Phasing Out SysVInit In openSUSE Raises Concerns
OpenSUSE 12.1 introduced support for systemd but it didn't defenestrate SysVinit as there are still some dependencies on this older init system. However, there's a proposal now to completely phase out SysVinit within openSUSE and it's been met by some mixed views.
Cristian RodrÃguez proposed phasing out SysVinit from future openSUSE releases due to the burden in supporting two different init systems -- systemd and SysVinit. Cristian proposed a three-phase process for doing away entirely with SysVinit.
The complete init migration process involves addressing outstanding systemd bugs, converting the rest of the services to having native systemd units, making corrections with rpmlint, lots of testing, and then stripping away SysVinit. The proposal can be read on the opensuse-factory mailing list.
However, so far this proposal has been met by much criticism concerning systemd. "Not until systemd can be proved to work correctly in all cases. So far, the track record is awful." "it is still alpha [quality] right now." "On one of my 64-bit systems, systemd would intermittently hang on boot and never succeed (I waited 2 hours one time.). The failure rate was about 2 in 5 tries." Etc...
It looks like it may be a few more releases before SysVinit can be done entirely away with in the openSUSE world. Meanwhile in the world of Fedora 17 they hope to finish porting SysVinit scripts over to systemd unit files for the remainder of their services, after their systemd upbringing began in Fedora 15.
How's systemd working out for you if you're a user of openSUSE, Fedora, or one of the other Linux distributions that's offered preliminary support for this new init system? Share your thoughts in the forums.
Cristian RodrÃguez proposed phasing out SysVinit from future openSUSE releases due to the burden in supporting two different init systems -- systemd and SysVinit. Cristian proposed a three-phase process for doing away entirely with SysVinit.
The complete init migration process involves addressing outstanding systemd bugs, converting the rest of the services to having native systemd units, making corrections with rpmlint, lots of testing, and then stripping away SysVinit. The proposal can be read on the opensuse-factory mailing list.
However, so far this proposal has been met by much criticism concerning systemd. "Not until systemd can be proved to work correctly in all cases. So far, the track record is awful." "it is still alpha [quality] right now." "On one of my 64-bit systems, systemd would intermittently hang on boot and never succeed (I waited 2 hours one time.). The failure rate was about 2 in 5 tries." Etc...
It looks like it may be a few more releases before SysVinit can be done entirely away with in the openSUSE world. Meanwhile in the world of Fedora 17 they hope to finish porting SysVinit scripts over to systemd unit files for the remainder of their services, after their systemd upbringing began in Fedora 15.
How's systemd working out for you if you're a user of openSUSE, Fedora, or one of the other Linux distributions that's offered preliminary support for this new init system? Share your thoughts in the forums.
24 Comments