have some guarantees in place. We already have a perfectly good one - the root file system is guaranteed to be mounted by the kernel before init is called. If that filesystem does not contain /usr then a rather sophisticated hack is available to ensure that /usr is available, and it is an initramfs. I do beleive the choice really is that clear - provide that guarantee or be stuck forever with old code, hardware and methods. Just because SysV worked well for ages does not mean it's rules must persist through time. Everything changes in this worls, and our game changes faster than most other things. Let's not cling to sacred cows when the world has observably moved on. None of this means I think systemd is good (or bad). Maybe it's over-engineered, but at least someone has the balls to stand up and try deal with the actual problem. -- Alan McKinnon alan.mckin...@gmail.com
