On Sun, Jan 15, 2017 at 08:13:42AM +0100, Pavel Raiskup wrote: > On Sunday, January 15, 2017 12:30:51 AM CET Nico Kadel-Garcia wrote: > > On Sat, Jan 14, 2017 at 7:20 PM, Zbigniew Jędrzejewski-Szmek > > <zbyszek@xxxxxxxxx> wrote: > > > On Sun, Jan 15, 2017 at 01:17:52AM +0100, Reindl Harald wrote: > > >> > > >> > > >> Am 15.01.2017 um 01:13 schrieb Zbigniew Jędrzejewski-Szmek: > > >> >== No need for static allocation, afaict > > >> >games, man, slocate, squid, named, postgres, mysql, nscd, > > >> >rpcuser, rpc, rpm, ntp, mailman, gdm, utempter, apache, smmsp, > > >> >tomcat, frontpage, nut, beagleindex, avahi, tcpdmp, privoxy, radvd, > > >> >imap, majordomo, polkituser, screen, clamav, saned, mock, ricci, luci > > >> > > >> this idea is horrible wenn you maintain many machines over many > > >> years because the 27/27 for mysqld and 48/48 for apache as example > > >> ensures that you can clone/move data between every redhat based > > >> distribution of the last 15 years and you wnat to go that capability > > >> to be gone > > > > > > How do you "clone/move data"? Normally tar/rsync/scp will use the user > > > or group name, not the number. > > > > Except when it doesn't. The mysql use on one system may not *exist* at > > the time of running rsync or tar, and it certainly does not work wekk > > across mounted disk images or NFSv3 and most NFSv4 mounts. > > > > Let's not change static, stable content that there isn't a compelling > > reason to change. > > Agreed; at least for the databases, please don't drop postgres and mysql, > thank you. > > The pattern to take into account might be that "external storage is often > used to store data" of the service. Without central arbiter (which users > are often a bit to setup for e.g. for "only" database server) it is > non-trivial to keep UIDs/GIDs synced across VM deployments. Ack. Zbyszek _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
